Ви не можете вибрати більше 25 тем Теми мають розпочинатися з літери або цифри, можуть містити дефіси (-) і не повинні перевищувати 35 символів.
 
 
 
 
 

365 рядки
9.4 KiB

  1. /*
  2. * This file is part of Aptdec.
  3. * Copyright (c) 2004-2009 Thierry Leconte (F4DWV), Xerbo (xerbo@protonmail.com) 2019-2020
  4. *
  5. * Aptdec is free software: you can redistribute it and/or modify
  6. * it under the terms of the GNU General Public License as published by
  7. * the Free Software Foundation, either version 2 of the License, or
  8. * (at your option) any later version.
  9. *
  10. * This program is distributed in the hope that it will be useful,
  11. * but WITHOUT ANY WARRANTY; without even the implied warranty of
  12. * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  13. * GNU General Public License for more details.
  14. *
  15. * You should have received a copy of the GNU General Public License
  16. * along with this program. If not, see <https://www.gnu.org/licenses/>.
  17. *
  18. */
  19. #include <stdio.h>
  20. #include <string.h>
  21. #include <sndfile.h>
  22. #include <math.h>
  23. #include <stdlib.h>
  24. #include "offsets.h"
  25. #include "messages.h"
  26. #define REGORDER 3
  27. typedef struct {
  28. double cf[REGORDER + 1];
  29. } rgparam_t;
  30. typedef struct {
  31. float *prow[MAX_HEIGHT]; // Row buffers
  32. int nrow; // Number of rows
  33. int chA, chB; // ID of each channel
  34. char name[256]; // Stripped filename
  35. } image_t;
  36. typedef struct {
  37. char *type; // Output image type
  38. char *effects;
  39. int satnum; // The satellite number
  40. char *map; // Path to a map file
  41. char *path; // Output directory
  42. int realtime;
  43. } options_t;
  44. extern void polyreg(const int m, const int n, const double x[], const double y[], double c[]);
  45. // Compute regression
  46. static void rgcomp(double x[16], rgparam_t * rgpr) {
  47. // { 0.106, 0.215, 0.324, 0.433, 0.542, 0.652, 0.78, 0.87, 0.0 }
  48. const double y[9] = { 31.07, 63.02, 94.96, 126.9, 158.86, 191.1, 228.62, 255.0, 0.0 };
  49. polyreg(REGORDER, 9, x, y, rgpr -> cf);
  50. }
  51. // Convert a value to 0-255 based off the provided regression curve
  52. static double rgcal(float x, rgparam_t *rgpr) {
  53. double y, p;
  54. int i;
  55. for (i = 0, y = 0.0, p = 1.0; i < REGORDER + 1; i++) {
  56. y += rgpr->cf[i] * p;
  57. p = p * x;
  58. }
  59. return(y);
  60. }
  61. static double tele[16];
  62. static double Cs;
  63. void histogramEqualise(float **prow, int nrow, int offset, int width){
  64. // Plot histogram
  65. int histogram[256] = { 0 };
  66. for(int y = 0; y < nrow; y++)
  67. for(int x = 0; x < width; x++)
  68. histogram[(int)floor(prow[y][x+offset])]++;
  69. // Calculate cumulative frequency
  70. long sum = 0, cf[256] = { 0 };
  71. for(int i = 0; i < 255; i++){
  72. sum += histogram[i];
  73. cf[i] = sum;
  74. }
  75. // Apply histogram
  76. int area = nrow * width;
  77. for(int y = 0; y < nrow; y++){
  78. for(int x = 0; x < width; x++){
  79. int k = prow[y][x+offset];
  80. prow[y][x+offset] = (256.0/area) * cf[k];
  81. }
  82. }
  83. }
  84. // Brightness calibrate, including telemetry
  85. void calibrateImage(float **prow, int nrow, int offset, int width, rgparam_t regr){
  86. offset -= SYNC_WIDTH+SPC_WIDTH;
  87. for (int n = 0; n < nrow; n++) {
  88. float *pixelv = prow[n];
  89. for (int i = 0; i < width+SYNC_WIDTH+SPC_WIDTH+TELE_WIDTH; i++) {
  90. float pv = rgcal(pixelv[i + offset], &regr);
  91. pixelv[i + offset] = CLIP(pv, 0, 255);
  92. }
  93. }
  94. }
  95. double teleNoise(double wedges[16]){
  96. int pattern[9] = { 31, 63, 95, 127, 159, 191, 223, 255, 0 };
  97. double noise = 0;
  98. for(int i = 0; i < 9; i++)
  99. noise += fabs(wedges[i] - (double)pattern[i]);
  100. return noise;
  101. }
  102. // Get telemetry data for thermal calibration/equalization
  103. int calibrate(float **prow, int nrow, int offset, int width) {
  104. double teleline[MAX_HEIGHT] = { 0.0 };
  105. double wedge[16];
  106. rgparam_t regr[30];
  107. int telestart, mtelestart = 0;
  108. int channel = -1;
  109. // The minimum rows required to decode a full frame
  110. if (nrow < 192) {
  111. fprintf(stderr, ERR_TELE_ROW);
  112. return 0;
  113. }
  114. // Calculate average of a row of telemetry
  115. for (int n = 0; n < nrow; n++) {
  116. float *pixelv = prow[n];
  117. // Average the center 40px
  118. for (int i = 3; i < 43; i++)
  119. teleline[n] += pixelv[i + offset + width];
  120. teleline[n] /= 40.0;
  121. }
  122. /* Wedge 7 is white and 8 is black, this will have the largest
  123. * difference in brightness, this will always be in the center of
  124. * the frame and can thus be used to find the start of the frame
  125. */
  126. double max = 0.0;
  127. for (int n = nrow / 3 - 64; n < 2 * nrow / 3 - 64; n++) {
  128. float df;
  129. // (sum 4px below) / (sum 4px above)
  130. df = (teleline[n - 4] + teleline[n - 3] + teleline[n - 2] + teleline[n - 1]) /
  131. (teleline[n + 0] + teleline[n + 1] + teleline[n + 2] + teleline[n + 3]);
  132. // Find the maximum difference
  133. if (df > max) {
  134. mtelestart = n;
  135. max = df;
  136. }
  137. }
  138. // Find the start of the first frame
  139. telestart = (mtelestart - FRAME_LEN/2) % FRAME_LEN;
  140. // Make sure that theres at least one full frame in the image
  141. if (nrow < telestart + FRAME_LEN) {
  142. fprintf(stderr, ERR_TELE_ROW);
  143. return(0);
  144. }
  145. // Find the least noisy frame
  146. double minNoise = -1;
  147. int bestFrame = telestart;
  148. for (int n = telestart, k = 0; n < nrow - FRAME_LEN; n += FRAME_LEN, k++) {
  149. // Turn pixels into wedge values
  150. for (int j = 0; j < 16; j++) {
  151. wedge[j] = 0.0;
  152. // Average the middle 6px
  153. for (int i = 1; i < 7; i++)
  154. wedge[j] += teleline[(j * 8) + i + n];
  155. wedge[j] /= 6;
  156. }
  157. double noise = teleNoise(wedge);
  158. if(noise < minNoise || minNoise == -1){
  159. minNoise = noise;
  160. bestFrame = k;
  161. // Compute & apply regression on the wedges
  162. rgcomp(wedge, &regr[k]);
  163. for (int j = 0; j < 16; j++)
  164. tele[j] = rgcal(wedge[j], &regr[k]);
  165. /* Compare the channel ID wedge to the reference
  166. * wedges, the wedge with the closest match will
  167. * be the channel ID
  168. */
  169. float min = -1;
  170. for (int j = 0; j < 6; j++) {
  171. float df = tele[15] - tele[j];
  172. df *= df;
  173. if (df < min || min == -1) {
  174. channel = j;
  175. min = df;
  176. }
  177. }
  178. }
  179. }
  180. calibrateImage(prow, nrow, offset, width, regr[bestFrame]);
  181. return channel + 1;
  182. }
  183. // --- Temperature Calibration --- //
  184. #include "satcal.h"
  185. typedef struct {
  186. double Nbb;
  187. double Cs;
  188. double Cb;
  189. int ch;
  190. } tempparam_t;
  191. // IR channel temperature compensation
  192. static void tempcomp(double t[16], int ch, int satnum, tempparam_t *tpr) {
  193. double Tbb, T[4];
  194. double C;
  195. tpr -> ch = ch - 4;
  196. // Compute equivalent T blackbody temperature
  197. for (int n = 0; n < 4; n++) {
  198. float d0, d1, d2;
  199. C = t[9 + n] * 4.0;
  200. d0 = satcal[satnum].d[n][0];
  201. d1 = satcal[satnum].d[n][1];
  202. d2 = satcal[satnum].d[n][2];
  203. T[n] = d0;
  204. T[n] += d1 * C;
  205. C = C * C;
  206. T[n] += d2 * C;
  207. }
  208. Tbb = (T[0] + T[1] + T[2] + T[3]) / 4.0;
  209. Tbb = satcal[satnum].rad[tpr->ch].A + satcal[satnum].rad[tpr->ch].B * Tbb;
  210. // Compute radiance blackbody
  211. C = satcal[satnum].rad[tpr->ch].vc;
  212. tpr->Nbb = c1 * C * C * C / (expm1(c2 * C / Tbb));
  213. // Store count blackbody and space
  214. tpr->Cs = Cs * 4.0;
  215. tpr->Cb = t[14] * 4.0;
  216. }
  217. // IR channel temperature calibration
  218. static double tempcal(float Ce, int satnum, tempparam_t * rgpr) {
  219. double Nl, Nc, Ns, Ne;
  220. double T, vc;
  221. Ns = satcal[satnum].cor[rgpr->ch].Ns;
  222. Nl = Ns + (rgpr->Nbb - Ns) * (rgpr->Cs - Ce * 4.0) / (rgpr->Cs - rgpr->Cb);
  223. Nc = satcal[satnum].cor[rgpr->ch].b[0] +
  224. satcal[satnum].cor[rgpr->ch].b[1] * Nl +
  225. satcal[satnum].cor[rgpr->ch].b[2] * Nl * Nl;
  226. Ne = Nl + Nc;
  227. vc = satcal[satnum].rad[rgpr->ch].vc;
  228. T = c2 * vc / log1p(c1 * vc * vc * vc / Ne);
  229. T = (T - satcal[satnum].rad[rgpr->ch].A) / satcal[satnum].rad[rgpr->ch].B;
  230. // Rescale to 0-255 for -60'C to +40'C
  231. T = (T - 273.15 + 60.0) / 100.0 * 256.0;
  232. return(T);
  233. }
  234. // Temperature calibration wrapper
  235. void temperature(options_t *opts, image_t *img, int offset, int width){
  236. tempparam_t temp;
  237. printf("Temperature... ");
  238. fflush(stdout);
  239. tempcomp(tele, img->chB, opts->satnum - 15, &temp);
  240. for (int y = 0; y < img->nrow; y++) {
  241. float *pixelv = img->prow[y];
  242. for (int x = 0; x < width; x++) {
  243. float pv = tempcal(pixelv[x + offset], opts->satnum - 15, &temp);
  244. pixelv[x + offset] = CLIP(pv, 0, 255);
  245. }
  246. }
  247. printf("Done\n");
  248. }
  249. void distrib(options_t *opts, image_t *img, char *chid) {
  250. int max = 0;
  251. // Options
  252. options_t options;
  253. options.path = opts->path;
  254. // Image options
  255. image_t distrib;
  256. strcpy(distrib.name, img->name);
  257. distrib.nrow = 256;
  258. // Assign memory
  259. for(int i = 0; i < 256; i++)
  260. distrib.prow[i] = (float *) malloc(sizeof(float) * 256);
  261. for(int n = 0; n < img->nrow; n++) {
  262. float *pixelv = img->prow[n];
  263. for(int i = 0; i < CH_WIDTH; i++) {
  264. int y = CLIP((int)pixelv[i + CHA_OFFSET], 0, 255);
  265. int x = CLIP((int)pixelv[i + CHB_OFFSET], 0, 255);
  266. distrib.prow[y][x]++;
  267. if(distrib.prow[y][x] > max)
  268. max = distrib.prow[y][x];
  269. }
  270. }
  271. // Scale to 0-255
  272. for(int x = 0; x < 256; x++)
  273. for(int y = 0; y < 256; y++)
  274. distrib.prow[y][x] = distrib.prow[y][x] / max * 255.0;
  275. extern int ImageOut(options_t *opts, image_t *img, int offset, int width, char *desc, char *chid, char *palette);
  276. ImageOut(&options, &distrib, 0, 256, "Distribution", chid, NULL);
  277. }
  278. extern float quick_select(float arr[], int n);
  279. // Recursive biased median denoise
  280. #define TRIG_LEVEL 40
  281. void denoise(float **prow, int nrow, int offset, int width){
  282. for(int y = 2; y < nrow-2; y++){
  283. for(int x = offset+1; x < offset+width-1; x++){
  284. if(prow[y][x+1] - prow[y][x] > TRIG_LEVEL ||
  285. prow[y][x-1] - prow[y][x] > TRIG_LEVEL ||
  286. prow[y+1][x] - prow[y][x] > TRIG_LEVEL ||
  287. prow[y-1][x] - prow[y][x] > TRIG_LEVEL){
  288. prow[y][x] = quick_select((float[]){
  289. prow[y+2][x-1], prow[y+2][x], prow[y+2][x+1],
  290. prow[y+1][x-1], prow[y+1][x], prow[y+1][x+1],
  291. prow[y-1][x-1], prow[y-1][x], prow[y-1][x+1],
  292. prow[y-2][x-1], prow[y-2][x], prow[y-2][x+1]
  293. }, 12);
  294. }
  295. }
  296. }
  297. }
  298. #undef TRIG_LEVEL