option to use autoThresholding. Improved version from bachelor thesis…

… JBIG2 compression http://is.muni.cz/th/208155/fi_b/.

and https://github.com/rhatlapa/jbig2enc

AUTHORS

@@ -7,3 +7,4 @@ Contributors:
 Misty De Meo <[email protected]>
 zdenop <[email protected]>
 Steven Lee http://www.rubypdf.com
+Radim Hatlapatka <[email protected]>

ChangeLog

@@ -5,6 +5,8 @@
   * version info (-V --version)
   * pdf.py now correctly retains DPI from input images (thanks to Steven Lee
     http://blog.rubypdf.com/2011/09/09/jbig2-pdf-py-patch-the-right-way-to-get-dpi/)
+  * R. Hatlapatka: option to use autoThresholding. Improved version from
+    bachelor thesis JBIG2 compression http://is.muni.cz/th/208155/fi_b/.
 
 0.27:
   * Update to the latest Leptonica (1.58)

src/Makefile.am

@@ -1,8 +1,9 @@
 AM_CXXFLAGS = -Wall
 
 lib_LTLIBRARIES = libjbig2enc.la
-libjbig2enc_la_SOURCES = jbig2enc.cc jbig2arith.cc jbig2sym.cc
-include_HEADERS = jbig2arith.h jbig2sym.h jbig2structs.h jbig2segments.h
+libjbig2enc_la_SOURCES = jbig2enc.cc jbig2arith.cc jbig2sym.cc result.cc jbig2comparator.cc
+libjbig2enc_la_LDFLAGS = -no-undefined -version-info $(GENERIC_LIBRARY_VERSION)
+include_HEADERS = jbig2arith.h jbig2sym.h jbig2structs.h jbig2segments.h result.h jbig2comparator.h
 
 bin_PROGRAMS = jbig2
 jbig2_SOURCES = jbig2.cc

src/jbig2.cc

@@ -54,6 +54,8 @@ usage(const char *argv0) {
   fprintf(stderr, "  -4: upsample 4x before thresholding\n");
   fprintf(stderr, "  -S: remove images from mixed input and save separately\n");
   fprintf(stderr, "  -j --jpeg-output: write images from mixed input as JPEG\n");
+  fprintf(stderr, "  -a --autoThresh: engage using autoThresholding for symbol coder\n");
+  fprintf(stderr, "  --nohash: only for debugging purposes to allow easily compare performance with older version\n");
   fprintf(stderr, "  -V --version: version info\n");
   fprintf(stderr, "  -v: be verbose\n");
 }
@@ -209,13 +211,15 @@ main(int argc, char **argv) {
   l_int32 img_fmt = IFF_PNG;
   const char *img_ext = "png";
   bool segment = false;
+  bool autoThresh = false;
+  bool hash = true;
   int i;
-
-  #ifdef WIN32
-    int result = _setmode(_fileno(stdout), _O_BINARY);
-    if (result == -1)
-      fprintf(stderr, "Cannot set mode to binary for stdout\n");
-  #endif
+
+  #ifdef WIN32
+    int result = _setmode(_fileno(stdout), _O_BINARY);
+    if (result == -1)
+      fprintf(stderr, "Cannot set mode to binary for stdout\n");
+  #endif
 
   for (i = 1; i < argc; ++i) {
     if (strcmp(argv[i], "-h") == 0 ||
@@ -327,6 +331,18 @@ main(int argc, char **argv) {
       continue;
     }
 
+    // engage auto thresholding
+    if (strcmp(argv[i], "--autoThresh") == 0 || 
+        strcmp(argv[i], "-a") == 0 ) {
+      autoThresh = true;
+      continue;
+    }
+
+    if (strcmp(argv[i], "--nohash") == 0) {
+      hash = false;
+      continue;
+    }
+
     if (strcmp(argv[i], "-v") == 0) {
       verbose = true;
       continue;
@@ -340,7 +356,7 @@ main(int argc, char **argv) {
     usage(argv[0]);
     return 4;
   }
-
+ 
   if (refine && !symbol_mode) {
     fprintf(stderr, "Refinement makes not sense unless in symbol mode!\n");
     fprintf(stderr, "(if you have -r, you must have -s)\n");
@@ -360,17 +376,17 @@ main(int argc, char **argv) {
     if (subimage==numsubimages) {
       subimage = numsubimages = 0;
       FILE *fp;
-      if (verbose) fprintf(stderr, "Processing \"%s\"...\n", argv[i]);
-      if ((fp=fopen(argv[i], "r"))==NULL) {
-        fprintf(stderr, "Unable to open \"%s\"\n", argv[i]);
+      if (verbose) fprintf(stderr, "Processing \"%s\"...\n", argv[i]);
+      if ((fp=lept_fopen(argv[i], "r"))==NULL) {
+        fprintf(stderr, "Unable to open \"%s\"\n", argv[i]);
         return 1;
       }
       l_int32 filetype;
       findFileFormatStream(fp, &filetype);
       if (filetype==IFF_TIFF && tiffGetCount(fp, &numsubimages)) {
         return 1;
       }
-      fclose(fp);
+      lept_fclose(fp);
     }
 
     PIX *source;
@@ -455,6 +471,14 @@ main(int argc, char **argv) {
     }
   }
 
+    if (autoThresh) {
+    if (hash) {
+      autoThresholdUsingHash(ctx);
+    } else {
+      autoThreshold(ctx);
+    }
+  }
+
   uint8_t *ret;
   int length;
   ret = jbig2_pages_complete(ctx, &length);