标签:hot 百度搜 toolbar pac ima from x86 ocr his
要识别中文,ocr引擎要下载一个中文包chi_sim放进Tesseract-OCR\tessdata里面。安装完ocr后还要配置一下调用路径,在Python36\Lib\site-packages\pytesseract找到pytesseract.py(这是我的windows路径),打开在里面添加一下路径:
1 # CHANGE THIS IF TESSERACT IS NOT IN YOUR PATH, OR IS NAMED DIFFERENTLY 2 tesseract_cmd = ‘tesseract‘ 3 tesseract_cmd = ‘C:/Program Files (x86)/Tesseract-OCR/tesseract.exe‘ 4 img_mode = ‘RGB‘
然后用AirDroid或者Vysor、360手机助手之类的将手机的屏幕投影到电脑上,用鼠标确定截图坐标,代码如下:
1 import pyautogui as pag 2 x,y = pag.position() 3 posStr = "position:"+str(x).rjust(4)+‘,‘+str(y).rjust(4) 4 print(posStr)
要获取两个坐标(截图开始坐标和结束坐标),然后利用获取的坐标运用如下代码截图并调用ocr引擎识别(识别出来的字是每个用空格分开的,所以要去除字符串中的空格),代码如下:
1 from PIL import Image
2 from PIL import ImageGrab
3 import pytesseract
4 import webbrowser
5
6 pos = (0,245,425, 327)
7 cut_img = ImageGrab.grab(pos)
8 cut_img.save(‘C:/imgSave/1.jpg‘) #保存截图到文件夹中
9 print("screenshots sucess")
10
11 text=pytesseract.image_to_string(Image.open(‘C:/imgSave/1.jpg‘),lang=‘chi_sim‘) #调用识别引擎识别(chi_sim是中文的识别引擎,如果需要识别英文需要下载英文的识别引擎)
12 text=text.replace(" ","") #去空格
13 print(text)
14 url = ‘http://www.baidu.com/s?wd=%s‘ % text #调用百度搜索
15 webbrowser.open(url)
标签:hot 百度搜 toolbar pac ima from x86 ocr his
原文地址:https://www.cnblogs.com/moying-wq/p/10704772.html