Python——百度识图-相似图片爬虫下载解决方案

Prince_Treasure

已于 2023-12-21 14:24:35 修改

阅读量221

点赞数 2

文章标签： python 爬虫开发语言

于 2023-09-20 14:27:59 首次发布

本文链接：https://blog.csdn.net/Prince999999/article/details/133079125

版权

# -*- coding:utf-8 _*-
import json
import os
import re
import cv2
import requests


from urllib.parse import urlparse, parse_qs
import numpy as np
data = {
    'image':open(r"C:\Users\Lenovo\Desktop/2.jpg",'rb')
}
headers = {
    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/68.0.3440.106 Safari/537.36'
}
r = requests.post('https://graph.baidu.com/upload?tn=pc&from=pc&image_source=PC_UPLOAD_IMAGE_MOVE&range={%22page_from%22:%20%22shituIndex%22}&extUiData%5bisLogoShow%5d=1',files=data,headers = headers).text
url = json.loads(r)["data"]["url"]
o = urlparse(url)
q = parse_qs(o.query, True)
sign = q['sign'][0]
r1 = requests.get(url,headers = headers).text
r0 = requests.get("https://graph.baidu.com/ajax/pcsimi?sign={}".format(sign)).text
l = json.loads(r0)["data"]["list"]
 
for one in l:
    img_url = one["thumbUrl"]
    try:
        response = requests.get(img_url, timeout=1)
    except Exception as e:
        print("ERROR: download img timeout.")
 
    try:
        # imgDataNp = np.fromstring(response.content, dtype='uint8')
        imgDataNp = np.frombuffer(response.content, dtype='uint8')
        img = cv2.imdecode(imgDataNp, cv2.IMREAD_UNCHANGED)
 
        img_name = img_url.split('/')[-1]
        print(img_name)
        save_path = os.path.join(os.getcwd(), img_name)
        cv2.imwrite(save_path, img)
    except Exception as e:
        print(e)
        print("ERROR: download img corruption.")