1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482
|
unit unit_contour;// Moore Neighbor Contour Tracing Algorithm
{Copyright (C) 2023 by Han Kleijn, www.hnsky.org
email: han.k.. at...hnsky.org
This Source Code Form is subject to the terms of the Mozilla Public
License, v. 2.0. If a copy of the MPL was not distributed with this
file, You can obtain one at https://mozilla.org/MPL/2.0/. }
interface
uses
Classes, SysUtils,graphics,forms,math,controls,lclintf,fpcanvas,
astap_main;
procedure contour( plot : boolean;img : image_array; var head: theader; blur, sigmafactor : double);//find contour and satellite lines in an image
function line_distance(fitsX,fitsY,slope,intercept: double) : double;
procedure trendline_without_outliers(xylist: star_list; len{length xylist} : integer; out slope, intercept,sd: double);//find linear trendline Y = magnitude_slope*X + intercept. Remove outliers in step 2
//procedure add_to_storage;//add streaks to storage
//procedure clear_storage;//clear streak storage
type
streak =record
slope : double;
intercept : double;
end;
var
streak_lines : array of streak; // storage for streaks of one image
nr_streak_lines : integer;
implementation
uses unit_stack,unit_gaussian_blur,unit_astrometric_solving;
procedure draw_streak_line(slope,intercept: double);//draw line y = slope * x + intercept
var
x,y, x1,y1,x2,y2: double;
w,h : integer;
flipV,fliph : boolean;
begin
with mainwindow do
begin
Flipv:=mainwindow.flip_vertical1.Checked;
Fliph:=mainwindow.Flip_horizontal1.Checked;
w:=image1.Canvas.Width-1;
h:=image1.Canvas.height-1;
end;
//start point line
x1:=0;
y1:=intercept;
if y1>h then
begin
y1:=h;
x1:=(h-intercept)/slope;
end
else
if y1<0 then
begin
y1:=0;
x1:=(-intercept)/slope;
end;
//end point line
x2:=w-1;
y2:=slope*(w-1)+intercept;
if y2>h then
begin
y2:=h;
x2:=(h-intercept)/slope;
end
else
if y2<0 then
begin
y2:=0;
x2:=(-intercept)/slope;
end;
//draw
if Fliph then
begin
x1:=w-x1;
x2:=w-x2;
end;
if Flipv=false then
begin
y1:=h-y1;
y2:=h-y2;
end;
mainwindow.image1.Canvas.MoveTo(round(x1),round(y1));
mainwindow.image1.Canvas.lineTo(round(x2),round(y2));
end;
function line_distance(fitsX,fitsY,slope,intercept: double) : double;
begin
//y:=ax+c => 0=by+ax+c
//0:=-y+ax+c and b=-1
//distance:=abs(a.fitsX+b.fitsY+c)/sqrt(sqr(a)+sqr(b)) See https://en.wikipedia.org/wiki/Distance_from_a_point_to_a_line
result:=abs(slope*fitsX -fitsY + intercept)/sqrt(sqr(slope)+1);
end;
procedure trendline(xylist: star_list; len{length xylist} : integer; out slope, intercept:double); //find linear trendline Y = magnitude_slope*X + intercept
var //idea from https://stackoverflow.com/questions/43224/how-do-i-calculate-a-trendline-for-a-graph
sumX,sumX2,sumY, sumXY,median,mad : double;
count, i : integer;
median_array : array of double;
begin
count:=0;
sumX:=0;
sumX2:=0;
sumY:=0;
sumXY:=0;
for i:=0 to len-1 do
begin
inc(count);
//memo2_message(#9+floattostr(xylist[0,i])+#9+floattostr(xylist[1,i]));
sumX:=sumX+xylist[0,i]; //sum X= sum B_V values = sum star colours;
sumX2:=sumx2+sqr(xylist[0,i]);
sumY:=sumY+xylist[1,i]; //sum Y, sum delta magnitudes;
sumXY:=sumXY+xylist[0,i]*xylist[1,i];
end;
Slope:=(count*sumXY - sumX*sumY) / (count*sumX2 - sqr(sumX)); // b = (n*Σ(xy) - ΣxΣy) / (n*Σ(x^2) - (Σx)^2)
Intercept:= (sumY - Slope * sumX)/count; // a = (Σy - bΣx)/n
end;
procedure trendline_without_outliers(xylist: star_list; len{length xylist} : integer; out slope, intercept,sd: double);//find linear trendline Y = magnitude_slope*X + intercept. Remove outliers in step 2
var
e : double;
xylist2 : star_list;
counter,i : integer;
begin
trendline(xylist, len{length xylist}, {out} slope, intercept);
// find standard deviation
sd:=0;
for i:=0 to len-1 do
sd:=sd + sqr(slope*xylist[0,i] - xylist[1,i] + intercept)/(sqr(slope)+1);// sum the sqr line distance. Note the line distance is abs(slope*fitsX -fitsY + intercept)/sqrt(sqr(slope)+1), See https://en.wikipedia.org/wiki/Distance_from_a_point_to_a_line
sd:=sqrt(sd/len); //sd
//calculate the trendline but ignore outliers in Y (b-v)
setlength(xylist2,2,len);
counter:=0;
for i:=0 to len-1 do
begin
e:=abs(xylist[1,i]{y original} - (slope * xylist[0,i]+intercept{y mean})); //calculate absolute error
if e<1.5 *sd then //not an outlier keep 86.64%
begin
xylist2[0,counter]:=xylist[0,i];// xy list without outliers
xylist2[1,counter]:=xylist[1,i];
inc(counter)
end;
end;
trendline(xylist2, counter{length xylist2}, {out} slope, intercept);
xylist2:=nil;
end;
procedure contour( plot : boolean;img : image_array; var head: theader; blur, sigmafactor : double);//find contour and satellite lines in an image
var
fitsX,fitsY,ww,hh,fontsize,minX,minY,maxX,maxY,x,y,detection_grid,binning : integer;
detection_level,surface,{leng,}maxleng,slope, intercept,sd : double;
restore_his, Fliph, Flipv : boolean;
img_sa,img_bk : image_array;
contour_array : array of array of integer;
contour_array2 : star_list;
bg,sd_bg : double;
procedure mark_pixel(x,y : integer);{flip if required for plotting. From array to image1 coordinates}
begin
// show_marker_shape(mainwindow.shape_var1,1,10,10,10{minimum},X,Y);
if Fliph then x:=ww-1-x;
if Flipv=false then y:=hh-1-y;
mainwindow.image1.Canvas.pixels[x*binning,y*binning]:=clYellow;
// application.processmessages;
end;
procedure mark_pixel_blue(x,y : integer);{flip if required for plotting. From array to image1 coordinates}
begin
// show_marker_shape(mainwindow.shape_var1,1,10,10,10{minimum},X,Y);
if Fliph then x:=ww-1-x;
if Flipv=false then y:=hh-1-y;
mainwindow.image1.Canvas.pixels[x*binning,y*binning]:=clBlue;
// application.processmessages;
end;
procedure writetext(x,y : integer; tex :string);
begin
if Fliph then x:=ww-1-x;
if Flipv=false then y:=hh-1-y;
mainwindow.image1.Canvas.textout(min(ww*binning-600,x*binning),y*binning,tex);{}
end;
// procedure local_background(x1,y1:integer; out bg,sd: double);
// var
// i,counter,startX,stopX,startY,stopY : integer;
// mad_bg : double;
// background : array [0..100] of double;
// begin
// startX:=max(0,x1-14);
// startY:=max(0,y1-14);
// stopX:=min(w,x1+14);
// stopY:=min(h,y1+14);
// counter:=0;
// for i:=startX to stopX do {calculate the mean outside the the detection area}
// begin
// background[counter]:=img_bk[0,i,startY];
// inc(counter);
// end;
// for i:=startX to stopX do {calculate the mean outside the the detection area}
// begin
// background[counter]:=img_bk[0,i,stopY];
// inc(counter);
// end;
// for i:=startY-1 to stopY-1 do {calculate the mean outside the the detection area}
// begin
// background[counter]:=img_bk[0,startX,i];
// inc(counter);
// end;
//}
// bg:=Smedian(background,counter);
// for i:=0 to counter-1 do background[i]:=abs(background[i] - bg);{fill background with offsets}
// mad_bg:=Smedian(background,counter); //median absolute deviation (MAD)
// sd:=mad_bg*1.4826; {Conversion from mad to sd for a normal distribution. See https://en.wikipedia.org/wiki/Median_absolute_deviation}
// {star_bg, sd_bg and r_aperture are global variables}
// end;
procedure find_contour(fx,fy : integer);// Moore Neighbor Contour Tracing Algorithm
function img_protected(xx,yy :integer) : boolean;//return true if pixel is above detection level but avoids errors by reading outside the image.
begin
if ((xx>=0) and (xx<ww-1) and (yy>=0) and (yy<hh-1)) then
result:=img_bk[0,yy,xx]>detection_level
else
result:=false;
end;
var detection : boolean;
direction, counter,counterC,startX,startY,i,j,k,offset : integer;
const
newdirection : array[0..7] of integer=(-1,0,0,+1,+1,+2,+2,-1);//delta directions
directions : array[0..7,0..1] of integer=((-1,-1), //3 south east, direction
(-1,0), //0 east
(-1,+1), //0 north east
(0,+1), //1, north
(+1,+1), //1 north west
(+1,0), //2 west
(+1,-1), //2 south west
(0,-1)); //3 south
begin
direction:=1;// , north=0, west=1, south=2. east=3
startX:=fx;
startY:=fy;
counter:=0;
counterC:=0;
setlength(contour_array,2,4*ww);
repeat
detection:=false;
for i:=0 to 7 do
begin
j:=((i+direction*2) and $7);
if img_protected(fx+directions[j,0],fy+directions[j,1])then //pixel detected
begin
fx:=fx+directions[j,0];
fy:=fy+directions[j,1];
detection:=true;
direction:=direction+newdirection[i]; //new direction
break;
end;
end;
if detection=false then
break
else
begin
if plot then mark_pixel(fx,fy);
contour_array[0,counterC]:=fx;
contour_array[1,counterC]:=fy;
inc(counterC);
end;
img_sa[0,fy,fx]:=img_sa[0,fy,fx]+1;//mark as inspected/used
if img_sa[0,fy,fx]>1 then break;//is looping local
inc(counter);
until (((fx=startX) and (fy=startY)) or (counter>4*ww));
//mark inner of contour
surface:=0;
maxX:=0;
minX:=999999;
maxY:=0;
minY:=999999;
for i:=0 to counterC-1 do
begin
minX:=min(contour_array[0,i],minX);
maxX:=max(contour_array[0,i],maxX);
minY:=min(contour_array[1,i],minY);
maxY:=max(contour_array[1,i],maxY);
for j:=0 to counterC-1 do
begin //mark inner of contour
if contour_array[1,i]=contour_array[1,j] then //y position the same
begin
for k:=min(contour_array[0,i],contour_array[0,j]) to max(contour_array[0,i],contour_array[0,j]) do //mark space between the mininum and maximum x values. With two pixel extra overlap.
begin
if img_sa[0,contour_array[1,i],k]<0 then
begin
surface:=surface+1;
img_sa[0,contour_array[1,i],k]:=+1;//mark as inspected/used
end;
// mark_pixel_blue(k,contour_array[1,i]);
//application.processmessages;
end;
end;
end;
end;
if surface>200*2 then
begin
maxleng:=sqrt(sqr(maxY-minY)+sqr(maxX-minX));
//writetext(contour_array[0,i],contour_array[1,i],floattostr(surface)+ ', '+floattostr(maxleng)+ ', '+floattostr(sqr(maxleng)/surface));
if ((maxleng>200) and (sqr(maxleng)/surface>10)) then
begin
setlength(contour_array2,2,counterC);
for i:=0 to counterC-1 do //convert to an array of singles instead of integers
begin
contour_array2[0,i]:=contour_array[0,i];
contour_array2[1,i]:=contour_array[1,i];
//memo2_message(#9+floattostr(contour_array[0,i])+#9+floattostr(contour_array[1,i]));
end;
trendline_without_outliers(contour_array2,counterC,slope, intercept,sd);
intercept:=intercept*binning;
sd:=sd*binning;
if sd<10 then
begin // A real line, sd max is about line thickness plus a nearby star.
if plot then
begin
mainwindow.image1.Canvas.Pen.mode:=pmXor;
mainwindow.image1.Canvas.Pen.Color := clred;
draw_streak_line(slope,intercept);//draw satellite streak
mainwindow.image1.Canvas.pen.color:=clyellow;
end;
if plot then writetext(min(ww*binning,contour_array[0,counterC div 2]),contour_array[1,counterC div 2],' Y='+floattostrf(slope,FFgeneral,5,0)+'*X + '+Floattostrf(intercept,FFgeneral,5,0)+ ', σ='+ Floattostrf(sd,FFgeneral,3,0));
memo2_message('Streak found: '+filename2+', Y='+floattostrf(slope,FFgeneral,5,0)+'*X + '+Floattostrf(intercept,FFgeneral,5,0)+ ', σ='+ Floattostrf(sd,FFgeneral,3,0));
contour_array2:=nil;
streak_lines[nr_streak_lines].slope:=slope;
streak_lines[nr_streak_lines].intercept:=intercept;
inc(nr_streak_lines);
if nr_streak_lines>=length(streak_lines) then
setlength(streak_lines,nr_streak_lines+20); //get more memory
end;
end;
end;
end;
begin
restore_his:=false;
binning:=1;
if head.naxis3>1 then {colour image}
begin
memo2_message('Converting image to mono');
binX1_crop(1, img, img_bk);{crop image, make mono, no binning}
get_hist(0,img_bk);{get histogram of img and his_total. Required to get correct background value}
restore_his:=true;
end
else
if (bayerpat<>'') then {raw Bayer image}
begin
memo2_message('Binning raw image for streak detection');
binX2_crop(1, img {out}, img_bk);{combine values of 4 pixels and crop is required, Result is mono}
get_hist(0,img_bk);{get histogram of img and his_total. Required to get correct background value}
restore_his:=true;
binning:=2;
end
else
img_bk:=img; {In dynamic arrays, the assignment statement duplicates only the reference to the array, while SetLength does the job of physically copying/duplicating it, leaving two separate, independent dynamic arrays.}
ww:=Length(img_bk[0,0]); {width}
hh:=Length(img_bk[0]); {height}
streak_lines:=nil;
nr_streak_lines:=0;
setlength(streak_lines,20);//allow 20 streak lines
with mainwindow do
begin
if plot then
begin
Flipv:=mainwindow.flip_vertical1.Checked;
Fliph:=mainwindow.Flip_horizontal1.Checked;
image1.Canvas.Pen.Mode := pmMerge;
image1.Canvas.brush.Style:=bsClear;
image1.Canvas.font.color:=clLime;
image1.Canvas.Pen.Color := clYellow;
image1.Canvas.Pen.width := round(1+head.height/image1.height);{thickness lines}
fontsize:=round(max(10,8*head.height/image1.height));{adapt font to image dimensions}
image1.Canvas.font.size:=fontsize;
end;
setlength(img_sa,1,hh,ww);{set length of image array}
gaussian_blur2(img_bk, blur);{apply gaussian blur }
get_background(0,img_bk,{cblack=0} false{histogram is already available},true {calculate noise level},{out}bck);{calculate background level from peek histogram}
detection_level:=sigmafactor*bck.noise_level+ bck.backgr;
detection_grid:=strtoint2(stackmenu1.detection_grid1.text,400) div binning;
for fitsY:=0 to hh-1 do
for fitsX:=0 to ww-1 do
img_sa[0,fitsY,fitsX]:=-1;{mark as star free area}
for fitsY:=0 to hh-1 do
begin
for fitsX:=0 to ww-1 do
begin
if ((detection_grid<=0) or (frac(fitsX/detection_grid)=0) or (frac(fitsy/detection_grid)=0)) then //overlay of vertical and horizontal lines
if (( img_sa[0,fitsY,fitsX]<0){untested area} and (img_bk[0,fitsY,fitsX]>detection_level){star}) then {new star}
begin
find_contour(fitsX,fitsY);
if frac(fitsY/300)= 0 then
begin
application.processmessages;
if esc_pressed then break;
end;
end;
end;
end;
end;{with mainwindow}
if restore_his then
begin
img_bk:=nil;
get_hist(0,img);{get histogram of img and his_total}
end;
img_sa:=nil;{free mem}
end;
end.
|