import os
from imagededup.methods import PHash # pip install imagededup
phasher = PHash()
def process_file(img_path):
# 生成图像目录中所有图像的二值hash编码
encodings = phasher.encode_images(image_dir=img_path)
duplicates = phasher.find_duplicates(encoding_map=encodings)
only_img = [] # 唯一图片
like_img = [] # 相似图片
for img, img_list in duplicates.items():
if img not in only_img and img not in like_img:
only_img.append(img)
like_img.extend(img_list)
for p in like_img:
os.remove(os.path.join(img_path,p))
print('相似图片{}张，唯一图片{}张'.format(len(like_img),len(only_img)))
if __name__ == '__main__':
image_path = r'F:\2023_07_10\zhangyifei\pic_data\pic_01'
process_file(image_path)
标签：__,like,img,python,only,重复,算法,path,图片 From： https://www.cnblogs.com/lvye001/p/17541414.html