Spaces:
Sleeping
Sleeping
File size: 2,775 Bytes
8eb0b3e |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 |
import cv2
import os
def preprocess(img):
"""Applies Otsu's thresholding to the input image."""
# Ensure input is grayscale if it's not already
gray_img = img
if len(img.shape) == 3 and img.shape[2] == 3:
gray_img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
_, thresh_otsu = cv2.threshold(gray_img, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
return thresh_otsu
# def preprocess(img):
# """Applies Otsu's thresholding to the input image."""
# import numpy as np
# # Ensure input is grayscale if it's not already
# # Apply adaptive Gaussian thresholding
# ## invert image
# block_size = 11 # Size of pixel neighborhood used for threshold calculation
# C = 10 # Constant subtracted from mean
# thresh_adaptive = cv2.adaptiveThreshold(img,
# maxValue=255,
# adaptiveMethod=cv2.ADAPTIVE_THRESH_GAUSSIAN_C,
# thresholdType=cv2.THRESH_BINARY,
# blockSize=block_size,
# C=C)
# kernel = np.ones((3,3), np.uint8)
# thresh_cleaned = cv2.morphologyEx(thresh_adaptive, cv2.MORPH_OPEN, kernel)
# # or
# thresh_cleaned = cv2.medianBlur(thresh_adaptive, 3)
# gray_img = cv2.bitwise_not(thresh_cleaned)
# return gray_img
def load_and_preprocess_image(image_path: str, config: dict):
if not os.path.exists(image_path):
raise FileNotFoundError(f"Image file not found at: {image_path}")
img_color = cv2.imread(image_path)
if img_color is None:
raise ValueError(f"Could not read image file: {image_path}")
img_gray = cv2.cvtColor(img_color, cv2.COLOR_BGR2GRAY)
# --- Upscaling Heuristic ---
cfg_proc = config.get('image_processing', {})
min_dimension_threshold = cfg_proc.get('min_dimension_threshold', 800)
upscale_factor = cfg_proc.get('upscale_factor', 2)
h, w = img_gray.shape
img_color_resized = img_color
img_gray_resized = img_gray
if h < min_dimension_threshold or w < min_dimension_threshold:
print(f"Image dimensions ({w}x{h}) below threshold ({min_dimension_threshold}px). Upscaling by {upscale_factor}x.")
new_w, new_h = w * upscale_factor, h * upscale_factor
img_gray_resized = cv2.resize(img_gray, (new_w, new_h), interpolation=cv2.INTER_LANCZOS4)
img_color_resized = cv2.resize(img_color, (new_w, new_h), interpolation=cv2.INTER_LANCZOS4)
# --- Initial Preprocessing (Inversion + Thresholding) ---
img_inverted = cv2.bitwise_not(img_gray_resized)
preprocessed_img = preprocess(img_inverted)
return preprocessed_img, img_color_resized, img_gray_resized |