유사한 사진 찾아주는 Python 코드

유사하거나 중복 사진을 찾아주는 파이썬 예제 코드입니다. 

 

아래 링크에 있는 코드를 수정하여 사용했습니다. 

https://github.com/cw-somil/Duplicate-Remover

2022. 11. 16  최초작성

from PIL import Image import imagehash import cv2 import os import numpy as np class DuplicateRemover:     def __init__(self, hash_size = 256, similarity= 80):         self.hash_size = hash_size         self.similarity = similarity         def get_hash(self, filename):         img = cv2.imread(filename)         img = img[:,:,:3] # discard alpha channel         img = Image.fromarray(img)         temp_hash_TF = imagehash.average_hash(img, self.hash_size).hash         return [filename, temp_hash_TF]                 def find_duplicates1(self, dirname, filename):                 fnames = os.listdir(dirname)         print("Finding Duplicates Now!\n")         threshold = 1 - self.similarity/100         diff_limit = int(threshold*(self.hash_size**2))         print('diff_limit', diff_limit)         list_item = []         for image in fnames:             f = os.path.join(dirname,image)             item = self.get_hash(f)             list_item.append(item)         new_item = self.get_hash(filename)         found = None         for item in list_item:                   diff = np.count_nonzero( item[1] != new_item[1])             if diff <= diff_limit:                 found = [item, diff]                 break                         if found != None:             print("Duplicate {} \nfound for Image {}!  {}\n".format(found[0][0], new_item[0], found[1]))     def find_duplicates2(self, dirname):                 fnames = os.listdir(dirname)         list_item = []         list_duplicate = []         print("Finding Duplicates Now!\n")         threshold = 1 - self.similarity/100         diff_limit = int(threshold*(self.hash_size**2))         print('diff_limit', diff_limit)         for image in fnames:                         f = os.path.join(dirname,image)             new_item = self.get_hash(f)             found= None             for item in list_item:                                 diff = np.count_nonzero(item[1] != new_item[1])                 if diff <= diff_limit:                     found = [item, diff]                     break                                 if found != None:                 print("Duplicate {} \nfound for Image {}! {}\n".format(found[0][0], new_item[0], found[1]))                 list_duplicate.append(found)             else:                 list_item.append(new_item) dr = DuplicateRemover(similarity=85) # dirname1 경로에 filename1과 유사한 사진이 있는지 체크합니다. dirname1 = "./test copy" filename1 = './test copy/rabbit1.jpg' dr.find_duplicates1(dirname1, filename1) print('-'*80) # dirname2 경로에 유사한 사진이 있는지 체크합니다. dirname2 = "./test" dr.find_duplicates2(dirname2)