Skip to content

I
imagej-elphel
Commit 1671728b authored by Andrey Filippov's avatar Andrey Filippov
Browse files
Options

Equalizing intensities

parent c4df622d
Hide whitespace changes
Inline Side-by-side
Showing with 916 additions and 583 deletions
src/main/java/com/elphel/imagej/common/PolynomialApproximation.java
View file @ 1671728b
package com.elphel.imagej.common;
import java.util.concurrent.atomic.AtomicInteger;
import com.elphel.imagej.tileprocessor.ImageDtt;
import com.elphel.imagej.tileprocessor.QuadCLT;
import Jama.LUDecomposition;
import Jama.Matrix;
public class PolynomialApproximation {
public int debugLevel=1;
// TODO Move other methods here
// TODO Move other methods here
public PolynomialApproximation(){}
public PolynomialApproximation(int debugLevel){
this.debugLevel=debugLevel;
......@@ -53,7 +58,7 @@ public class PolynomialApproximation {
System.out.println("polynomialApproximation1d() B:");
B.print(10, 5);
}
// while (!(new LUDecomposition(M)).isNonsingular() && (N1>0)){
// while (!(new LUDecomposition(M)).isNonsingular() && (N1>0)){
while (!(new LUDecomposition(M)).isNonsingular() && (N1>=0)){ // make N=0 legal ?
aM=new double [N1][N1];
aB=new double [N1][1];
......@@ -81,14 +86,14 @@ public class PolynomialApproximation {
for (int i=0;i<=N;i++) result[i]=(i<=N1)?aR[i][0]:0.0;
return result;
}
/**
* Linear approximates each of 3 functions of 3 variables and finds where they are all zero
* @param data: for each sample (1-st index):
* 0 - {x,y,z}
* 1 - {f1, f2, f3},
* 2 - {weight}
* @return {x0, y0, z0} where A1*x0+B1*y0+C1*z0+D1=0, A2*x0+B2*y0+C2*z0+D2=0, A3*x0+B3*y0+C3*z0+D3=0
*/
/**
* Linear approximates each of 3 functions of 3 variables and finds where they are all zero
* @param data: for each sample (1-st index):
* 0 - {x,y,z}
* 1 - {f1, f2, f3},
* 2 - {weight}
* @return {x0, y0, z0} where A1*x0+B1*y0+C1*z0+D1=0, A2*x0+B2*y0+C2*z0+D2=0, A3*x0+B3*y0+C3*z0+D3=0
*/
public double [] linear3d(double [][][] data){
/*
* Approximating each of the 3 measured parameters (Far/near, tilt x and tilt y) with linear approximation in the vicinity of the last position
......@@ -102,10 +107,10 @@ public class PolynomialApproximation {
if (this.debugLevel>3){
System.out.println(
parNum+","+data[nSample][0][0]+","+data[nSample][0][1]+","+data[nSample][0][2]+","+
data[nSample][1][0]+","+data[nSample][1][1]+","+data[nSample][1][2]+","+data[nSample][2][0]);
data[nSample][1][0]+","+data[nSample][1][1]+","+data[nSample][1][2]+","+data[nSample][2][0]);
}
//TODO replace with PolynomialApproximation class
//TODO replace with PolynomialApproximation class
double w=(data[nSample].length>2)?data[nSample][2][0]:1.0;
double [] xyz=data[nSample][0];
S0+=w;
......@@ -168,151 +173,151 @@ public class PolynomialApproximation {
}
public double[] quadraticMax2d (double [][][] data){
return quadraticMax2d (data,1.0E-15);
}
public double[] quadraticMax2d (double [][][] data, double thresholdQuad){
return quadraticMax2d (data,thresholdQuad, debugLevel);
}
public double[] quadraticMax2d (double [][][] data,double thresholdQuad, int debugLevel){
double [][] coeff=quadraticApproximation(data, false, 1.0E-20, thresholdQuad,debugLevel);
if (coeff==null) return null;
if (coeff[0].length<6) return null;
double [][] aM={
{2*coeff[0][0], coeff[0][2]}, // | 2A, C |
{ coeff[0][2],2*coeff[0][1]}}; // | C, 2B |
Matrix M=(new Matrix(aM));
double nmQ=normMatix(aM);
if (debugLevel>3) System.out.println("M.det()="+M.det()+" normMatix(aM)="+nmQ+" data.length="+data.length);
if ((nmQ==0.0) || (Math.abs(M.det())/normMatix(aM)<thresholdQuad)) {
if (debugLevel>3) System.out.println("quadraticMax2d() failed: M.det()="+M.det()+" normMatix(aM)="+normMatix(aM));
return null;
}
double [][] aB={
{-coeff[0][3]}, // | - D |
{-coeff[0][4]}}; // | - E |
double [] xy=M.solve(new Matrix(aB)).getColumnPackedCopy();
return xy;
}
// returns maximum's coordinates, value, and coefficients for x2, y2 and xy
public double[] quadraticMaxV2dX2Y2XY (double [][][] data,double thresholdQuad, int debugLevel){
double [][] coeff=quadraticApproximation(data, false, 1.0E-20, thresholdQuad,debugLevel);
if (coeff==null) return null;
if (coeff[0].length<6) return null;
double [][] aM={
{2*coeff[0][0], coeff[0][2]}, // | 2A, C |
{ coeff[0][2],2*coeff[0][1]}}; // | C, 2B |
Matrix M=(new Matrix(aM));
double nmQ=normMatix(aM);
if (debugLevel>3) System.out.println("M.det()="+M.det()+" normMatix(aM)="+nmQ+" data.length="+data.length);
if ((nmQ==0.0) || (Math.abs(M.det())/normMatix(aM)<thresholdQuad)) {
if (debugLevel>3) System.out.println("quadraticMax2d() failed: M.det()="+M.det()+" normMatix(aM)="+normMatix(aM));
return null;
}
double [][] aB={
{-coeff[0][3]}, // | - D |
{-coeff[0][4]}}; // | - E |
double [] xy=M.solve(new Matrix(aB)).getColumnPackedCopy();
double vmax = coeff[0][0]* xy[0]*xy[0]+coeff[0][1]*xy[1]*xy[1]+coeff[0][2]*xy[0]*xy[1]+coeff[0][3]*xy[0]+coeff[0][4]*xy[1]+coeff[0][5];
double [] xyx2y2 = {xy[0], xy[1], vmax, coeff[0][0], coeff[0][1], coeff[0][2], coeff[0][3], coeff[0][4], coeff[0][5]};
return xyx2y2;
}
public double[] quadraticMax2d (double [][][] data){
return quadraticMax2d (data,1.0E-15);
}
public double[] quadraticMax2d (double [][][] data, double thresholdQuad){
return quadraticMax2d (data,thresholdQuad, debugLevel);
}
public double[] quadraticMax2d (double [][][] data,double thresholdQuad, int debugLevel){
double [][] coeff=quadraticApproximation(data, false, 1.0E-20, thresholdQuad,debugLevel);
if (coeff==null) return null;
if (coeff[0].length<6) return null;
double [][] aM={
{2*coeff[0][0], coeff[0][2]}, // | 2A, C |
{ coeff[0][2],2*coeff[0][1]}}; // | C, 2B |
Matrix M=(new Matrix(aM));
double nmQ=normMatix(aM);
if (debugLevel>3) System.out.println("M.det()="+M.det()+" normMatix(aM)="+nmQ+" data.length="+data.length);
if ((nmQ==0.0) || (Math.abs(M.det())/normMatix(aM)<thresholdQuad)) {
if (debugLevel>3) System.out.println("quadraticMax2d() failed: M.det()="+M.det()+" normMatix(aM)="+normMatix(aM));
return null;
}
double [][] aB={
{-coeff[0][3]}, // | - D |
{-coeff[0][4]}}; // | - E |
double [] xy=M.solve(new Matrix(aB)).getColumnPackedCopy();
return xy;
}
// returns maximum's coordinates, value, and coefficients for x2, y2 and xy
public double[] quadraticMaxV2dX2Y2XY (double [][][] data,double thresholdQuad, int debugLevel){
double [][] coeff=quadraticApproximation(data, false, 1.0E-20, thresholdQuad,debugLevel);
if (coeff==null) return null;
if (coeff[0].length<6) return null;
double [][] aM={
{2*coeff[0][0], coeff[0][2]}, // | 2A, C |
{ coeff[0][2],2*coeff[0][1]}}; // | C, 2B |
Matrix M=(new Matrix(aM));
double nmQ=normMatix(aM);
if (debugLevel>3) System.out.println("M.det()="+M.det()+" normMatix(aM)="+nmQ+" data.length="+data.length);
if ((nmQ==0.0) || (Math.abs(M.det())/normMatix(aM)<thresholdQuad)) {
if (debugLevel>3) System.out.println("quadraticMax2d() failed: M.det()="+M.det()+" normMatix(aM)="+normMatix(aM));
return null;
}
double [][] aB={
{-coeff[0][3]}, // | - D |
{-coeff[0][4]}}; // | - E |
double [] xy=M.solve(new Matrix(aB)).getColumnPackedCopy();
double vmax = coeff[0][0]* xy[0]*xy[0]+coeff[0][1]*xy[1]*xy[1]+coeff[0][2]*xy[0]*xy[1]+coeff[0][3]*xy[0]+coeff[0][4]*xy[1]+coeff[0][5];
double [] xyx2y2 = {xy[0], xy[1], vmax, coeff[0][0], coeff[0][1], coeff[0][2], coeff[0][3], coeff[0][4], coeff[0][5]};
return xyx2y2;
}
/* ======================================================================== */
/**
* See below, this version is for backward compatibility with no damping
* @param data
* @param forceLinear
* @return
*/
public double [][] quadraticApproximation(
double [][][] data,
boolean forceLinear) // use linear approximation
{
return quadraticApproximation(
data,
forceLinear,
null);
}
/**
* See below, this version is for backward compatibility with no damping
* @param data
* @param forceLinear
* @return
*/
public double [][] quadraticApproximation(
double [][][] data,
boolean forceLinear) // use linear approximation
{
return quadraticApproximation(
data,
forceLinear,
null);
}
/**
* Approximate function z(x,y) as a second degree polynomial (or just linear)
* f(x,y)=A*x^2+B*y^2+C*x*y+D*x+E*y+F or f(x,y)=D*x+E*y+F
* @param data array consists of lines of either 2 or 3 vectors:
* 2-element vector x,y
* variable length vector z (should be the same for all samples)
* optional 1- element vector w (weight of the sample)
* @param forceLinear force linear approximation
* @param damping optional (may be null) array of 3 (for linear) or 6 (quadratic) values that adds cost
* for corresponding coefficient be non-zero. This can be used to find reasonable solutions when
* data is insufficient. Just one data point would result in a plane of constant z, co-linear
* points will produce a plane with a gradient along this line
* @return array of vectors or null
* each vector (one per each z component) is either 6-element- (A,B,C,D,E,F) if quadratic is possible and enabled
* or 3-element - (D,E,F) if linear is possible and quadratic is not possible or disabled
* returns null if not enough data even for the linear approximation
*/
public double [][] quadraticApproximation(
double [][][] data,
boolean forceLinear, // use linear approximation
double [] damping)
{
return quadraticApproximation(
data,
forceLinear, // use linear approximation
damping,
this.debugLevel);
}
/**
* Approximate function z(x,y) as a second degree polynomial (or just linear)
* f(x,y)=A*x^2+B*y^2+C*x*y+D*x+E*y+F or f(x,y)=D*x+E*y+F
* @param data array consists of lines of either 2 or 3 vectors:
* 2-element vector x,y
* variable length vector z (should be the same for all samples)
* optional 1- element vector w (weight of the sample)
* @param forceLinear force linear approximation
* @param damping optional (may be null) array of 3 (for linear) or 6 (quadratic) values that adds cost
* for corresponding coefficient be non-zero. This can be used to find reasonable solutions when
* data is insufficient. Just one data point would result in a plane of constant z, co-linear
* points will produce a plane with a gradient along this line
* @return array of vectors or null
* each vector (one per each z component) is either 6-element- (A,B,C,D,E,F) if quadratic is possible and enabled
* or 3-element - (D,E,F) if linear is possible and quadratic is not possible or disabled
* returns null if not enough data even for the linear approximation
*/
public double [][] quadraticApproximation(
double [][][] data,
boolean forceLinear, // use linear approximation
double [] damping)
{
return quadraticApproximation(
data,
forceLinear, // use linear approximation
damping,
this.debugLevel);
}
public double [][] quadraticApproximation( // no use
double [][][] data,
boolean forceLinear, // use linear approximation
int debugLevel
){
return quadraticApproximation(
data,
forceLinear, // use linear approximation
null,
debugLevel);
}
public double [][] quadraticApproximation( // no use
double [][][] data,
boolean forceLinear, // use linear approximation
int debugLevel
){
return quadraticApproximation(
data,
forceLinear, // use linear approximation
null,
debugLevel);
}
public double [][] quadraticApproximation(
double [][][] data,
boolean forceLinear, // use linear approximation
double [] damping,
int debugLevel
){
return quadraticApproximation(
data,
forceLinear, // use linear approximation
damping,
1.0E-10, // threshold ratio of matrix determinant to norm for linear approximation (det too low - fail) 11.0E-10 failed where it shouldn't?
1.0E-15, // threshold ratio of matrix determinant to norm for quadratic approximation (det too low - fail)
debugLevel);
}
public double [][] quadraticApproximation(
double [][][] data,
boolean forceLinear, // use linear approximation
double [] damping,
int debugLevel
){
return quadraticApproximation(
data,
forceLinear, // use linear approximation
damping,
1.0E-10, // threshold ratio of matrix determinant to norm for linear approximation (det too low - fail) 11.0E-10 failed where it shouldn't?
1.0E-15, // threshold ratio of matrix determinant to norm for quadratic approximation (det too low - fail)
debugLevel);
}
public double [][] quadraticApproximation(
double [][][] data,
boolean forceLinear, // use linear approximation
double thresholdLin, // threshold ratio of matrix determinant to norm for linear approximation (det too low - fail)
double thresholdQuad) // threshold ratio of matrix determinant to norm for quadratic approximation (det too low - fail)
{
return quadraticApproximation(
data,
forceLinear, // use linear approximation
null,
1.0E-10, // threshold ratio of matrix determinant to norm for linear approximation (det too low - fail) 11.0E-10 failed where it shouldn't?
1.0E-15, // threshold ratio of matrix determinant to norm for quadratic approximation (det too low - fail)
this.debugLevel);
}
/*
public double [][] quadraticApproximation(
double [][][] data,
boolean forceLinear, // use linear approximation
double thresholdLin, // threshold ratio of matrix determinant to norm for linear approximation (det too low - fail)
double thresholdQuad) // threshold ratio of matrix determinant to norm for quadratic approximation (det too low - fail)
{
return quadraticApproximation(
data,
forceLinear, // use linear approximation
null,
1.0E-10, // threshold ratio of matrix determinant to norm for linear approximation (det too low - fail) 11.0E-10 failed where it shouldn't?
1.0E-15, // threshold ratio of matrix determinant to norm for quadratic approximation (det too low - fail)
this.debugLevel);
}
/*
public double [][] quadraticApproximation( // no use
double [][][] data,
boolean forceLinear, // use linear approximation
......@@ -328,39 +333,39 @@ public class PolynomialApproximation {
1.0E-15, // threshold ratio of matrix determinant to norm for quadratic approximation (det too low - fail)
this.debugLevel);
}
*/
public double [][] quadraticApproximation(
double [][][] data,
boolean forceLinear, // use linear approximation
double thresholdLin, // threshold ratio of matrix determinant to norm for linear approximation (det too low - fail)
double thresholdQuad, // threshold ratio of matrix determinant to norm for quadratic approximation (det too low - fail)
int debugLevel)
{
return quadraticApproximation(
data,
forceLinear, // use linear approximation
null,
thresholdLin, // threshold ratio of matrix determinant to norm for linear approximation (det too low - fail) 11.0E-10 failed where it shouldn't?
thresholdQuad, // threshold ratio of matrix determinant to norm for quadratic approximation (det too low - fail)
debugLevel);
*/
public double [][] quadraticApproximation(
double [][][] data,
boolean forceLinear, // use linear approximation
double thresholdLin, // threshold ratio of matrix determinant to norm for linear approximation (det too low - fail)
double thresholdQuad, // threshold ratio of matrix determinant to norm for quadratic approximation (det too low - fail)
int debugLevel)
{
return quadraticApproximation(
data,
forceLinear, // use linear approximation
null,
thresholdLin, // threshold ratio of matrix determinant to norm for linear approximation (det too low - fail) 11.0E-10 failed where it shouldn't?
thresholdQuad, // threshold ratio of matrix determinant to norm for quadratic approximation (det too low - fail)
debugLevel);
}
}
public double [][] quadraticApproximation(
double [][][] data,
boolean forceLinear, // use linear approximation
double [] damping,
double thresholdLin, // threshold ratio of matrix determinant to norm for linear approximation (det too low - fail)
double thresholdQuad, // threshold ratio of matrix determinant to norm for quadratic approximation (det too low - fail)
int debugLevel
){
if (debugLevel>3) System.out.println("quadraticApproximation(...), debugLevel="+debugLevel+":");
if ((data == null) || (data.length == 0)) {
return null;
}
/* ix, iy - the location of the point with maximal value. We'll approximate the vicinity of that maximum using a
* second degree polynomial:
public double [][] quadraticApproximation(
double [][][] data,
boolean forceLinear, // use linear approximation
double [] damping,
double thresholdLin, // threshold ratio of matrix determinant to norm for linear approximation (det too low - fail)
double thresholdQuad, // threshold ratio of matrix determinant to norm for quadratic approximation (det too low - fail)
int debugLevel
){
if (debugLevel>3) System.out.println("quadraticApproximation(...), debugLevel="+debugLevel+":");
if ((data == null) || (data.length == 0)) {
return null;
}
/* ix, iy - the location of the point with maximal value. We'll approximate the vicinity of that maximum using a
* second degree polynomial:
Z(x,y)~=A*x^2+B*y^2+C*x*y+D*x+E*y+F
by minimizing sum of squared differenceS00between the actual (Z(x,uy)) and approximated values.
and then find the maximum on the approximated surface. Here iS00the math:
......@@ -424,183 +429,338 @@ public class PolynomialApproximation {
(4) 0= A*S30 + B*S12 + C*S21 + D*S20 + E*S11 + F*S10 - SZ10
(5) 0= A*S21 + B*S03 + C*S12 + D*S11 + E*S02 + F*S01 - SZ01
(6) 0= A*S20 + B*S02 + C*S11 + D*S10 + E*S01 + F*S00 - SZ00
*/
Matrix mDampingLin = null;
Matrix mDampingQuad = null;
if (damping != null){
mDampingLin = new Matrix(3,3);
for (int i = 0; i < 3; i++){
int j = damping.length + i - 3;
if (j >= 0) mDampingLin.set(i,i,damping[j]);
}
if (!forceLinear) {
mDampingQuad = new Matrix(6,6);
for (int i = 0; i < 6; i++){
int j = damping.length + i - 6;
if (j >= 0) mDampingQuad.set(i,i,damping[j]);
}
}
}
int zDim = 0; // =data[0][1].length;
for (int i = 0; i < data.length; i++) if (data[i] != null){
zDim =data[i][1].length;
break;
}
double w,z,x,x2,x3,x4,y,y2,y3,y4,wz;
int i,j,n=0;
double S00=0.0,
S10=0.0,S01=0.0,
S20=0.0,S11=0.0,S02=0.0,
S30=0.0,S21=0.0,S12=0.0,S03=0.0,
S40=0.0,S31=0.0,S22=0.0,S13=0.0,S04=0.0;
double [] SZ00=new double [zDim];
double [] SZ01=new double [zDim];
double [] SZ10=new double [zDim];
double [] SZ11=new double [zDim];
double [] SZ02=new double [zDim];
double [] SZ20=new double [zDim];
for (i=0;i<zDim;i++) {
SZ00[i]=0.0;
SZ01[i]=0.0;
SZ10[i]=0.0;
SZ11[i]=0.0;
SZ02[i]=0.0;
SZ20[i]=0.0;
}
for (i=0;i<data.length;i++) if (data[i] != null){
w=(data[i].length>2)? data[i][2][0]:1.0;
if (w>0) {
n++;
x=data[i][0][0];
y=data[i][0][1];
x2=x*x;
y2=y*y;
S00+=w;
S10+=w*x;
S01+=w*y;
S11+=w*x*y;
S20+=w*x2;
S02+=w*y2;
if (!forceLinear) {
x3=x2*x;
x4=x3*x;
y3=y2*y;
y4=y3*y;
S30+=w*x3;
S21+=w*x2*y;
S12+=w*x*y2;
S03+=w*y3;
S40+=w*x4;
S31+=w*x3*y;
S22+=w*x2*y2;
S13+=w*x*y3;
S04+=w*y4;
}
for (j=0;j<zDim;j++) {
z=data[i][1][j];
wz=w*z;
SZ00[j]+=wz;
SZ10[j]+=wz*x;
SZ01[j]+=wz*y;
if (!forceLinear) {
SZ20[j]+=wz*x2;
SZ11[j]+=wz*x*y;
SZ02[j]+=wz*y2;
}
}
}
}
//need to decide if there is enough data for linear and quadratic
double [][] mAarrayL= {
{S20,S11,S10},
{S11,S02,S01},
{S10,S01,S00}};
Matrix mLin=new Matrix (mAarrayL);
if (mDampingLin != null){
mLin.plusEquals(mDampingLin);
}
// TODO Maybe bypass determinant checks for damped ?
Matrix Z;
if (debugLevel>3) System.out.println(">>> n="+n+" det_lin="+mLin.det()+" norm_lin="+normMatix(mAarrayL));
double nmL=normMatix(mAarrayL);
if ((nmL==0.0) || (Math.abs(mLin.det())/nmL<thresholdLin)){
// return average value for each channel
if (S00==0.0) return null; // not even average
double [][] ABCDEF=new double[zDim][3];
for (i=0;i<zDim;i++) {
ABCDEF[i][0]=0.0;
ABCDEF[i][1]=0.0;
ABCDEF[i][2]=SZ00[i]/S00;
}
return ABCDEF;
}
double []zAarrayL=new double [3];
double [][] ABCDEF=new double[zDim][];
// double [] zAarrayL={SZ10,SZ01,SZ00};
for (i=0;i<zDim;i++) {
zAarrayL[0]=SZ10[i];
zAarrayL[1]=SZ01[i];
zAarrayL[2]=SZ00[i];
Z=new Matrix (zAarrayL,3);
ABCDEF[i]= mLin.solve(Z).getRowPackedCopy();
}
if (forceLinear) return ABCDEF;
// quote try quadratic approximation
double [][] mAarrayQ= {
{S40,S22,S31,S30,S21,S20},
{S22,S04,S13,S12,S03,S02},
{S31,S13,S22,S21,S12,S11},
{S30,S12,S21,S20,S11,S10},
{S21,S03,S12,S11,S02,S01},
{S20,S02,S11,S10,S01,S00}};
Matrix mQuad=new Matrix (mAarrayQ);
if (mDampingQuad != null){
mQuad.plusEquals(mDampingQuad);
}
if (debugLevel>3) {
System.out.println(" n="+n+" det_quad="+mQuad.det()+" norm_quad="+normMatix(mAarrayQ)+" data.length="+data.length);
mQuad.print(10,5);
}
double nmQ=normMatix(mAarrayQ);
if ((nmQ==0.0) || (Math.abs(mQuad.det())/normMatix(mAarrayQ)<thresholdQuad)) {
if (debugLevel>0) System.out.println("Using linear approximation, M.det()="+mQuad.det()+
" normMatix(mAarrayQ)="+normMatix(mAarrayQ)+
", thresholdQuad="+thresholdQuad+
", nmQ="+nmQ+
", Math.abs(M.det())/normMatix(mAarrayQ)="+(Math.abs(mQuad.det())/normMatix(mAarrayQ))); //did not happen
return ABCDEF; // not enough data for the quadratic approximation, return linear
}
// double [] zAarrayQ={SZ20,SZ02,SZ11,SZ10,SZ01,SZ00};
double [] zAarrayQ=new double [6];
for (i=0;i<zDim;i++) {
zAarrayQ[0]=SZ20[i];
zAarrayQ[1]=SZ02[i];
zAarrayQ[2]=SZ11[i];
zAarrayQ[3]=SZ10[i];
zAarrayQ[4]=SZ01[i];
zAarrayQ[5]=SZ00[i];
Z=new Matrix (zAarrayQ,6);
ABCDEF[i]= mQuad.solve(Z).getRowPackedCopy();
}
return ABCDEF;
}
// calculate "volume" made of the matrix row-vectors, placed orthogonally
// to be compared to determinant
public double normMatix(double [][] a) {
double d,norm=1.0;
for (int i=0;i<a.length;i++) {
d=0;
for (int j=0;j<a[i].length;j++) d+=a[i][j]*a[i][j];
norm*=Math.sqrt(d);
}
return norm;
*/
Matrix mDampingLin = null;
Matrix mDampingQuad = null;
if (damping != null){
mDampingLin = new Matrix(3,3);
for (int i = 0; i < 3; i++){
int j = damping.length + i - 3;
if (j >= 0) mDampingLin.set(i,i,damping[j]);
}
if (!forceLinear) {
mDampingQuad = new Matrix(6,6);
for (int i = 0; i < 6; i++){
int j = damping.length + i - 6;
if (j >= 0) mDampingQuad.set(i,i,damping[j]);
}
}
}
int zDim = 0; // =data[0][1].length;
for (int i = 0; i < data.length; i++) if (data[i] != null){
zDim =data[i][1].length;
break;
}
double w,z,x,x2,x3,x4,y,y2,y3,y4,wz;
int i,j,n=0;
double S00=0.0,
S10=0.0,S01=0.0,
S20=0.0,S11=0.0,S02=0.0,
S30=0.0,S21=0.0,S12=0.0,S03=0.0,
S40=0.0,S31=0.0,S22=0.0,S13=0.0,S04=0.0;
double [] SZ00=new double [zDim];
double [] SZ01=new double [zDim];
double [] SZ10=new double [zDim];
double [] SZ11=new double [zDim];
double [] SZ02=new double [zDim];
double [] SZ20=new double [zDim];
for (i=0;i<zDim;i++) {
SZ00[i]=0.0;
SZ01[i]=0.0;
SZ10[i]=0.0;
SZ11[i]=0.0;
SZ02[i]=0.0;
SZ20[i]=0.0;
}
for (i=0;i<data.length;i++) if (data[i] != null){
w=(data[i].length>2)? data[i][2][0]:1.0;
if (w>0) {
n++;
x=data[i][0][0];
y=data[i][0][1];
x2=x*x;
y2=y*y;
S00+=w;
S10+=w*x;
S01+=w*y;
S11+=w*x*y;
S20+=w*x2;
S02+=w*y2;
if (!forceLinear) {
x3=x2*x;
x4=x3*x;
y3=y2*y;
y4=y3*y;
S30+=w*x3;
S21+=w*x2*y;
S12+=w*x*y2;
S03+=w*y3;
S40+=w*x4;
S31+=w*x3*y;
S22+=w*x2*y2;
S13+=w*x*y3;
S04+=w*y4;
}
for (j=0;j<zDim;j++) {
z=data[i][1][j];
wz=w*z;
SZ00[j]+=wz;
SZ10[j]+=wz*x;
SZ01[j]+=wz*y;
if (!forceLinear) {
SZ20[j]+=wz*x2;
SZ11[j]+=wz*x*y;
SZ02[j]+=wz*y2;
}
}
}
}
//need to decide if there is enough data for linear and quadratic
double [][] mAarrayL= {
{S20,S11,S10},
{S11,S02,S01},
{S10,S01,S00}};
Matrix mLin=new Matrix (mAarrayL);
if (mDampingLin != null){
mLin.plusEquals(mDampingLin);
}
// TODO Maybe bypass determinant checks for damped ?
Matrix Z;
if (debugLevel>3) System.out.println(">>> n="+n+" det_lin="+mLin.det()+" norm_lin="+normMatix(mAarrayL));
double nmL=normMatix(mAarrayL);
if ((nmL==0.0) || (Math.abs(mLin.det())/nmL<thresholdLin)){
// return average value for each channel
if (S00==0.0) return null; // not even average
double [][] ABCDEF=new double[zDim][3];
for (i=0;i<zDim;i++) {
ABCDEF[i][0]=0.0;
ABCDEF[i][1]=0.0;
ABCDEF[i][2]=SZ00[i]/S00;
}
return ABCDEF;
}
double []zAarrayL=new double [3];
double [][] ABCDEF=new double[zDim][];
// double [] zAarrayL={SZ10,SZ01,SZ00};
for (i=0;i<zDim;i++) {
zAarrayL[0]=SZ10[i];
zAarrayL[1]=SZ01[i];
zAarrayL[2]=SZ00[i];
Z=new Matrix (zAarrayL,3);
ABCDEF[i]= mLin.solve(Z).getRowPackedCopy();
}
if (forceLinear) return ABCDEF;
// quote try quadratic approximation
double [][] mAarrayQ= {
{S40,S22,S31,S30,S21,S20},
{S22,S04,S13,S12,S03,S02},
{S31,S13,S22,S21,S12,S11},
{S30,S12,S21,S20,S11,S10},
{S21,S03,S12,S11,S02,S01},
{S20,S02,S11,S10,S01,S00}};
Matrix mQuad=new Matrix (mAarrayQ);
if (mDampingQuad != null){
mQuad.plusEquals(mDampingQuad);
}
if (debugLevel>3) {
System.out.println(" n="+n+" det_quad="+mQuad.det()+" norm_quad="+normMatix(mAarrayQ)+" data.length="+data.length);
mQuad.print(10,5);
}
double nmQ=normMatix(mAarrayQ);
if ((nmQ==0.0) || (Math.abs(mQuad.det())/normMatix(mAarrayQ)<thresholdQuad)) {
if (debugLevel>0) System.out.println("Using linear approximation, M.det()="+mQuad.det()+
" normMatix(mAarrayQ)="+normMatix(mAarrayQ)+
", thresholdQuad="+thresholdQuad+
", nmQ="+nmQ+
", Math.abs(M.det())/normMatix(mAarrayQ)="+(Math.abs(mQuad.det())/normMatix(mAarrayQ))); //did not happen
return ABCDEF; // not enough data for the quadratic approximation, return linear
}
// double [] zAarrayQ={SZ20,SZ02,SZ11,SZ10,SZ01,SZ00};
double [] zAarrayQ=new double [6];
for (i=0;i<zDim;i++) {
zAarrayQ[0]=SZ20[i];
zAarrayQ[1]=SZ02[i];
zAarrayQ[2]=SZ11[i];
zAarrayQ[3]=SZ10[i];
zAarrayQ[4]=SZ01[i];
zAarrayQ[5]=SZ00[i];
Z=new Matrix (zAarrayQ,6);
ABCDEF[i]= mQuad.solve(Z).getRowPackedCopy();
}
return ABCDEF;
}
// calculate "volume" made of the matrix row-vectors, placed orthogonally
// to be compared to determinant
public double normMatix(double [][] a) {
double d,norm=1.0;
for (int i=0;i<a.length;i++) {
d=0;
for (int j=0;j<a[i].length;j++) d+=a[i][j]*a[i][j];
norm*=Math.sqrt(d);
}
return norm;
}
//RuntimeException
public static double [] getYXRegression(
final double [] data_x,
final double [] data_y,
final boolean [] mask) {
final Thread[] threads = ImageDtt.newThreadArray(QuadCLT.THREADS_MAX);
final AtomicInteger ai = new AtomicInteger(0);
final double [] as0 = new double[threads.length];
final double [] asx = new double[threads.length];
final double [] asx2 = new double[threads.length];
final double [] asy = new double[threads.length];
final double [] asxy= new double[threads.length];
final AtomicInteger ati = new AtomicInteger(0);
ai.set(0);
for (int ithread = 0; ithread < threads.length; ithread++) {
threads[ithread] = new Thread() {
public void run() {
int thread_num = ati.getAndIncrement();
for (int ipix = ai.getAndIncrement(); ipix < data_x.length; ipix = ai.getAndIncrement()) if ((mask == null) || mask[ipix]){
double x = data_x[ipix];
double y = data_y[ipix];
if (!Double.isNaN(x) && !Double.isNaN(y)) {
as0 [thread_num] += 1;
asx [thread_num] += x;
asx2[thread_num] += x * x;
asy [thread_num] += y;
asxy[thread_num] += x * y;
}
} // for (int ipix
}
};
}
ImageDtt.startAndJoin(threads);
double s0 = 0.0;
double sx = 0.0;
double sx2 = 0.0;
double sy = 0.0;
double sxy= 0.0;
for (int i = 0; i < threads.length; i++) {
s0+= as0[i];
sx+= asx[i];
sx2+= asx2[i];
sy+= asy[i];
sxy+= asxy[i];
}
double dnm = s0 * sx2 - sx*sx;
double a = (sxy * s0 - sy * sx) / dnm;
double b = (sy * sx2 - sxy * sx) / dnm;
return new double[] {a,b};
}
/**
* Get best fit ax+b symmetrical for X and Y, same weights
* https://en.wikipedia.org/wiki/Deming_regression#Orthogonal_regression
* @param data_x
* @param data_y
* @param mask
* @return
*/
public static double [] getOrthoRegression( // symmetrical for X and Y, same eror weights
final double [] data_x,
final double [] data_y,
final boolean [] mask_in) {
final boolean [] mask = new boolean [data_x.length];
final Thread[] threads = ImageDtt.newThreadArray(QuadCLT.THREADS_MAX);
final AtomicInteger ai = new AtomicInteger(0);
final double [] as0 = new double[threads.length];
final double [] asx = new double[threads.length];
final double [] asy = new double[threads.length];
final AtomicInteger ati = new AtomicInteger(0);
ai.set(0);
// Find centroid first
for (int ithread = 0; ithread < threads.length; ithread++) {
threads[ithread] = new Thread() {
public void run() {
int thread_num = ati.getAndIncrement();
for (int ipix = ai.getAndIncrement(); ipix < data_x.length; ipix = ai.getAndIncrement()) if ((mask_in == null) || mask_in[ipix]){
double x = data_x[ipix];
double y = data_y[ipix];
if (!Double.isNaN(x) && !Double.isNaN(y)) {
as0 [thread_num] += 1;
asx [thread_num] += x;
asy [thread_num] += y;
mask[ipix] = true;
}
} // for (int ipix
}
};
}
ImageDtt.startAndJoin(threads);
double s0 = 0.0;
double sx = 0.0;
double sy = 0.0;
for (int i = 0; i < threads.length; i++) {
s0+= as0[i];
sx+= asx[i];
sy+= asy[i];
}
double [] z_mean = {sx/s0, sy/s0}; // complex
final double [] as_re = new double[threads.length];
final double [] as_im = new double[threads.length];
ai.set(0);
ati.set(0);
for (int ithread = 0; ithread < threads.length; ithread++) {
threads[ithread] = new Thread() {
public void run() {
int thread_num = ati.getAndIncrement();
for (int ipix = ai.getAndIncrement(); ipix < data_x.length; ipix = ai.getAndIncrement()) if ((mask == null) || mask[ipix]){
double x = data_x[ipix]-z_mean[0];
double y = data_y[ipix]-z_mean[1];
as_re[thread_num] += x*x - y*y;
as_im[thread_num] += 2*x*y;
} // for (int ipix
}
};
}
ImageDtt.startAndJoin(threads);
double s_re = 0.0;
double s_im = 0.0;
for (int i = 0; i < threads.length; i++) {
s_re+= as_re[i];
s_im+= as_im[i];
}
// https://en.wikipedia.org/wiki/Square_root#Algebraic_formula
// sqrt (s_re+i*s_im)
double sqrt_re = Math.sqrt(0.5 * (Math.sqrt(s_re*s_re + s_im*s_im) +s_re));
double sqrt_im = ((s_im > 0)? 1 : -1) * Math.sqrt(0.5 * (Math.sqrt(s_re*s_re + s_im*s_im) - s_re));
double a = sqrt_im/sqrt_re;
double b = z_mean[1]- a * z_mean[0];
return new double[] {a,b};
}
public static void applyRegression(
final double [] data, // clone by caller
final double [] regression) {
final Thread[] threads = ImageDtt.newThreadArray(QuadCLT.THREADS_MAX);
final AtomicInteger ai = new AtomicInteger(0);
for (int ithread = 0; ithread < threads.length; ithread++) {
threads[ithread] = new Thread() {
public void run() {
for (int ipix = ai.getAndIncrement(); ipix < data.length; ipix = ai.getAndIncrement()){
data[ipix] = regression[0] * data[ipix] + regression[1];
}
}
};
}
ImageDtt.startAndJoin(threads);
}
public static double [] invertRegression(double [] regression) {
return new double [] {1.0/regression[0], -regression[1]/regression[0]};
}
}
src/main/java/com/elphel/imagej/orthomosaic/ComboMatch.java
View file @ 1671728b
......@@ -171,7 +171,7 @@ public class ComboMatch {
boolean pattern_match = true; // false;
boolean bounds_to_indices = true;
int temp_mode = 1;
int temp_mode = 0;
boolean restore_temp = true;
double frac_remove = clt_parameters.imp.pmtch_frac_remove; // 0.15;
double metric_error = clt_parameters.imp.pmtch_metric_err; // 0.05; // 0.02;// 2 cm
......@@ -1030,10 +1030,13 @@ public class ComboMatch {
}
if (render_match) {
String title=String.format("multi_%03d-%03d_%s-%s_zoom%d_%d",gpu_pair[0],gpu_pair[1],gpu_spair[0],gpu_spair[1],min_zoom_lev,zoom_lev);
// Avoid renderMulti() - it duplicates code renderMultiDouble()
int eq_mode = 2; // calculate
ImagePlus imp_img_pair = maps_collection.renderMulti (
//_zoom<integer> is needed for opening with "Extract Objects" command
title, // String title,
OrthoMapsCollection.MODE_IMAGE, // int mode, // 0 - regular image, 1 - altitudes, 2 - black/white mask // boolean use_alt,
title, // String title,
// OrthoMapsCollection.MODE_IMAGE, // int mode, // 0 - regular image, 1 - altitudes, 2 - black/white mask // boolean use_alt,
eq_mode, //int eq_mode, // 0 - ignore equalization, 1 - use stored equalization, 2 - calculate equalization
gpu_pair, // int [] indices, // null or which indices to use (normally just 2 for pairwise comparison)
bounds_to_indices, // boolean bounds_to_indices,
temp_mode, // int temp_mode, // 0 - do nothing, 1 - equalize average,2 - try to correct
......
src/main/java/com/elphel/imagej/orthomosaic/OrthoMap.java
View file @ 1671728b
......@@ -115,6 +115,7 @@ public class OrthoMap implements Comparable <OrthoMap>, Serializable{
public transient double agl = Double.NaN;
public transient int num_scenes = -1;; // number of scenes that made up this image
public transient double sfm_gain = Double.NaN; // maximal SfM gain of this map
public transient double [] equalize = {1,0}; // rectified value = equalize[0]*source_value+equalize[1]
private void writeObject(ObjectOutputStream oos) throws IOException {
oos.defaultWriteObject();
oos.writeObject(path);
......@@ -137,8 +138,9 @@ public class OrthoMap implements Comparable <OrthoMap>, Serializable{
oos.writeObject(agl);
oos.writeObject(num_scenes);
oos.writeObject(sfm_gain);
oos.writeObject(equalize);
}
private void readObject(ObjectInputStream ois) throws ClassNotFoundException, IOException {
// sfm_gain = Double.NaN;
// num_scenes = -1;
......@@ -160,12 +162,26 @@ public class OrthoMap implements Comparable <OrthoMap>, Serializable{
agl = (double) ois.readObject();
num_scenes = (int) ois.readObject();
sfm_gain = (double) ois.readObject();
// equalize = new double[] {1,0};
equalize = (double []) ois.readObject();
images = new HashMap <Integer, FloatImageData>(); // field images was not saved
averageImagePixel = Double.NaN; // average image pixel value (to combine with raw)
// pairwise_matches is not transient
// pairwise_matches = new HashMap<Double, PairwiseOrthoMatch>();
}
double getEqualized(double d) {
return d * equalize[0] + equalize[1];
}
double [] getEqualize() {
return equalize;
}
void setEqualize(double [] equalize) {
this.equalize = equalize;
}
@Override
public int compareTo(OrthoMap otherPlayer) {
......@@ -1717,9 +1733,9 @@ public class OrthoMap implements Comparable <OrthoMap>, Serializable{
};
}
ImageDtt.startAndJoin(threads);
double [] ab = getDatiRegression(
temp, // final double [] temp,
dati, // final double [] dati,
double [] ab = PolynomialApproximation.getYXRegression(
temp, // final double [] data_x,
dati, // final double [] data_y,
flat); // final boolean [] mask);
double a = ab[0];
double b = ab[1];
......@@ -1774,56 +1790,6 @@ public class OrthoMap implements Comparable <OrthoMap>, Serializable{
return flat;
}
private static double [] getDatiRegression(
final double [] temp,
final double [] dati,
final boolean [] mask) {
final Thread[] threads = ImageDtt.newThreadArray(QuadCLT.THREADS_MAX);
final AtomicInteger ai = new AtomicInteger(0);
final double [] as0 = new double[threads.length];
final double [] asx = new double[threads.length];
final double [] asx2 = new double[threads.length];
final double [] asy = new double[threads.length];
final double [] asxy= new double[threads.length];
final AtomicInteger ati = new AtomicInteger(0);
ai.set(0);
for (int ithread = 0; ithread < threads.length; ithread++) {
threads[ithread] = new Thread() {
public void run() {
int thread_num = ati.getAndIncrement();
for (int ipix = ai.getAndIncrement(); ipix < temp.length; ipix = ai.getAndIncrement()) if (mask[ipix]){
double x = temp[ipix];
double y = dati[ipix];
if (!Double.isNaN(x+y)) {
as0 [thread_num] += 1;
asx [thread_num] += x;
asx2[thread_num] += x * x;
asy [thread_num] += y;
asxy[thread_num] += x * y;
}
} // for (int ipix
}
};
}
ImageDtt.startAndJoin(threads);
double s0 = 0.0;
double sx = 0.0;
double sx2 = 0.0;
double sy = 0.0;
double sxy= 0.0;
for (int i = 0; i < threads.length; i++) {
s0+= as0[i];
sx+= asx[i];
sx2+= asx2[i];
sy+= asy[i];
sxy+= asxy[i];
}
double dnm = s0 * sx2 - sx*sx;
double a = (sxy * s0 - sy * sx) / dnm;
double b = (sy * sx2 - sxy * sx) / dnm;
return new double[] {a,b};
}
private static int [][] getCirclePoints(
double radius) {
......
src/main/java/com/elphel/imagej/orthomosaic/OrthoMapsCollection.java
View file @ 1671728b
......@@ -30,6 +30,7 @@ import com.elphel.imagej.calibration.CalibrationFileManagement;
import com.elphel.imagej.cameras.CLTParameters;
import com.elphel.imagej.common.DoubleGaussianBlur;
import com.elphel.imagej.common.GenericJTabbedDialog;
import com.elphel.imagej.common.PolynomialApproximation;
import com.elphel.imagej.common.ShowDoubleFloatArrays;
import com.elphel.imagej.gpu.GPUTileProcessor;
import com.elphel.imagej.gpu.TpTask;
......@@ -537,32 +538,10 @@ public class OrthoMapsCollection implements Serializable{
ol.setPixels(xy_src);
}
}
public ImagePlus renderMulti (
String title,
int mode, // 0 - regular image, 1 - altitudes, 2 - black/white mask
boolean show_centers,
int zoom_level,
int temp_mode, // 0 - do nothing, 1 - equalize average,2 - try to correct
int [] origin){
return renderMulti (
title, // String title,
mode, // int mode, // 0 - regular image, 1 - altitudes, 2 - black/white mask
null, // int [] indices, // null or which indices to use (normally just 2 for pairwise comparison)
false, // bounds_to_indices, // boolean bounds_to_indices,
temp_mode, // int temp_mode, // 0 - do nothing, 1 - equalize average,2 - try to correct
null, // double [][][] affines, // null or [indices.length][2][3]
null, // FineXYCorr warp,
show_centers, // boolean show_centers,
zoom_level, // int zoom_level,
origin); // int [] origin)
}
public ImagePlus renderMulti (
String title,
// boolean use_alt,
int mode, // 0 - regular image, 1 - altitudes, 2 - black/white mask
int eq_mode, // 0 - ignore equalization, 1 - use stored equalization, 2 - calculate equalization
int [] indices, // null or which indices to use (normally just 2 for pairwise comparison)
boolean bounds_to_indices,
int temp_mode, // 0 - do nothing, 1 - equalize average,2 - try to correct
......@@ -570,82 +549,157 @@ public class OrthoMapsCollection implements Serializable{
FineXYCorr warp, // use for a single pair only
boolean show_centers,
int zoom_level,
int [] origin){
int num_images = (indices != null)? indices.length : ortho_maps.length;
int [] wh = new int[2];
double [][] centers = new double [(indices !=null)? indices.length: ortho_maps.length][];
/*
float [][] multi = renderMulti (
// use_alt, // boolean use_alt,
mode, // int mode, // 0 - regular image, 1 - altitudes, 2 - black/white mask
OrthoMapsCollection.MODE_IMAGE, // int mode, // 0 - regular image, 1 - altitudes, 2 - black/white mask
indices, // int [] indices, // null or which indices to use (normally just 2 for pairwise comparison)
bounds_to_indices, // boolean bounds_to_indices,
affines, // double [][][] affines, // null or [indices.length][2][3]
warp, // FineXYCorr warp,,
warp, // FineXYCorr warp,
zoom_level, // int zoom_level,
wh, // int [] wh,
origin, // int [] origin){ // maps[0] as a reference
centers); // double [][] centers)
int num_images = (indices != null)? indices.length : ortho_maps.length;
String [] map_names = new String[num_images + (((num_images==2) && (mode == MODE_IMAGE))? 1 : 0)];
float [][] extra_multi = new float [map_names.length][];
*/
double [][] dmulti = renderMultiDouble (
null, // double [][] ground_planes, // null - images, non-null altitudes. use new double[2][3] for old way alt
indices, // int [] indices, // null or which indices to use (normally just 2 for pairwise comparison)
bounds_to_indices, // boolean bounds_to_indices,
affines, // affines, // double [][][] affines, // null or [indices.length][2][3]
null, // double [][] equalize,
(eq_mode != 1), // boolean ignore_equalize,
null, // warp, // FineXYCorr warp,,
zoom_level, // int zoom_level,
wh, // int [] wh,
origin, // int [] origin){ // maps[0] as a reference
centers); // double [][] centers)
if ((eq_mode == 2) && (num_images == 2)) {
boolean [] mask = new boolean [dmulti[0].length];
// int hist_min = -140,hist_max=135; // 700,200
// int hist_min = -100,hist_max=100; // 700,200
int hist_min = -250,hist_max=100; // 700,200
int hist_width= hist_max-hist_min;
double [] hist = new double [hist_width*hist_width];
for (int i = 0; i < dmulti[0].length; i++) if (!Double.isNaN(dmulti[0][i]) && !Double.isNaN(dmulti[1][i])){
int x = (int) Math.round(dmulti[0][i]) - hist_min;
int y = (int) Math.round(dmulti[1][i]) - hist_min;
if ((x>=0) && (y>=0) && (x < hist_width) && (y < hist_width)) {
int indx = y*hist_width+x;
hist[indx] += 1.0;
mask[i] = true;
}
}
ShowDoubleFloatArrays.showArrays(
hist,
hist_width,
hist_width,
title+"hist_"+hist_min+"_"+hist_max);
double [] dmulti1 = dmulti[1].clone();
double [] regression = PolynomialApproximation.getOrthoRegression(
dmulti[0], // final double [] data_x,
dmulti[1], // final double [] data_y,
mask); // final boolean [] mask)
double [] inv_regression = PolynomialApproximation.invertRegression(regression);
double [] regression10 = PolynomialApproximation.getOrthoRegression(
dmulti[1], // final double [] data_x,
dmulti[0], // final double [] data_y,
mask); // final boolean [] mask)
double [] regression00 = PolynomialApproximation.getOrthoRegression(
dmulti[0], // final double [] data_x,
dmulti[0], // final double [] data_y,
mask); // final boolean [] mask)
double [] regression11 = PolynomialApproximation.getOrthoRegression(
dmulti[1], // final double [] data_x,
dmulti[1], // final double [] data_y,
mask); // final boolean [] mask)
PolynomialApproximation.applyRegression(
dmulti1, // final double [] data, // clone by caller
inv_regression); // final double [] regression) {
double [] regression01 = PolynomialApproximation.getOrthoRegression(
dmulti[0], // final double [] data_x,
dmulti1, // final double [] data_y,
mask); // final boolean [] mask)
double [] dmulti12 = dmulti[1].clone();
for (int i = 0; i < dmulti12.length; i++) {
dmulti12[i] = 2.0*dmulti12[i]+10;
}
double [] regression012 = PolynomialApproximation.getOrthoRegression(
dmulti[0], // final double [] data_x,
dmulti12, // final double [] data_y,
mask); // final boolean [] mask)
double [] dmulti13 = dmulti[1].clone();
for (int i = 0; i < dmulti13.length; i++) {
dmulti13[i] = dmulti13[i]+10;
}
double [] regression013 = PolynomialApproximation.getOrthoRegression(
dmulti[0], // final double [] data_x,
dmulti13, // final double [] data_y,
mask); // final boolean [] mask)
double [] dmulti14 = dmulti[1].clone();
for (int i = 0; i < dmulti14.length; i++) {
dmulti14[i] = 2*dmulti14[i];
}
double [] regression014 = PolynomialApproximation.getOrthoRegression(
dmulti[0], // final double [] data_x,
dmulti14, // final double [] data_y,
mask); // final boolean [] mask)
double [] regression_single = getYXRegression(
dmulti[0], // final double [] data_x,
dmulti[1]); // final double [] data_y,
System.out.println();
}
String [] map_names = new String[num_images + (((num_images==2))? 1 : 0)];
// float [][] extra_multi = new float [map_names.length][];
double [][] extra_multi = new double [map_names.length][];
for (int n = 0; n <num_images; n++) {
int mapn= (indices != null)? indices[n] : n;
map_names[n] = ortho_maps[mapn].getName()+"_"+ortho_maps[mapn].getLocalDateTime().toString().replace("T","_")+"_UTC";
map_names[n] += String.format(" raw=%7.1f", ortho_maps[mapn].getTemperature());
extra_multi[n] = multi[n];
extra_multi[n] = dmulti[n];
}
if (map_names.length > num_images) { // inefficient way, only for display
// correct(offset) pixel values relative to multi[0]
int map0= (indices != null)? indices[0] : 0;
if (mode == MODE_IMAGE) {
for (int n = 1; n < num_images; n++) {
//temp_mode, // 0 - do nothing, 1 - equalize average,2 - try to correct
int mapn= (indices != null)? indices[n] : n;
double offs = 0;
switch (temp_mode) {
case 1:
offs = ortho_maps[map0].getAveragePixel()-ortho_maps[mapn].getAveragePixel();
break;
case 2:
offs = ortho_maps[mapn].getTemperature()-ortho_maps[mapn].getAveragePixel()
-(ortho_maps[map0].getTemperature()-ortho_maps[map0].getAveragePixel());
break;
}
float foffs= (float) offs;
for (int i = 0; i < extra_multi[n].length; i++) if (!Float.isNaN(extra_multi[n][i])) {
extra_multi[n][i]+= foffs;
}
for (int n = 1; n < num_images; n++) {
//temp_mode, // 0 - do nothing, 1 - equalize average,2 - try to correct
int mapn= (indices != null)? indices[n] : n;
double offs = 0;
switch (temp_mode) {
case 1:
offs = ortho_maps[map0].getAveragePixel()-ortho_maps[mapn].getAveragePixel();
break;
case 2:
offs = ortho_maps[mapn].getTemperature()-ortho_maps[mapn].getAveragePixel()
-(ortho_maps[map0].getTemperature()-ortho_maps[map0].getAveragePixel());
break;
}
// float foffs= (float) offs;
for (int i = 0; i < extra_multi[n].length; i++) if (!Double.isNaN(extra_multi[n][i])) {
extra_multi[n][i]+= offs;
}
}
map_names[num_images] = "Diff 0-1";
extra_multi[num_images] = new float[extra_multi[0].length];
extra_multi[num_images] = new double[extra_multi[0].length];
for (int i = 0; i < extra_multi[num_images].length; i++) {
extra_multi[num_images][i] = extra_multi[1][i]-extra_multi[0][i];
}
}
ImageStack stack;
if (mode == MODE_MASK) {
title +="-MASK";
byte [][] byte_pix = new byte[multi.length][]; // multi[0].length];
for (int n = 0; n < multi.length; n++) { // slow
byte_pix[n] = new byte[multi[n].length];
for (int i = 0; i < multi[n].length; i++) {
byte_pix[n][i] = (byte) (Float.isNaN(multi[n][i]) ? 0 : 0xff);
}
}
stack = ShowDoubleFloatArrays.makeStack(
byte_pix,
wh[0],
wh[1],
map_names);
} else {
stack = ShowDoubleFloatArrays.makeStack(
extra_multi,
wh[0],
wh[1],
map_names,
false);
}
ImagePlus imp = new ImagePlus(title, stack);
if (show_centers) {
PointRoi roi = new PointRoi();
......@@ -657,9 +711,28 @@ public class OrthoMapsCollection implements Serializable{
}
imp.show(); // debugging
return imp;
}
public static double [] getYXRegression(
final double [] data_x,
final double [] data_y) {
double s0=0,sx=0,sx2=0,sy = 0, sxy=0;
for (int i = 0; i < data_x.length; i++) {
double x = data_x[i];
double y = 2*data_x[i];
if (!Double.isNaN(x) && !Double.isNaN(y)) {
s0+= 1.0;
sx+= x;
sx2+=x*x;
sy+= y;
sxy += x*y;
}
}
double d = (s0*sx2-sx*sx);
double a = (sxy*s0-sy*sx)/d;
double b = (sy*sx2-sx*sxy)/d;
return new double [] {a,b};
}
/**
* Rectify and render multiple images (as slices) matching vert_meters to
......@@ -671,27 +744,8 @@ public class OrthoMapsCollection implements Serializable{
* @param centers - null or double[maps.length][] - will return image centers coordinates
* @return
*/
public float [][] renderMulti (
int mode, // 0 - regular image, 1 - altitudes, 2 - black/white mask
int zoom_level,
int [] wh,
int [] origin, // maps[0] as a reference
double [][] centers){
return renderMulti (
// use_alt, // boolean use_alt,
mode, // int mode, // 0 - regular image, 1 - altitudes, 2 - black/white mask
null, // int [] indices,
false, // boolean bounds_to_indices,
null, // double [][][] affines,
null, // FineXYCorr warp,
zoom_level, // int zoom_level,
wh, // int [] wh,
origin, // int [] origin, // maps[0] as a reference
centers); // double [][] centers)
}
public float [][] renderMulti (
// boolean use_alt,
int mode, // 0 - regular image, 1 - altitudes, 2 - black/white mask
int [] indices, // null or which indices to use (normally just 2 for pairwise comparison)
boolean bounds_to_indices,
......@@ -703,9 +757,17 @@ public class OrthoMapsCollection implements Serializable{
double [][] centers){
final boolean use_alt = mode == MODE_ALT;
final int dbg_x=2783, dbg_y=-5228;
int [][] bounds = getBoundsPixels( // should be for rectified, {-bounds[0][0], -bounds[0][1]} - exact center
zoom_level,
bounds_to_indices? indices: null);
int [][] bounds;
if (affines == null) {
bounds = getBoundsPixels( // should be for rectified, {-bounds[0][0], -bounds[0][1]} - exact center
zoom_level,
bounds_to_indices? indices: null);
} else {
bounds = getBoundsPixels( // should be for rectified, {-bounds[0][0], -bounds[0][1]} - exact center
zoom_level,
bounds_to_indices? indices: null,
affines);
}
int width = bounds[0][1] - bounds[0][0]; // bounds[x][0] - negative
int height = bounds[1][1] - bounds[1][0];
if (wh != null) {
......@@ -724,27 +786,24 @@ public class OrthoMapsCollection implements Serializable{
}
final float [][] fpixels = new float [indices.length][width * height];
// for (int nmap:indices) { // = 0; nmap< ortho_maps.length; nmap++) {
for (int indx = 0; indx < indices.length; indx++) { //:indices) { // = 0; nmap< ortho_maps.length; nmap++) {
final int findx= indx;
final int nmap = indices[indx]; // nmap;
final double [][] affine = (affines !=null) ? affines[indx]: ortho_maps[nmap].affine; // only here use provided
Arrays.fill(fpixels[findx], Float.NaN);
final double scale = 1.0/OrthoMap.getPixelSizeMeters(zoom_level);
final double src_scale = 1.0/OrthoMap.getPixelSizeMeters(ortho_maps[nmap].orig_zoom_level); // pix per meter
// metric bounds of the rectified image relative to its origin
// double [][] mbounds = (affines !=null) ?
// ortho_maps[nmap].getBoundsMeters(true, affines[indx]): // use provided affine
// ortho_maps[nmap].getBoundsMeters(true);
double [][] mbounds = ortho_maps[nmap].getBoundsMeters(true); // keep original bounds
// double [][] mbounds = ortho_maps[nmap].getBoundsMeters(true); // keep original bounds
double [][] mbounds = ortho_maps[nmap].getBoundsMeters(true,affine); // keep original bounds
double [] enu_offset = ortho_maps[indices[0]].enuOffsetTo(ortho_maps[nmap]);
final double [] scaled_out_center = { // xy center to apply affine to
final double [] scaled_out_center = { // xy center to apply affine to in output pixels
-bounds[0][0] + scale * enu_offset[0],
-bounds[1][0] - scale * enu_offset[1]};
if (centers != null) {
centers[findx] = scaled_out_center;
}
final int [][] obounds = new int [2][2]; // output (rectified, combined) image bounds, relative to thje top-left
for (int n = 0; n< 2; n++) {
final int [][] obounds = new int [2][2]; // output (rectified, combined) image bounds, relative to the top-left
for (int n = 0; n< 2; n++) { // output pixels
obounds[n][0] = (int) Math.floor(scaled_out_center[n] + scale*mbounds[n][0]);
obounds[n][1] = (int) Math.ceil (scaled_out_center[n] + scale*mbounds[n][1]);
}
......@@ -752,44 +811,69 @@ public class OrthoMapsCollection implements Serializable{
final int ownd_width = obounds[0][1] - obounds[0][0];
final int ownd_height = obounds[1][1] - obounds[1][0];
final int ownd_len = ownd_width * ownd_height;
// double [][] src_bounds=(affines !=null) ?
// ortho_maps[nmap].getBoundsMeters (true, affines[indx]): // use provided affine
// ortho_maps[nmap].getBoundsMeters (true);
double [][] src_bounds=ortho_maps[nmap].getBoundsMeters (true); // using original affines
// double [][] src_bounds=ortho_maps[nmap].getBoundsMeters (true); // using original affines
double [][] src_bounds=ortho_maps[nmap].getBoundsMeters (false,affine); // using provided affines
final double [] src_center = {-src_bounds[0][0],-src_bounds[1][0]}; // x,y center offset in the source image
final double [][] affine = (affines !=null) ? affines[indx]: ortho_maps[nmap].affine; // only here use provided
final int src_width = use_alt? ortho_maps[nmap].getAltData().width: ortho_maps[nmap].getImageData().width;
final int src_height = use_alt? ortho_maps[nmap].getAltData().height : ortho_maps[nmap].getImageData().height;
final float [] src_img = use_alt? ortho_maps[nmap].getAltData().data : ortho_maps[nmap].getImageData().data;
if ((indx==0) && (warp != null)) { // set center from the first image
warp.setRender(
zoom_level, // int zoom_lev,
scaled_out_center[0], // double px0, // in render pixels
scaled_out_center[1]); // // double py0);
}
final float [] src_img = use_alt? ortho_maps[nmap].getAltData().data : ortho_maps[nmap].getImageData().data;
final Rectangle warp_woi =((indx==1) && (warp != null))? warp.getRenderWOI():null;
final Thread[] threads = ImageDtt.newThreadArray();
final AtomicInteger ai = new AtomicInteger(0);
/*
boolean tilt_alt = (ground_planes != null) && (ground_planes[indx] != null);
final float [] src_img = use_alt?
(tilt_alt? ortho_maps[nmap].getAltData().data.clone(): ortho_maps[nmap].getAltData().data) :
ortho_maps[nmap].getImageData().data;
if (tilt_alt) {
// apply tilt to the source altitudes
double [] vert_meters = ortho_maps[nmap].getVertMeters();
// altitude at top-left pixel;
final double top_left_alt =
ground_planes[indx][2] - vert_meters[0]*ground_planes[indx][0] - vert_meters[1]*ground_planes[indx][1];
final double [] tilt_XY = {ground_planes[indx][0]/src_scale, ground_planes[indx][1]/src_scale};
for (int ithread = 0; ithread < threads.length; ithread++) {
threads[ithread] = new Thread() {
public void run() {
for (int nPix = ai.getAndIncrement(); nPix < src_img.length; nPix = ai.getAndIncrement()) {
int py = nPix / src_width;
int px = nPix % src_width;
src_img[nPix] -= (top_left_alt + px * tilt_XY[0] + py * tilt_XY[1]);
}
}
};
}
ImageDtt.startAndJoin(threads);
ai.set(0);
}
*/
for (int ithread = 0; ithread < threads.length; ithread++) {
threads[ithread] = new Thread() {
public void run() {
for (int nPix = ai.getAndIncrement(); nPix < ownd_len; nPix = ai.getAndIncrement()) {
int opX = nPix % ownd_width + obounds[0][0]; // absolute output pX, pY
int opY = nPix / ownd_width + obounds[1][0];
double dX = (opX - scaled_out_center[0]) /scale; // in original image scale
double dX = (opX - scaled_out_center[0]) /scale; // in meters
double dY = (opY - scaled_out_center[1]) /scale;
double [] xy_src = { // pixels of the source image
src_scale * (affine[0][0]*dX + affine[0][1]*dY + affine[0][2] + src_center[0]),
src_scale * (affine[1][0]*dX + affine[1][1]*dY + affine[1][2] + src_center[1])};
// limit to the source image
if ((((int) opX)==dbg_x) && (((int) opY)==dbg_y)) {
System.out.println("opX="+opX+", opy="+opY);
}
if ((warp_woi != null) && (warp_woi.contains(opX,opY))) {
if ((opX==dbg_x) && (opY==dbg_y)) {
System.out.println("opX="+opX+", opy="+opY);
}
double [] dxy = warp.getWarp(opX,opY);
xy_src[0] += dxy[0];
xy_src[1] += dxy[1];
}
if ((xy_src[0] >= 0) && (xy_src[0] < (src_width-1)) &&
(xy_src[1] >= 0) && (xy_src[1] < (src_height-1))) {
int [] ixy_src = {(int) Math.floor(xy_src[0]), (int)Math.floor(xy_src[1]) };
......@@ -821,14 +905,16 @@ public class OrthoMapsCollection implements Serializable{
double [][] ground_planes, // null - images, non-null altitudes. use new double[2][] for old way alt
int [] indices, // null or which indices to use (normally just 2 for pairwise comparison)
boolean bounds_to_indices,
double [][][] affines, // null or [indices.length][2][3]
double [][][] affines, // null or [indices.length][2][3]
double [][] equalize_in,
boolean ignore_equalize,
FineXYCorr warp,
int zoom_level,
int [] wh,
int [] origin, // maps[0] as a reference
double [][] centers){
boolean use_alt = ground_planes != null;
final int dbg_x=707, dbg_y=615;
final int dbg_x=707, dbg_y=-615;
int [][] bounds;
if (affines == null) {
bounds = getBoundsPixels( // should be for rectified, {-bounds[0][0], -bounds[0][1]} - exact center
......@@ -862,6 +948,7 @@ public class OrthoMapsCollection implements Serializable{
final int findx= indx;
final int nmap = indices[indx]; // nmap;
final double [][] affine = (affines !=null) ? affines[indx]: ortho_maps[nmap].affine; // only here use provided
final double [] equalize = ignore_equalize ? null:((equalize_in != null) ? equalize_in[indx] : ortho_maps[nmap].equalize);
Arrays.fill(dpixels[findx], Float.NaN);
final double scale = 1.0/OrthoMap.getPixelSizeMeters(zoom_level);
final double src_scale = 1.0/OrthoMap.getPixelSizeMeters(ortho_maps[nmap].orig_zoom_level); // pix per meter
......@@ -960,7 +1047,10 @@ public class OrthoMapsCollection implements Serializable{
d01* kxy[0] *(1.0 - kxy[1])+
d10*(1.0 - kxy[0])* kxy[1]+
d11* kxy[0] * kxy[1];
dpixels[findx][opX + opY*width] = (float) d;
if (equalize != null) {
d = equalize[0] * d + equalize[1];
}
dpixels[findx][opX + opY*width] = d;
}
}
}
......@@ -2351,10 +2441,12 @@ public class OrthoMapsCollection implements Serializable{
int [] origin = new int[2];
double [][] centers = new double [indices.length][];
double [][] dmulti = renderMultiDouble (
null, // double [][] ground_planes, // null - images, non-null altitudes. use new double[2][3] for old way alt
null, // double [][] ground_planes, // null - images, non-null altitudes. use new double[2][3] for old way alt
indices, // int [] indices, // null or which indices to use (normally just 2 for pairwise comparison)
true, // boolean bounds_to_indices,
true, // boolean bounds_to_indices,
affines, // double [][][] affines, // null or [indices.length][2][3]
null, // double [][] equalize,
false, // boolean ignore_equalize,
warp, // FineXYCorr warp,,
zoom_level, // int zoom_level,
wh, // int [] wh,
......@@ -2367,12 +2459,13 @@ public class OrthoMapsCollection implements Serializable{
indices, // int [] indices, // null or which indices to use (normally just 2 for pairwise comparison)
true, // boolean bounds_to_indices,
affines, // double [][][] affines, // null or [indices.length][2][3]
null, // double [][] equalize,
true, // boolean ignore_equalize,
warp, // FineXYCorr warp,,
zoom_level, // int zoom_level,
wh, // int [] wh,
origin, // int [] origin){ // maps[0] as a reference
null); // centers); // double [][] centers) // already set with dmulti
}
......@@ -4071,15 +4164,17 @@ public class OrthoMapsCollection implements Serializable{
int [] origin = new int[2];
double [][] centers =new double [indices.length][];
double [][] dmulti = renderMultiDouble (
null, // double [][] ground_planes, // null - images, non-null altitudes. use new double[2][3] for old way alt
indices, // int [] indices, // null or which indices to use (normally just 2 for pairwise comparison)
null, // double [][] ground_planes, // null - images, non-null altitudes. use new double[2][3] for old way alt
indices, // int [] indices, // null or which indices to use (normally just 2 for pairwise comparison)
bounds_to_indices, // boolean bounds_to_indices,
null, // affines, // double [][][] affines, // null or [indices.length][2][3]
null, // warp, // FineXYCorr warp,,
zoom_lev, // int zoom_level,
wh, // int [] wh,
origin, // int [] origin){ // maps[0] as a reference
centers); // double [][] centers)
null, // affines, // double [][][] affines, // null or [indices.length][2][3]
null, // double [][] equalize,
true, // boolean ignore_equalize,
null, // warp, // FineXYCorr warp,,
zoom_lev, // int zoom_level,
wh, // int [] wh,
origin, // int [] origin){ // maps[0] as a reference
centers); // double [][] centers)
double [] overlaps = getFracOverlaps(
dmulti); // final double [][] dmulti) {
......@@ -4418,7 +4513,7 @@ public class OrthoMapsCollection implements Serializable{
if (!direct && (inv_match != null)) { // prefer inverse
use_inverse = true;
} else if (direct_match != null) {
if (skip_existing) {
if (skip_existing && (direct_match.overlap > 0)) { // do not skip old pairs - refine them
skip = true;
} else if (refine_existing) {
use_direct = true;
......@@ -4751,15 +4846,17 @@ public class OrthoMapsCollection implements Serializable{
int [] origin = new int[2];
double [][] centers = show_centers? (new double [indices.length][]): null;
double [][] dmulti = renderMultiDouble (
null, // double [][] ground_planes, // null - images, non-null altitudes. use new double[2][3] for old way alt
indices, // int [] indices, // null or which indices to use (normally just 2 for pairwise comparison)
null, // double [][] ground_planes, // null - images, non-null altitudes. use new double[2][3] for old way alt
indices, // int [] indices, // null or which indices to use (normally just 2 for pairwise comparison)
bounds_to_indices, // boolean bounds_to_indices,
null, // affines, // double [][][] affines, // null or [indices.length][2][3]
null, // warp, // FineXYCorr warp,,
zoom_lev, // int zoom_level,
wh, // int [] wh,
origin, // int [] origin){ // maps[0] as a reference
centers); // double [][] centers)
null, // affines, // double [][][] affines, // null or [indices.length][2][3]
null, // double [][] equalize,
true, // boolean ignore_equalize,
null, // warp, // FineXYCorr warp,,
zoom_lev, // int zoom_level,
wh, // int [] wh,
origin, // int [] origin){ // maps[0] as a reference
centers); // double [][] centers)
if (margins > 0) {
addMargins(
dmulti, // double [][] dmulti,
......@@ -4788,12 +4885,14 @@ public class OrthoMapsCollection implements Serializable{
new double [indices.length][] , // double [][] ground_planes, // null - images, non-null altitudes. use new double[2][3] for old way alt
indices, // int [] indices, // null or which indices to use (normally just 2 for pairwise comparison)
bounds_to_indices, // boolean bounds_to_indices,
null, // affines, // double [][][] affines, // null or [indices.length][2][3]
null, // warp, // FineXYCorr warp,,
zoom_lev, // int zoom_level,
wh, // int [] wh,
origin, // int [] origin){ // maps[0] as a reference
centers); // double [][] centers)
null, // affines, // double [][][] affines, // null or [indices.length][2][3]
null, // double [][] equalize,
true, // boolean ignore_equalize,
null, // warp, // FineXYCorr warp,,
zoom_lev, // int zoom_level,
wh, // int [] wh,
origin, // int [] origin){ // maps[0] as a reference
centers); // double [][] centers)
if (margins > 0) {
addMargins(
dmulti, // double [][] dmulti,
......@@ -4818,15 +4917,17 @@ public class OrthoMapsCollection implements Serializable{
int [] origin = new int[2];
double [][] centers = show_centers? (new double [indices.length][]): null;
double [][] dmulti = renderMultiDouble (
null, // double [][] ground_planes, // null - images, non-null altitudes. use new double[2][3] for old way alt
indices, // int [] indices, // null or which indices to use (normally just 2 for pairwise comparison)
null, // double [][] ground_planes, // null - images, non-null altitudes. use new double[2][3] for old way alt
indices, // int [] indices, // null or which indices to use (normally just 2 for pairwise comparison)
bounds_to_indices, // boolean bounds_to_indices,
null, // affines, // double [][][] affines, // null or [indices.length][2][3]
null, // warp, // FineXYCorr warp,,
zoom_lev, // int zoom_level,
wh, // int [] wh,
origin, // int [] origin){ // maps[0] as a reference
centers); // double [][] centers)
null, // affines, // double [][][] affines, // null or [indices.length][2][3]
null, // double [][] equalize,
true, // boolean ignore_equalize,
null, // warp, // FineXYCorr warp,,
zoom_lev, // int zoom_level,
wh, // int [] wh,
origin, // int [] origin){ // maps[0] as a reference
centers); // double [][] centers)
if (margins > 0) {
addMargins(
dmulti, // double [][] dmulti,
......@@ -4857,15 +4958,17 @@ public class OrthoMapsCollection implements Serializable{
int [] origin = new int[2];
double [][] centers = show_centers? (new double [indices.length][]): null;
double [][] dmulti = renderMultiDouble (
null, // double [][] ground_planes, // null - images, non-null altitudes. use new double[2][3] for old way alt
indices, // int [] indices, // null or which indices to use (normally just 2 for pairwise comparison)
null, // double [][] ground_planes, // null - images, non-null altitudes. use new double[2][3] for old way alt
indices, // int [] indices, // null or which indices to use (normally just 2 for pairwise comparison)
bounds_to_indices, // boolean bounds_to_indices,
null, // affines, // double [][][] affines, // null or [indices.length][2][3]
null, // warp, // FineXYCorr warp,,
zoom_lev, // int zoom_level,
wh, // int [] wh,
origin, // int [] origin){ // maps[0] as a reference
centers); // double [][] centers)
null, // affines, // double [][][] affines, // null or [indices.length][2][3]
null, // double [][] equalize,
true, // boolean ignore_equalize,
null, // warp, // FineXYCorr warp,,
zoom_lev, // int zoom_level,
wh, // int [] wh,
origin, // int [] origin){ // maps[0] as a reference
centers); // double [][] centers)
if (margins > 0) {
addMargins(
dmulti, // double [][] dmulti,
......
src/main/java/com/elphel/imagej/orthomosaic/OrthoMultiLMA.java
View file @ 1671728b
......@@ -40,7 +40,8 @@ import ij.ImagePlus;
import ij.gui.PointRoi;
public class OrthoMultiLMA {
final boolean move_only;
final boolean move_only;
final boolean corr_avg; // correct average skew, tilt, scale
private double [] last_rms = null; // {rms, rms_pure}, matching this.vector
private double [] good_or_bad_rms = null; // just for diagnostics, to read last (failed) rms
private double [] initial_rms = null; // {rms, rms_pure}, first-calcualted rms
......@@ -61,7 +62,11 @@ public class OrthoMultiLMA {
private double [][] offsets = null; // scene offsets (rd)
private int num_scenes = 0;
private int num_pairs = 0;
public OrthoMultiLMA (boolean move_only) {
public OrthoMultiLMA (
boolean corr_avg,
boolean move_only
) {
this.corr_avg = corr_avg && !move_only;
this.move_only = move_only;
}
public static boolean testMultiLMA(
......@@ -122,6 +127,8 @@ public class OrthoMultiLMA {
indices, // int [] indices, // null or which indices to use (normally just 2 for pairwise comparison)
true, // boolean bounds_to_indices,
affines_a, // affines_0d, // affines, // double [][][] affines, // null or [indices.length][2][3]
null, // double [][] equalize,
true, // boolean ignore_equalize,
null, // warp, // FineXYCorr warp,,
zoom_level, // int zoom_level,
wh, // int [] wh,
......@@ -147,6 +154,8 @@ public class OrthoMultiLMA {
indices, // int [] indices, // null or which indices to use (normally just 2 for pairwise comparison)
true, // boolean bounds_to_indices,
affines_sym,// affines, // double [][][] affines, // null or [indices.length][2][3]
null, // double [][] equalize,
true, // boolean ignore_equalize,
null, // warp, // FineXYCorr warp,,
zoom_level, // int zoom_level,
wh, // int [] wh,
......@@ -213,19 +222,26 @@ public class OrthoMultiLMA {
int debugLevel = 1;
boolean move_only = false;
double [] val_coord = null; // 1 - valid, 0 - invalid, minimize coordinates errors
double position_pull = 0.001;
boolean use_inv = false;
double overlap_pow = 2.0; // match weight as overlap fraction to this power
double skew_pull = 1.0;
double tilt_pull = 1.0;
double scale_pull = 0.1; // .0;
double position_pull = 0.0001;
boolean corr_avg= (skew_pull > 0) || (tilt_pull > 0) || (scale_pull > 0);
int [] indices = maps_collection.getScenesSelection(
null, // boolean select_all,
" to build a map"); // String purpose)
OrthoMultiLMA oml = new OrthoMultiLMA(move_only);
OrthoMultiLMA oml = new OrthoMultiLMA(
corr_avg,
move_only);
double lambda = 0.1;
double lambda_scale_good = 0.5;
double lambda_scale_bad = 8.0;
double lambda_max = 100;
double rms_diff = 0.000001;
int num_iter = 20;
int num_iter = 100; // 50;
boolean last_run = false;
oml.prepareLMA (
clt_parameters, // CLTParameters clt_parameters,
......@@ -233,6 +249,9 @@ public class OrthoMultiLMA {
indices, // int [] indices,
val_coord, // double [] val_coord, // 1 - valid, 0 - invalid, minimize coordinates errors
position_pull, // double position_pull,
skew_pull, // double skew_pull,
tilt_pull, // double tilt_pull,
scale_pull, // double scale_pull,
use_inv, // boolean use_inv,
overlap_pow, // double overlap_pow, // match weight as overlap fraction to this power
debugLevel); // int debugLevel)
......@@ -256,7 +275,7 @@ public class OrthoMultiLMA {
//Get and apply affines
//oml.updateAffines(maps_collection);
double [][][] affines = oml.getAffines();
double [] fx = oml.getFx();
// test
/*
PairwiseOrthoMatch match = maps_collection.ortho_maps[indices[0]].getMatch(maps_collection.ortho_maps[indices[1]].getName());
......@@ -284,6 +303,8 @@ public class OrthoMultiLMA {
indices, // int [] indices, // null or which indices to use (normally just 2 for pairwise comparison)
true, // boolean bounds_to_indices,
affines, // null, // affines, // double [][][] affines, // null or [indices.length][2][3]
null, // double [][] equalize,
true, // boolean ignore_equalize,
null, // warp, // FineXYCorr warp,,
zoom_level, // int zoom_level,
wh, // int [] wh,
......@@ -307,6 +328,7 @@ public class OrthoMultiLMA {
roi.setOptions("label");
imp_multi.setRoi(roi);
imp_multi.show();
/*
double [][] affine = getAffineAndDerivatives(
move_only, //boolean move_only,
affines[0], // double [][] affine00,
......@@ -317,7 +339,7 @@ public class OrthoMultiLMA {
affines[0],
affines[1],
oml.offsets[0]);
*/
/*
......@@ -336,6 +358,16 @@ public class OrthoMultiLMA {
return 0;
}
public double [] getParameters() {
return parameters_vector;
}
public double [] getFx() {
return getFxDerivs(
parameters_vector, // double [] vector,
null, // final double [][] jt, // should be null or initialized with [vector.length][]
0); // final int debug_level)
}
public int prepareLMA (
CLTParameters clt_parameters,
......@@ -343,6 +375,9 @@ public class OrthoMultiLMA {
int [] indices,
double [] val_coord, // 1 - valid, 0 - invalid, minimize coordinates errors
double position_pull,
double skew_pull,
double tilt_pull,
double scale_pull,
boolean use_inv,
double overlap_pow, // match weight as overlap fraction to this power
int debugLevel) {
......@@ -383,7 +418,6 @@ public class OrthoMultiLMA {
for (int np = 0; np < num_pairs; np++) {
Point pair = pairs_list.get(np);
pairs[np] = new int[] {pair.x, pair.y};
// double [] enuOffset = maps_collection.ortho_maps[indices[pair.x]].enuOffsetTo(maps_collection.ortho_maps[indices[pair.y]]);
double [] enuOffset = maps_collection.ortho_maps[indices[pair.y]].enuOffsetTo(maps_collection.ortho_maps[indices[pair.x]]);
offsets[np] = new double[] {enuOffset[0], -enuOffset[1]}; // {right,down} of the image
}
......@@ -396,9 +430,8 @@ public class OrthoMultiLMA {
int n62 = move_only ? 2 : 6;
int n13 = move_only ? 1 : 3;
y_vector = new double [n62*num_pairs + 2 * num_scenes]; // last 2 * num_scenes will stay 0
weights = new double [n62*num_pairs + 2 * num_scenes];
y_vector = new double [n62*num_pairs + 2 * num_scenes + (corr_avg ? 3:0)]; // last 2 * num_scenes + 3 will stay 0
weights = new double [n62*num_pairs + 2 * num_scenes + (corr_avg ? 3:0)];
parameters_vector = new double [n62*num_scenes]; // maybe will need move-only mode?
for (int n = 0; n < num_scenes; n++) {
double [][] affine = maps_collection.ortho_maps[indices[n]].getAffine();
......@@ -441,7 +474,12 @@ public class OrthoMultiLMA {
weights[wi++] = w;
weights[wi++] = w;
sw+= 2*w;
}
}
if (corr_avg) {
weights[wi++] = skew_pull;
weights[wi++] = tilt_pull;
weights[wi++] = scale_pull;
}
pure_weight = swp/sw;
double s = 1.0/sw;
for (int i = 0; i <weights.length;i++) {
......@@ -1166,6 +1204,49 @@ public class OrthoMultiLMA {
};
}
ImageDtt.startAndJoin(threads);
ai.set(0);
if (corr_avg) { // && ((skew_pull > 0) || (tilt_pull > 0) || (scale_pull > 0))) {
final int pull_indices = weights.length-3;
for (int nScene = 0; nScene < num_scenes; nScene++){
int indx = 6 * nScene;
double a00 = vector[indx+0],a01=vector[indx+1],a10=vector[indx+3],a11=vector[indx+4];
double aa1 = 0.5*(a00*a00 + a10*a10);
double aa2 = 0.5*(a01*a01 + a11*a11);
fx[pull_indices + 0] += a00*a01+a10*a11;
fx[pull_indices + 1] += aa1-aa2;
fx[pull_indices + 2] += aa1+aa2 - 1.0;
}
if (jt!= null) {
for (int ithread = 0; ithread < threads.length; ithread++) {
threads[ithread] = new Thread() {
public void run() {
for (int nScene = ai.getAndIncrement(); nScene < num_scenes; nScene = ai.getAndIncrement()){
int indx = 6 * nScene;
double a00=vector[indx+0],a01=vector[indx+1],a10=vector[indx+3],a11=vector[indx+4];
jt[indx + 0][pull_indices + 0] = a01;
jt[indx + 1][pull_indices + 0] = a00;
jt[indx + 3][pull_indices + 0] = a11;
jt[indx + 4][pull_indices + 0] = a10;
jt[indx + 0][pull_indices + 1] = a00;
jt[indx + 1][pull_indices + 1] =-a01;
jt[indx + 3][pull_indices + 1] = a10;
jt[indx + 4][pull_indices + 1] =-a11;
jt[indx + 0][pull_indices + 2] = a00;
jt[indx + 1][pull_indices + 2] = a01;
jt[indx + 3][pull_indices + 2] = a10;
jt[indx + 4][pull_indices + 2] = a11;
}
}
};
}
ImageDtt.startAndJoin(threads);
ai.set(0);
}
}
return fx;
}
......
src/main/java/com/elphel/imagej/orthomosaic/PairwiseOrthoMatch.java
View file @ 1671728b
......@@ -11,10 +11,12 @@ public class PairwiseOrthoMatch implements Serializable {
private static final long serialVersionUID = 1L;
public double [][] affine = new double[2][3];
public transient double [][] jtj = new double [6][6];
public int zoom_lev;
public double rms = Double.NaN;
public transient int [] nxy = null; // not saved, just to communicate for logging
public transient double overlap = 0.0;
public int zoom_lev;
public double rms = Double.NaN;
public transient int [] nxy = null; // not saved, just to communicate for logging
public transient double overlap = 0.0;
public transient double [] equalize1to0 = {1,0}; // value1 = equalize2to1[0]*value2+equalize2to1[1]
public PairwiseOrthoMatch() {
}
......@@ -30,6 +32,11 @@ public class PairwiseOrthoMatch implements Serializable {
this.rms = rms;
this.overlap = overlap;
}
public void setEqualize2to1 (double [] equalize2to1) {
this.equalize1to0 = equalize2to1;
}
public PairwiseOrthoMatch clone() {
double [][] affine = {this.affine[0].clone(),this.affine[1].clone()};
double [][] jtj = new double [this.jtj.length][];
......@@ -45,6 +52,7 @@ public class PairwiseOrthoMatch implements Serializable {
if (nxy != null) {
pom.nxy = nxy.clone();
}
pom.equalize1to0 = this.equalize1to0.clone();
return pom;
}
/**
......@@ -102,6 +110,7 @@ public class PairwiseOrthoMatch implements Serializable {
rd[0] * affine[1][0]+ rd[1]*(affine[1][1]-1.0)};
affine[0][2] += corr[0];
affine[1][2] += corr[1];
inverted_match.setEqualize2to1(new double [] {1/equalize1to0[0], -equalize1to0[1]/equalize1to0[0]});
return inverted_match;
}
......@@ -168,10 +177,16 @@ public class PairwiseOrthoMatch implements Serializable {
affine = new double[][] {
{A.get(0,0),A.get(0,1), B.get(0,0)},
{A.get(1,0),A.get(1,1), B.get(1,0)}};
// jtj = null;
// rms = Double.NaN; // double rms,
// zoom_lev = 0; // int zoom_lev)
}
public void combineEqualize(
double [] equalize0,
double [] equalize1 ) {
setEqualize2to1(new double[] {
equalize1[0]/equalize0[0],
equalize1[1]-equalize1[0]/equalize0[0]*equalize0[1]});
}
public static double [][] combineAffines(
double [][] affine0,
double [][] affine, // differential
......@@ -203,6 +218,8 @@ public class PairwiseOrthoMatch implements Serializable {
}
}
oos.writeObject(overlap);
oos.writeObject(equalize1to0);
}
private void readObject(ObjectInputStream ois) throws ClassNotFoundException, IOException {
ois.defaultReadObject();
......@@ -216,6 +233,8 @@ public class PairwiseOrthoMatch implements Serializable {
}
}
overlap = (Double) ois.readObject();
// equalize1to0 = new double[] {1,0};
equalize1to0 = (double[]) ois.readObject();
}
//private void readObjectNoData() throws ObjectStreamException; // used to modify default values
}
src/main/java/com/elphel/imagej/tileprocessor/TileNeibs.java
View file @ 1671728b
......@@ -1356,4 +1356,5 @@ public class TileNeibs{
}
return filled;
}
}
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment