|
|
|
|
|
import os
|
|
|
|
|
|
import time
|
|
|
|
|
|
from PIL import Image
|
|
|
|
|
|
|
|
|
|
|
|
# 导入配置
|
|
|
|
|
|
from Config import target_width, image_quality, compressed_images_dir
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def process_image(img_path, target_width=target_width):
|
|
|
|
|
|
"""
|
|
|
|
|
|
处理图像
|
|
|
|
|
|
:param img_path: 输入图像路径
|
|
|
|
|
|
:param target_width: 目标宽度（默认587像素，根据用户提供的好用图片尺寸）
|
|
|
|
|
|
:param output_path: 输出结果的文件路径（可选）
|
|
|
|
|
|
:param raw_output_path: 原始OCR结果的输出文件路径（可选）
|
|
|
|
|
|
:return: OCR识别结果
|
|
|
|
|
|
"""
|
|
|
|
|
|
start_time = time.time()
|
|
|
|
|
|
|
|
|
|
|
|
# 图片预处理 - 缩小图片尺寸到指定宽度
|
|
|
|
|
|
img = Image.open(img_path)
|
|
|
|
|
|
width, height = img.size
|
|
|
|
|
|
print(f"原始图片尺寸: {width}x{height}")
|
|
|
|
|
|
|
|
|
|
|
|
# 计算缩放比例，保持纵横比
|
|
|
|
|
|
scale = target_width / width
|
|
|
|
|
|
new_width = target_width
|
|
|
|
|
|
new_height = int(height * scale)
|
|
|
|
|
|
|
|
|
|
|
|
print(f"目标图片尺寸: {new_width}x{new_height}")
|
|
|
|
|
|
|
|
|
|
|
|
# 缩放图像
|
|
|
|
|
|
resized_img = img.resize((new_width, new_height), Image.LANCZOS)
|
|
|
|
|
|
|
|
|
|
|
|
# 确保压缩图片目录存在
|
|
|
|
|
|
os.makedirs(compressed_images_dir, exist_ok=True)
|
|
|
|
|
|
|
|
|
|
|
|
# 获取原始文件名并创建新的输出路径
|
|
|
|
|
|
file_name = os.path.basename(img_path)
|
|
|
|
|
|
output_name = os.path.splitext(file_name)[0] + "_compressed.jpg"
|
|
|
|
|
|
output_path = os.path.join(compressed_images_dir, output_name)
|
|
|
|
|
|
|
|
|
|
|
|
# 保存缩小后的图片
|
|
|
|
|
|
resized_img.convert('RGB').save(output_path, 'JPEG', quality=image_quality, optimize=True)
|
|
|
|
|
|
print(f"图片已缩小: {width}x{height} -> {new_width}x{new_height}")
|
|
|
|
|
|
|
|
|
|
|
|
# 释放内存
|
|
|
|
|
|
del resized_img
|
|
|
|
|
|
img.close()
|
|
|
|
|
|
del img
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
if __name__ == "__main__":
|
|
|
|
|
|
import glob
|
|
|
|
|
|
from Config import images_dir
|
|
|
|
|
|
|
|
|
|
|
|
# 处理Images目录下的所有PNG图片
|
|
|
|
|
|
image_files = glob.glob(os.path.join(images_dir, "page_*.png"))
|
|
|
|
|
|
|
|
|
|
|
|
print(f"找到{len(image_files)}个图片文件需要处理")
|
|
|
|
|
|
|
|
|
|
|
|
for i, img_path in enumerate(image_files):
|
|
|
|
|
|
print(f"[{i+1}/{len(image_files)}] 处理图片: {os.path.basename(img_path)}")
|
|
|
|
|
|
process_image(img_path)
|
|
|
|
|
|
|
|
|
|
|
|
print(f"所有图片处理完成，压缩后的图片保存在: {compressed_images_dir}")
|
