Fixed issues 899/1220/1246 (mixed eng+ara)

lsfjGitHub · Sep 18, 2014 · 2f197cd · 2f197cd
1 parent e46b605
commit 2f197cd
Show file tree

Hide file tree

Showing 4 changed files with 20 additions and 6 deletions.
diff --git a/ccmain/control.cpp b/ccmain/control.cpp
@@ -357,7 +357,7 @@ bool Tesseract::recog_all_words(PAGE_RES* page_res,
 
   // ****************** Pass 2 *******************
   if (tessedit_tess_adaption_mode != 0x0 && !tessedit_test_adaption &&
-      tessedit_ocr_engine_mode != OEM_CUBE_ONLY ) {
+      AnyTessLang()) {
     page_res_it.restart_page();
     GenericVector<WordData> words;
     SetupAllWordsPassN(2, target_word_box, word_config, page_res, &words);
@@ -371,8 +371,7 @@ bool Tesseract::recog_all_words(PAGE_RES* page_res,
 
   // The next passes can only be run if tesseract has been used, as cube
   // doesn't set all the necessary outputs in WERD_RES.
-  if (tessedit_ocr_engine_mode == OEM_TESSERACT_ONLY ||
-      tessedit_ocr_engine_mode == OEM_TESSERACT_CUBE_COMBINED) {
+  if (AnyTessLang()) {
     // ****************** Pass 3 *******************
     // Fix fuzzy spaces.
     set_global_loc_code(LOC_FUZZY_SPACE);
@@ -1098,6 +1097,9 @@ void Tesseract::classify_word_pass2(const WordData& word_data,
       tessedit_ocr_engine_mode != OEM_TESSERACT_CUBE_COMBINED &&
       word_data.word->best_choice != NULL)
     return;
+  if (tessedit_ocr_engine_mode == OEM_CUBE_ONLY) {
+    return;
+  }
   ROW* row = word_data.row;
   BLOCK* block = word_data.block;
   WERD_RES* word = *in_word;

diff --git a/ccmain/cube_control.cpp b/ccmain/cube_control.cpp
@@ -383,8 +383,8 @@ bool Tesseract::cube_recognize(CubeObject *cube_obj, BLOCK* block,
   for (int i = 0; i < num_chars; ++i) {
     UNICHAR_ID uch_id =
         cube_cntxt_->CharacterSet()->UnicharID(char_samples[i]->StrLabel());
-    choices[i] = new BLOB_CHOICE(uch_id, 0.0, cube_certainty, -1, -1,
-                                 0, 0, 0, 0, BCC_STATIC_CLASSIFIER);
+    choices[i] = new BLOB_CHOICE(uch_id, -cube_certainty, cube_certainty,
+                                 -1, -1, 0, 0, 0, 0, BCC_STATIC_CLASSIFIER);
   }
   word->FakeClassifyWord(num_chars, choices);
   // within a word, cube recognizes the word in reading order.

diff --git a/ccmain/tesseractclass.h b/ccmain/tesseractclass.h
@@ -245,6 +245,15 @@ class Tesseract : public Wordrec {
   Tesseract* get_sub_lang(int index) const {
     return sub_langs_[index];
   }
+  // Returns true if any language uses Tesseract (as opposed to cube).
+  bool AnyTessLang() const {
+    if (tessedit_ocr_engine_mode != OEM_CUBE_ONLY) return true;
+    for (int i = 0; i < sub_langs_.size(); ++i) {
+      if (sub_langs_[i]->tessedit_ocr_engine_mode != OEM_CUBE_ONLY)
+        return true;
+    }
+    return false;
+  }
 
   void SetBlackAndWhitelist();
 

diff --git a/ccstruct/pageres.cpp b/ccstruct/pageres.cpp
@@ -485,7 +485,10 @@ void WERD_RES::DebugWordChoices(bool debug, const char* word_to_debug) {
 void WERD_RES::DebugTopChoice(const char* msg) const {
   tprintf("Best choice: accepted=%d, adaptable=%d, done=%d : ",
           tess_accepted, tess_would_adapt, done);
-  best_choice->print(msg);
+  if (best_choice == NULL)
+    tprintf("<Null choice>\n");
+  else
+    best_choice->print(msg);
 }
 
 // Removes from best_choices all choices which are not within a reasonable