qurator-spk
/
eynollah
mirror of https://github.com/qurator-spk/eynollah.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

introduce self.batch_processing_mode to clarify when data is read from dir_in

Browse Source
refactoring-2024-08-merged
kba 4 months ago
parent 532ee6fe41
commit 9ee9c4403b
1 changed files with 36 additions and 35 deletions
Show Stats Download Patch File Download Diff File

71
qurator/eynollah/eynollah.py
Unescape Escape View File

@ -109,6 +109,7 @@ class Eynollah():
): ):
self.dirs = dirs self.dirs = dirs
self.logger = logger self.logger = logger
self.batch_processing_mode = bool(dirs.dir_in)
if not dirs.dir_in: if not dirs.dir_in:
if image_pil: if image_pil:
self._imgs = self._cache_images(image_pil=image_pil) self._imgs = self._cache_images(image_pil=image_pil)
@ -160,7 +161,7 @@ class Eynollah():
self.models : dict[str, tf.keras.Model] = {} self.models : dict[str, tf.keras.Model] = {}
if self.dirs.dir_in and light_version: if self.batch_processing_mode and light_version:
config = tf.compat.v1.ConfigProto() config = tf.compat.v1.ConfigProto()
config.gpu_options.allow_growth = True config.gpu_options.allow_growth = True
session = tf.compat.v1.Session(config=config) session = tf.compat.v1.Session(config=config)
@ -176,7 +177,7 @@ class Eynollah():
self.ls_imgs = listdir(self.dirs.dir_in) self.ls_imgs = listdir(self.dirs.dir_in)
if self.dirs.dir_in and not light_version: if self.batch_processing_mode and not light_version:
config = tf.compat.v1.ConfigProto() config = tf.compat.v1.ConfigProto()
config.gpu_options.allow_growth = True config.gpu_options.allow_growth = True
session = tf.compat.v1.Session(config=config) session = tf.compat.v1.Session(config=config)
@ -392,7 +393,7 @@ class Eynollah():
img = self.imread() img = self.imread()
_, page_coord = self.early_page_for_num_of_column_classification(img) _, page_coord = self.early_page_for_num_of_column_classification(img)
if not self.dirs.dir_in: if not self.batch_processing_mode:
model_num_classifier = self.load_model(self.model_dir_of_col_classifier) model_num_classifier = self.load_model(self.model_dir_of_col_classifier)
if self.input_binary: if self.input_binary:
img_in = np.copy(img) img_in = np.copy(img)
@ -416,7 +417,7 @@ class Eynollah():
img_in[0, :, :, 1] = img_1ch[:, :] img_in[0, :, :, 1] = img_1ch[:, :]
img_in[0, :, :, 2] = img_1ch[:, :] img_in[0, :, :, 2] = img_1ch[:, :]
if not self.dirs.dir_in: if not self.batch_processing_mode:
label_p_pred = model_num_classifier.predict(img_in, verbose=0) label_p_pred = model_num_classifier.predict(img_in, verbose=0)
else: else:
label_p_pred = self.model_classifier.predict(img_in, verbose=0) label_p_pred = self.model_classifier.predict(img_in, verbose=0)
@ -439,7 +440,7 @@ class Eynollah():
self.logger.info("Detected %s DPI", dpi) self.logger.info("Detected %s DPI", dpi)
if self.input_binary: if self.input_binary:
img = self.imread() img = self.imread()
if self.dirs.dir_in: if self.batch_processing_mode:
prediction_bin = self.do_prediction(True, img, self.model_bin) prediction_bin = self.do_prediction(True, img, self.model_bin)
else: else:
@ -461,7 +462,7 @@ class Eynollah():
t1 = time.time() t1 = time.time()
_, page_coord = self.early_page_for_num_of_column_classification(img_bin) _, page_coord = self.early_page_for_num_of_column_classification(img_bin)
if not self.dirs.dir_in: if not self.batch_processing_mode:
model_num_classifier = self.load_model(self.model_dir_of_col_classifier) model_num_classifier = self.load_model(self.model_dir_of_col_classifier)
if self.input_binary: if self.input_binary:
@ -483,7 +484,7 @@ class Eynollah():
img_in[0, :, :, 2] = img_1ch[:, :] img_in[0, :, :, 2] = img_1ch[:, :]
if self.dirs.dir_in: if self.batch_processing_mode:
label_p_pred = self.model_classifier.predict(img_in, verbose=0) label_p_pred = self.model_classifier.predict(img_in, verbose=0)
else: else:
label_p_pred = model_num_classifier.predict(img_in, verbose=0) label_p_pred = model_num_classifier.predict(img_in, verbose=0)
@ -873,10 +874,10 @@ class Eynollah():
if not self.ignore_page_extraction: if not self.ignore_page_extraction:
img = cv2.GaussianBlur(self.image, (5, 5), 0) img = cv2.GaussianBlur(self.image, (5, 5), 0)
if not self.dirs.dir_in: if not self.batch_processing_mode:
model_page = self.load_model(self.model_page_dir) model_page = self.load_model(self.model_page_dir)
if not self.dirs.dir_in: if not self.batch_processing_mode:
img_page_prediction = self.do_prediction(False, img, model_page) img_page_prediction = self.do_prediction(False, img, model_page)
else: else:
img_page_prediction = self.do_prediction(False, img, self.model_page) img_page_prediction = self.do_prediction(False, img, self.model_page)
@ -921,11 +922,11 @@ class Eynollah():
img = img.astype(np.uint8) img = img.astype(np.uint8)
else: else:
img = self.imread() img = self.imread()
if not self.dirs.dir_in: if not self.batch_processing_mode:
model_page = self.load_model(self.model_page_dir) model_page = self.load_model(self.model_page_dir)
img = cv2.GaussianBlur(img, (5, 5), 0) img = cv2.GaussianBlur(img, (5, 5), 0)
if self.dirs.dir_in: if self.batch_processing_mode:
img_page_prediction = self.do_prediction(False, img, self.model_page) img_page_prediction = self.do_prediction(False, img, self.model_page)
else: else:
img_page_prediction = self.do_prediction(False, img, model_page) img_page_prediction = self.do_prediction(False, img, model_page)
@ -954,7 +955,7 @@ class Eynollah():
self.logger.debug("enter extract_text_regions") self.logger.debug("enter extract_text_regions")
img_height_h = img.shape[0] img_height_h = img.shape[0]
img_width_h = img.shape[1] img_width_h = img.shape[1]
if not self.dirs.dir_in: if not self.batch_processing_mode:
model_region = self.load_model(self.model_region_dir_fully if patches else self.model_region_dir_fully_np) model_region = self.load_model(self.model_region_dir_fully if patches else self.model_region_dir_fully_np)
else: else:
model_region = self.model_region_fl if patches else self.model_region_fl_np model_region = self.model_region_fl if patches else self.model_region_fl_np
@ -1421,19 +1422,19 @@ class Eynollah():
def textline_contours(self, img, patches, scaler_h, scaler_w): def textline_contours(self, img, patches, scaler_h, scaler_w):
self.logger.debug('enter textline_contours') self.logger.debug('enter textline_contours')
if not self.dirs.dir_in: if not self.batch_processing_mode:
model_textline = self.load_model(self.model_textline_dir if patches else self.model_textline_dir_np) model_textline = self.load_model(self.model_textline_dir if patches else self.model_textline_dir_np)
img = img.astype(np.uint8) img = img.astype(np.uint8)
img_org = np.copy(img) img_org = np.copy(img)
img_h = img_org.shape[0] img_h = img_org.shape[0]
img_w = img_org.shape[1] img_w = img_org.shape[1]
img = resize_image(img_org, int(img_org.shape[0] * scaler_h), int(img_org.shape[1] * scaler_w)) img = resize_image(img_org, int(img_org.shape[0] * scaler_h), int(img_org.shape[1] * scaler_w))
if not self.dirs.dir_in: if not self.batch_processing_mode:
prediction_textline = self.do_prediction(patches, img, model_textline) prediction_textline = self.do_prediction(patches, img, model_textline)
else: else:
prediction_textline = self.do_prediction(patches, img, self.model_textline) prediction_textline = self.do_prediction(patches, img, self.model_textline)
prediction_textline = resize_image(prediction_textline, img_h, img_w) prediction_textline = resize_image(prediction_textline, img_h, img_w)
if not self.dirs.dir_in: if not self.batch_processing_mode:
prediction_textline_longshot = self.do_prediction(False, img, model_textline) prediction_textline_longshot = self.do_prediction(False, img, model_textline)
else: else:
prediction_textline_longshot = self.do_prediction(False, img, self.model_textline) prediction_textline_longshot = self.do_prediction(False, img, self.model_textline)
@ -1514,7 +1515,7 @@ class Eynollah():
img_h_new = int(img_org.shape[0] / float(img_org.shape[1]) * img_w_new) img_h_new = int(img_org.shape[0] / float(img_org.shape[1]) * img_w_new)
img_resized = resize_image(img,img_h_new, img_w_new ) img_resized = resize_image(img,img_h_new, img_w_new )
if not self.dirs.dir_in: if not self.batch_processing_mode:
model_bin = self.load_model(self.model_dir_of_binarization) model_bin = self.load_model(self.model_dir_of_binarization)
prediction_bin = self.do_prediction(True, img_resized, model_bin) prediction_bin = self.do_prediction(True, img_resized, model_bin)
else: else:
@ -1533,7 +1534,7 @@ class Eynollah():
textline_mask_tot_ea = self.run_textline(img_bin) textline_mask_tot_ea = self.run_textline(img_bin)
if not self.dirs.dir_in: if not self.batch_processing_mode:
model_region = self.load_model(self.model_region_dir_p_ens_light) model_region = self.load_model(self.model_region_dir_p_ens_light)
prediction_regions_org = self.do_prediction_new_concept(True, img_bin, model_region) prediction_regions_org = self.do_prediction_new_concept(True, img_bin, model_region)
else: else:
@ -1578,14 +1579,14 @@ class Eynollah():
img_height_h = img_org.shape[0] img_height_h = img_org.shape[0]
img_width_h = img_org.shape[1] img_width_h = img_org.shape[1]
if not self.dirs.dir_in: if not self.batch_processing_mode:
model_region = self.load_model(self.model_region_dir_p_ens) model_region = self.load_model(self.model_region_dir_p_ens)
ratio_y=1.3 ratio_y=1.3
ratio_x=1 ratio_x=1
img = resize_image(img_org, int(img_org.shape[0]*ratio_y), int(img_org.shape[1]*ratio_x)) img = resize_image(img_org, int(img_org.shape[0]*ratio_y), int(img_org.shape[1]*ratio_x))
if not self.dirs.dir_in: if not self.batch_processing_mode:
prediction_regions_org_y = self.do_prediction(True, img, model_region) prediction_regions_org_y = self.do_prediction(True, img, model_region)
else: else:
prediction_regions_org_y = self.do_prediction(True, img, self.model_region) prediction_regions_org_y = self.do_prediction(True, img, self.model_region)
@ -1607,7 +1608,7 @@ class Eynollah():
img = resize_image(img_org, int(img_org.shape[0]), int(img_org.shape[1]*(1.2 if is_image_enhanced else 1))) img = resize_image(img_org, int(img_org.shape[0]), int(img_org.shape[1]*(1.2 if is_image_enhanced else 1)))
if self.dirs.dir_in: if self.batch_processing_mode:
prediction_regions_org = self.do_prediction(True, img, self.model_region) prediction_regions_org = self.do_prediction(True, img, self.model_region)
else: else:
prediction_regions_org = self.do_prediction(True, img, model_region) prediction_regions_org = self.do_prediction(True, img, model_region)
@ -1617,12 +1618,12 @@ class Eynollah():
prediction_regions_org[(prediction_regions_org[:,:]==1) & (mask_zeros_y[:,:]==1)]=0 prediction_regions_org[(prediction_regions_org[:,:]==1) & (mask_zeros_y[:,:]==1)]=0
if not self.dirs.dir_in: if not self.batch_processing_mode:
model_region = self.load_model(self.model_region_dir_p2) model_region = self.load_model(self.model_region_dir_p2)
img = resize_image(img_org, int(img_org.shape[0]), int(img_org.shape[1])) img = resize_image(img_org, int(img_org.shape[0]), int(img_org.shape[1]))
if self.dirs.dir_in: if self.batch_processing_mode:
prediction_regions_org2 = self.do_prediction(True, img, self.model_region_p2, 0.2) prediction_regions_org2 = self.do_prediction(True, img, self.model_region_p2, 0.2)
else: else:
prediction_regions_org2 = self.do_prediction(True, img, model_region, 0.2) prediction_regions_org2 = self.do_prediction(True, img, model_region, 0.2)
@ -1656,7 +1657,7 @@ class Eynollah():
if self.input_binary: if self.input_binary:
prediction_bin = np.copy(img_org) prediction_bin = np.copy(img_org)
else: else:
if not self.dirs.dir_in: if not self.batch_processing_mode:
model_bin = self.load_model(self.model_dir_of_binarization) model_bin = self.load_model(self.model_dir_of_binarization)
prediction_bin = self.do_prediction(True, img_org, model_bin) prediction_bin = self.do_prediction(True, img_org, model_bin)
else: else:
@ -1669,7 +1670,7 @@ class Eynollah():
prediction_bin =np.repeat(prediction_bin[:, :, np.newaxis], 3, axis=2) prediction_bin =np.repeat(prediction_bin[:, :, np.newaxis], 3, axis=2)
if not self.dirs.dir_in: if not self.batch_processing_mode:
model_region = self.load_model(self.model_region_dir_p_ens) model_region = self.load_model(self.model_region_dir_p_ens)
ratio_y=1 ratio_y=1
ratio_x=1 ratio_x=1
@ -1677,7 +1678,7 @@ class Eynollah():
img = resize_image(prediction_bin, int(img_org.shape[0]*ratio_y), int(img_org.shape[1]*ratio_x)) img = resize_image(prediction_bin, int(img_org.shape[0]*ratio_y), int(img_org.shape[1]*ratio_x))
if not self.dirs.dir_in: if not self.batch_processing_mode:
prediction_regions_org = self.do_prediction(True, img, model_region) prediction_regions_org = self.do_prediction(True, img, model_region)
else: else:
prediction_regions_org = self.do_prediction(True, img, self.model_region) prediction_regions_org = self.do_prediction(True, img, self.model_region)
@ -1709,7 +1710,7 @@ class Eynollah():
if self.input_binary: if self.input_binary:
prediction_bin = np.copy(img_org) prediction_bin = np.copy(img_org)
if not self.dirs.dir_in: if not self.batch_processing_mode:
model_bin = self.load_model(self.model_dir_of_binarization) model_bin = self.load_model(self.model_dir_of_binarization)
prediction_bin = self.do_prediction(True, img_org, model_bin) prediction_bin = self.do_prediction(True, img_org, model_bin)
else: else:
@ -1724,7 +1725,7 @@ class Eynollah():
prediction_bin =np.repeat(prediction_bin[:, :, np.newaxis], 3, axis=2) prediction_bin =np.repeat(prediction_bin[:, :, np.newaxis], 3, axis=2)
if not self.dirs.dir_in: if not self.batch_processing_mode:
model_region = self.load_model(self.model_region_dir_p_ens) model_region = self.load_model(self.model_region_dir_p_ens)
else: else:
@ -1734,7 +1735,7 @@ class Eynollah():
img = resize_image(prediction_bin, int(img_org.shape[0]*ratio_y), int(img_org.shape[1]*ratio_x)) img = resize_image(prediction_bin, int(img_org.shape[0]*ratio_y), int(img_org.shape[1]*ratio_x))
if not self.dirs.dir_in: if not self.batch_processing_mode:
prediction_regions_org = self.do_prediction(True, img, model_region) prediction_regions_org = self.do_prediction(True, img, model_region)
else: else:
prediction_regions_org = self.do_prediction(True, img, self.model_region) prediction_regions_org = self.do_prediction(True, img, self.model_region)
@ -2733,12 +2734,12 @@ class Eynollah():
t0_tot = time.time() t0_tot = time.time()
if not self.dirs.dir_in: if not self.batch_processing_mode:
self.ls_imgs = [1] self.ls_imgs = [1]
for img_name in self.ls_imgs: for img_name in self.ls_imgs:
t0 = time.time() t0 = time.time()
if self.dirs.dir_in: if self.batch_processing_mode:
self.reset_file_name_dir(join(self.dirs.dir_in,img_name)) self.reset_file_name_dir(join(self.dirs.dir_in,img_name))
img_res, is_image_enhanced, num_col_classifier, num_column_is_classified = self.run_enhancement(self.light_version) img_res, is_image_enhanced, num_col_classifier, num_column_is_classified = self.run_enhancement(self.light_version)
@ -2767,7 +2768,7 @@ class Eynollah():
self.logger.info("No columns detected, outputting an empty PAGE-XML") self.logger.info("No columns detected, outputting an empty PAGE-XML")
pcgts = self.writer.build_pagexml_no_full_layout([], page_coord, [], [], [], [], [], [], [], [], [], [], cont_page, [], []) pcgts = self.writer.build_pagexml_no_full_layout([], page_coord, [], [], [], [], [], [], [], [], [], [], cont_page, [], [])
self.logger.info("Job done in %.1fs", time.time() - t1) self.logger.info("Job done in %.1fs", time.time() - t1)
if self.dirs.dir_in: if self.batch_processing_mode:
self.writer.write_pagexml(pcgts) self.writer.write_pagexml(pcgts)
continue continue
else: else:
@ -2995,7 +2996,7 @@ class Eynollah():
pcgts = self.writer.build_pagexml_full_layout(contours_only_text_parent, contours_only_text_parent_h, page_coord, order_text_new, id_of_texts_tot, all_found_textline_polygons, all_found_textline_polygons_h, all_box_coord, all_box_coord_h, polygons_of_images, contours_tables, polygons_of_drop_capitals, polygons_of_marginals, all_found_textline_polygons_marginals, all_box_coord_marginals, slopes, slopes_h, slopes_marginals, cont_page, polygons_lines_xml) pcgts = self.writer.build_pagexml_full_layout(contours_only_text_parent, contours_only_text_parent_h, page_coord, order_text_new, id_of_texts_tot, all_found_textline_polygons, all_found_textline_polygons_h, all_box_coord, all_box_coord_h, polygons_of_images, contours_tables, polygons_of_drop_capitals, polygons_of_marginals, all_found_textline_polygons_marginals, all_box_coord_marginals, slopes, slopes_h, slopes_marginals, cont_page, polygons_lines_xml)
self.logger.info("Job done in %.1fs", time.time() - t0) self.logger.info("Job done in %.1fs", time.time() - t0)
if not self.dirs.dir_in: if not self.batch_processing_mode:
return pcgts return pcgts
else: else:
contours_only_text_parent_h = None contours_only_text_parent_h = None
@ -3006,11 +3007,11 @@ class Eynollah():
order_text_new, id_of_texts_tot = self.do_order_of_regions(contours_only_text_parent_d_ordered, contours_only_text_parent_h, boxes_d, textline_mask_tot_d) order_text_new, id_of_texts_tot = self.do_order_of_regions(contours_only_text_parent_d_ordered, contours_only_text_parent_h, boxes_d, textline_mask_tot_d)
pcgts = self.writer.build_pagexml_no_full_layout(txt_con_org, page_coord, order_text_new, id_of_texts_tot, all_found_textline_polygons, all_box_coord, polygons_of_images, polygons_of_marginals, all_found_textline_polygons_marginals, all_box_coord_marginals, slopes, slopes_marginals, cont_page, polygons_lines_xml, contours_tables) pcgts = self.writer.build_pagexml_no_full_layout(txt_con_org, page_coord, order_text_new, id_of_texts_tot, all_found_textline_polygons, all_box_coord, polygons_of_images, polygons_of_marginals, all_found_textline_polygons_marginals, all_box_coord_marginals, slopes, slopes_marginals, cont_page, polygons_lines_xml, contours_tables)
self.logger.info("Job done in %.1fs", time.time() - t0) self.logger.info("Job done in %.1fs", time.time() - t0)
if not self.dirs.dir_in: if not self.batch_processing_mode:
return pcgts return pcgts
if self.dirs.dir_in: if self.batch_processing_mode:
self.writer.write_pagexml(pcgts) self.writer.write_pagexml(pcgts)
#self.logger.info("Job done in %.1fs", time.time() - t0) #self.logger.info("Job done in %.1fs", time.time() - t0)
if self.dirs.dir_in: if self.batch_processing_mode:
self.logger.info("All jobs done in %.1fs", time.time() - t0_tot) self.logger.info("All jobs done in %.1fs", time.time() - t0_tot)

Write Preview
Loading…
Cancel
Save