further untangle run

sbb_newspapers_org_image/eynollah.py

@@ -705,7 +705,9 @@ class eynollah:
         del img
         del imgray
 
+        K.clear_session()
         gc.collect()
+        self.logger.debug("exit extract_page")
         return croped_page, page_coord
 
     def extract_text_regions(self, img, patches, cols):
@@ -2140,6 +2142,45 @@ class eynollah:
             return self.do_order_of_regions_full_layout(*args, **kwargs)
         return self.do_order_of_regions_no_full_layout(*args, **kwargs)
 
+    def run_graphics_and_columns(self, text_regions_p_1, num_column_is_classified):
+        img_g = cv2.imread(self.image_filename, cv2.IMREAD_GRAYSCALE)
+        img_g = img_g.astype(np.uint8)
+
+        img_g3 = np.zeros((img_g.shape[0], img_g.shape[1], 3))
+        img_g3 = img_g3.astype(np.uint8)
+        img_g3[:, :, 0] = img_g[:, :]
+        img_g3[:, :, 1] = img_g[:, :]
+        img_g3[:, :, 2] = img_g[:, :]
+
+        image_page, page_coord = self.extract_page()
+        if self.plotter:
+            self.plotter.save_page_image(image_page)
+
+        img_g3_page = img_g3[page_coord[0] : page_coord[1], page_coord[2] : page_coord[3], :]
+
+        text_regions_p_1 = text_regions_p_1[page_coord[0] : page_coord[1], page_coord[2] : page_coord[3]]
+
+        mask_images = (text_regions_p_1[:, :] == 2) * 1
+        mask_images = mask_images.astype(np.uint8)
+        mask_images = cv2.erode(mask_images[:, :], self.kernel, iterations=10)
+
+        mask_lines = (text_regions_p_1[:, :] == 3) * 1
+        mask_lines = mask_lines.astype(np.uint8)
+
+        img_only_regions_with_sep = ((text_regions_p_1[:, :] != 3) & (text_regions_p_1[:, :] != 0)) * 1
+        img_only_regions_with_sep = img_only_regions_with_sep.astype(np.uint8)
+        img_only_regions = cv2.erode(img_only_regions_with_sep[:, :], self.kernel, iterations=6)
+
+        try:
+            num_col, peaks_neg_fin = find_num_col(img_only_regions, multiplier=6.0)
+            if not num_column_is_classified:
+                num_col_classifier = num_col + 1
+        except:
+            num_col = None
+            peaks_neg_fin = []
+            num_col_classifier = None
+        return num_col, num_col_classifier, img_only_regions, page_coord, image_page, mask_images, mask_lines
+
     def run_enhancement(self):
         self.logger.info("resize and enhance image")
         is_image_enhanced, img_org, img_res, _, num_column_is_classified = self.resize_and_enhance_image_with_column_classifier()
@@ -2163,6 +2204,7 @@ class eynollah:
                 self.get_image_and_scales_after_enhancing(img_org, img_res)
         return img_res, is_image_enhanced, num_column_is_classified
 
+
     def run(self):
         """
         Get image and scales, then extract the page of scanned image
@@ -2177,65 +2219,17 @@ class eynollah:
         t1 = time.time()
         text_regions_p_1 = self.get_regions_from_xy_2models(img_res, is_image_enhanced)
         self.logger.info("Textregion detection took %ss ", str(time.time() - t1))
-        t1 = time.time()
 
-        img_g = cv2.imread(self.image_filename, cv2.IMREAD_GRAYSCALE)
-        img_g = img_g.astype(np.uint8)
-
-        img_g3 = np.zeros((img_g.shape[0], img_g.shape[1], 3))
-        img_g3 = img_g3.astype(np.uint8)
-        img_g3[:, :, 0] = img_g[:, :]
-        img_g3[:, :, 1] = img_g[:, :]
-        img_g3[:, :, 2] = img_g[:, :]
-
-        image_page, page_coord = self.extract_page()
-        # print(image_page.shape,'page')
-        if self.plotter:
-            self.plotter.save_page_image(image_page)
-        K.clear_session()
-        gc.collect()
-
-        img_g3_page = img_g3[page_coord[0] : page_coord[1], page_coord[2] : page_coord[3], :]
-        del img_g3
-        del img_g
-
-        text_regions_p_1 = text_regions_p_1[page_coord[0] : page_coord[1], page_coord[2] : page_coord[3]]
-
-        mask_images = (text_regions_p_1[:, :] == 2) * 1
-        mask_images = mask_images.astype(np.uint8)
-        mask_images = cv2.erode(mask_images[:, :], self.kernel, iterations=10)
-
-        mask_lines = (text_regions_p_1[:, :] == 3) * 1
-        mask_lines = mask_lines.astype(np.uint8)
-
-        img_only_regions_with_sep = ((text_regions_p_1[:, :] != 3) & (text_regions_p_1[:, :] != 0)) * 1
-        img_only_regions_with_sep = img_only_regions_with_sep.astype(np.uint8)
-        img_only_regions = cv2.erode(img_only_regions_with_sep[:, :], self.kernel, iterations=6)
-
-        try:
-            num_col, peaks_neg_fin = find_num_col(img_only_regions, multiplier=6.0)
-            if not num_column_is_classified:
-                num_col_classifier = num_col + 1
-        except:
-            num_col = None
-            peaks_neg_fin = []
+        t1 = time.time()
+        num_col, num_col_classifier, img_only_regions, page_coord, image_page, mask_images, mask_lines = self.run_graphics_and_columns(text_regions_p_1, num_column_is_classified)
+        self.logger.info("Graphics detection took %ss ", str(time.time() - t1))
 
         #print(num_col, "num_colnum_col")
         if not num_col:
             self.logger.info("No columns detected, outputting an empty PAGE-XML")
-            txt_con_org = []
-            order_text_new = []
-            id_of_texts_tot = []
-            all_found_texline_polygons = []
-            all_box_coord = []
-            polygons_of_images = []
-            polygons_of_marginals = []
-            all_found_texline_polygons_marginals = []
-            all_box_coord_marginals = []
-            slopes = []
-            slopes_marginals = []
-            self.write_into_page_xml(txt_con_org, page_coord, self.dir_out, order_text_new, id_of_texts_tot, all_found_texline_polygons, all_box_coord, polygons_of_images, polygons_of_marginals, all_found_texline_polygons_marginals, all_box_coord_marginals, self.curved_line, slopes, slopes_marginals)
-        else:
+            self.write_into_page_xml([], page_coord, self.dir_out, [], [], [], [], [], [], [], [], self.curved_line, [], [])
+            self.logger.info("Job done in %ss", str(time.time() - t1))
+            return
         patches = True
         scaler_h_textline = 1  # 1.2#1.2
         scaler_w_textline = 1  # 0.9#1