From 291953838233b6bc202de5d198bc1fd8849b91d4 Mon Sep 17 00:00:00 2001
From: Robert Sachunsky <sachunsky@informatik.uni-leipzig.de>
Date: Tue, 1 Apr 2025 23:33:26 +0000
Subject: [PATCH] minor fixes to avoid frequent warnings
---
src/eynollah/eynollah.py | 15 ++++++---------
src/eynollah/sbb_binarize.py | 14 ++++----------
src/eynollah/utils/contour.py | 2 +-
src/eynollah/utils/pil_cv2.py | 8 +++++---
src/eynollah/utils/separate_lines.py | 2 +-
5 files changed, 17 insertions(+), 24 deletions(-)
diff --git a/src/eynollah/eynollah.py b/src/eynollah/eynollah.py
index 23e2458..ef8bcc6 100644
--- a/src/eynollah/eynollah.py
+++ b/src/eynollah/eynollah.py
@@ -32,7 +32,7 @@ from scipy.ndimage import gaussian_filter1d
from numba import cuda
from ocrd import OcrdPage
-from ocrd_utils import getLogger
+from ocrd_utils import getLogger, tf_disable_interactive_logs
try:
import torch
@@ -47,14 +47,11 @@ try:
except ImportError:
TrOCRProcessor = VisionEncoderDecoderModel = None
-os.environ["TF_CPP_MIN_LOG_LEVEL"] = "3"
#os.environ['CUDA_VISIBLE_DEVICES'] = '-1'
-stderr = sys.stderr
-sys.stderr = open(os.devnull, "w")
+tf_disable_interactive_logs()
import tensorflow as tf
from tensorflow.python.keras import backend as K
from tensorflow.keras.models import load_model
-sys.stderr = stderr
tf.get_logger().setLevel("ERROR")
warnings.filterwarnings("ignore")
# use tf1 compatibility for keras backend
@@ -3614,7 +3611,7 @@ class Eynollah:
for ij in range(len(all_found_textline_polygons[j])):
con_ind = all_found_textline_polygons[j][ij]
area = cv2.contourArea(con_ind)
- con_ind = con_ind.astype(np.float)
+ con_ind = con_ind.astype(float)
x_differential = np.diff( con_ind[:,0,0])
y_differential = np.diff( con_ind[:,0,1])
@@ -3718,7 +3715,7 @@ class Eynollah:
con_ind = all_found_textline_polygons[j]
#print(len(con_ind[:,0,0]),'con_ind[:,0,0]')
area = cv2.contourArea(con_ind)
- con_ind = con_ind.astype(np.float)
+ con_ind = con_ind.astype(float)
x_differential = np.diff( con_ind[:,0,0])
y_differential = np.diff( con_ind[:,0,1])
@@ -3821,7 +3818,7 @@ class Eynollah:
con_ind = all_found_textline_polygons[j][ij]
area = cv2.contourArea(con_ind)
- con_ind = con_ind.astype(np.float)
+ con_ind = con_ind.astype(float)
x_differential = np.diff( con_ind[:,0,0])
y_differential = np.diff( con_ind[:,0,1])
@@ -4053,7 +4050,7 @@ class Eynollah:
for j in range(len(all_found_textline_polygons)):
for i in range(len(all_found_textline_polygons[j])):
con_ind = all_found_textline_polygons[j][i]
- con_ind = con_ind.astype(np.float)
+ con_ind = con_ind.astype(float)
x_differential = np.diff( con_ind[:,0,0])
y_differential = np.diff( con_ind[:,0,1])
diff --git a/src/eynollah/sbb_binarize.py b/src/eynollah/sbb_binarize.py
index 36e9ab0..f43b6ba 100644
--- a/src/eynollah/sbb_binarize.py
+++ b/src/eynollah/sbb_binarize.py
@@ -4,24 +4,18 @@ Tool to load model and binarize a given image.
import sys
from glob import glob
-from os import environ, devnull
-from os.path import join
-from warnings import catch_warnings, simplefilter
import os
+import logging
import numpy as np
from PIL import Image
import cv2
-environ['TF_CPP_MIN_LOG_LEVEL'] = '3'
-stderr = sys.stderr
-sys.stderr = open(devnull, 'w')
+from ocrd_utils import tf_disable_interactive_logs
+tf_disable_interactive_logs()
import tensorflow as tf
from tensorflow.keras.models import load_model
from tensorflow.python.keras import backend as tensorflow_backend
-sys.stderr = stderr
-
-import logging
def resize_image(img_in, input_height, input_width):
return cv2.resize(img_in, (input_width, input_height), interpolation=cv2.INTER_NEAREST)
@@ -53,7 +47,7 @@ class SbbBinarizer:
del self.session
def load_model(self, model_name):
- model = load_model(join(self.model_dir, model_name), compile=False)
+ model = load_model(os.path.join(self.model_dir, model_name), compile=False)
model_height = model.layers[len(model.layers)-1].output_shape[1]
model_width = model.layers[len(model.layers)-1].output_shape[2]
n_classes = model.layers[len(model.layers)-1].output_shape[3]
diff --git a/src/eynollah/utils/contour.py b/src/eynollah/utils/contour.py
index be00db0..1adb943 100644
--- a/src/eynollah/utils/contour.py
+++ b/src/eynollah/utils/contour.py
@@ -247,7 +247,7 @@ def get_textregion_contours_in_org_image_light(cnts, img, slope_first, map=map):
img = cv2.resize(img, (int(img.shape[1]/6), int(img.shape[0]/6)), interpolation=cv2.INTER_NEAREST)
##cnts = list( (np.array(cnts)/2).astype(np.int16) )
#cnts = cnts/2
- cnts = [(i/6).astype(np.int) for i in cnts]
+ cnts = [(i/6).astype(int) for i in cnts]
results = map(partial(do_back_rotation_and_get_cnt_back,
img=img,
slope_first=slope_first,
diff --git a/src/eynollah/utils/pil_cv2.py b/src/eynollah/utils/pil_cv2.py
index 83ae47d..9f6913e 100644
--- a/src/eynollah/utils/pil_cv2.py
+++ b/src/eynollah/utils/pil_cv2.py
@@ -1,3 +1,4 @@
+from contextlib import nullcontext
from PIL import Image
import numpy as np
from ocrd_models import OcrdExif
@@ -17,12 +18,13 @@ def pil2cv(img):
def check_dpi(img):
try:
if isinstance(img, Image.Image):
- pil_image = img
+ pil_image = nullcontext(img)
elif isinstance(img, str):
pil_image = Image.open(img)
else:
- pil_image = cv2pil(img)
- exif = OcrdExif(pil_image)
+ pil_image = nullcontext(cv2pil(img))
+ with pil_image:
+ exif = OcrdExif(pil_image)
resolution = exif.resolution
if resolution == 1:
raise Exception()
diff --git a/src/eynollah/utils/separate_lines.py b/src/eynollah/utils/separate_lines.py
index 7e77afe..111f58b 100644
--- a/src/eynollah/utils/separate_lines.py
+++ b/src/eynollah/utils/separate_lines.py
@@ -1616,7 +1616,7 @@ def do_work_of_slopes_new(
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)
+ y_diff_mean = find_contours_mean_y_diff(textline_con_fil) if len(textline_con_fil) > 1 else np.NaN
if np.isnan(y_diff_mean):
slope_for_all = MAX_SLOPE
else: