fixed #38

qurator/eynollah/eynollah.py

@@ -159,6 +159,9 @@ class Eynollah:
             key += '_uint8'
         return self._imgs[key].copy()
     
+    def isNaN(self, num):
+        return num != num
+
 
     def predict_enhancement(self, img):
         self.logger.debug("enter predict_enhancement")
@@ -920,16 +923,16 @@ class Eynollah:
                     textline_con, hierarchy = return_contours_of_image(img_int_p)
                     textline_con_fil = filter_contours_area_of_image(img_int_p, textline_con, hierarchy, max_area=1, min_area=0.0008)
                     y_diff_mean = find_contours_mean_y_diff(textline_con_fil)
+                    if self.isNaN(y_diff_mean):
+                        slope_for_all = MAX_SLOPE
+                    else:
                         sigma_des = max(1, int(y_diff_mean * (4.0 / 40.0)))
-
                         img_int_p[img_int_p > 0] = 1
                         slope_for_all = return_deskew_slop(img_int_p, sigma_des, plotter=self.plotter)
 
                         if abs(slope_for_all) < 0.5:
                             slope_for_all = [slope_deskew][0]
-                    # old method
-                    # slope_for_all=self.textline_contours_to_get_slope_correctly(self.all_text_region_raw[mv],denoised,contours[mv])
-                    # text_patch_processed=textline_contours_postprocessing(gada)
+
                 except Exception as why:
                     self.logger.error(why)
                     slope_for_all = MAX_SLOPE
@@ -1031,6 +1034,9 @@ class Eynollah:
                     textline_con, hierarchy = return_contours_of_image(img_int_p)
                     textline_con_fil = filter_contours_area_of_image(img_int_p, textline_con, hierarchy, max_area=1, min_area=0.00008)
                     y_diff_mean = find_contours_mean_y_diff(textline_con_fil)
+                    if self.isNaN(y_diff_mean):
+                        slope_for_all = MAX_SLOPE
+                    else:
                         sigma_des = int(y_diff_mean * (4.0 / 40.0))
                         if sigma_des < 1:
                             sigma_des = 1
@@ -1891,6 +1897,7 @@ class Eynollah:
                 areas_cnt_text_d = np.array([cv2.contourArea(contours_only_text_parent_d[j]) for j in range(len(contours_only_text_parent_d))])
                 areas_cnt_text_d = areas_cnt_text_d / float(text_only_d.shape[0] * text_only_d.shape[1])
                 
+                if len(areas_cnt_text_d)>0:
                     contours_biggest_d = contours_only_text_parent_d[np.argmax(areas_cnt_text_d)]
                     index_con_parents_d=np.argsort(areas_cnt_text_d)
                     contours_only_text_parent_d=list(np.array(contours_only_text_parent_d)[index_con_parents_d] )
@@ -1937,6 +1944,12 @@ class Eynollah:
                 else:
                     contours_only_text_parent_d_ordered = []
                     contours_only_text_parent_d = []
+                    contours_only_text_parent = []
+                    
+            else:
+                contours_only_text_parent_d_ordered = []
+                contours_only_text_parent_d = []
+                contours_only_text_parent = []
                 
         else:
             contours_only_text, hir_on_text = return_contours_of_image(text_only)
@@ -1968,7 +1981,6 @@ class Eynollah:
         if not self.curved_line:
             slopes, all_found_texline_polygons, boxes_text, txt_con_org, contours_only_text_parent, all_box_coord, index_by_text_par_con = self.get_slopes_and_deskew_new(txt_con_org, contours_only_text_parent, textline_mask_tot_ea, image_page_rotated, boxes_text, slope_deskew)
             slopes_marginals, all_found_texline_polygons_marginals, boxes_marginals, _, polygons_of_marginals, all_box_coord_marginals, _ = self.get_slopes_and_deskew_new(polygons_of_marginals, polygons_of_marginals, textline_mask_tot_ea, image_page_rotated, boxes_marginals, slope_deskew)
-
         else:
             
             scale_param = 1