How to get the letter coordinate retrieved by Tesseract ocr

问题

I'm trying to handle tesseract in python to just do simple job: - open a picture - run ocr - get the string - get the characters coordinates

The last one is my pain!

Here is my first code:

import tesseract
import glob
import cv2

api = tesseract.TessBaseAPI()
api.SetVariable("tessedit_char_whitelist", "0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZéèô%")
api.SetPageSegMode(tesseract.PSM_AUTO)

imagepath = "C:\\Project\\Bob\\"
imagePathList = glob.glob(imagepath + "*.jpg")

for image in imagePathList:
    mBuffer=open(imagePathList[10],"rb").read()
    result = tesseract.ProcessPagesBuffer(mBuffer,len(mBuffer),api)
    img = cv2.imread(image)
    cv2.putText(img,result,(20,20), cv2.FONT_HERSHEY_PLAIN, 1.0,(0,255,0))       
    cv2.imshow("Original",img)
    cv2.waitKey()

As my picture get various layouts, with different words at different positions, I would like to get a box for every char.

I have seen talking about: - api.getBoxText - Hocr

But no way has been found to implement it in Python.

回答1:

tesserocr provides the capability to access pretty much all of tesseract's API functionality. Here's an example that might be what you want:

from PIL import Image
from tesserocr import PyTessBaseAPI, RIL

image = Image.open('/usr/src/tesseract/testing/phototest.tif')
with PyTessBaseAPI() as api:
    api.SetImage(image)
    boxes = api.GetComponentImages(RIL.TEXTLINE, True)
    print 'Found {} textline image components.'.format(len(boxes))
    for i, (im, box, _, _) in enumerate(boxes):
        # im is a PIL image object
        # box is a dict with x, y, w and h keys
        api.SetRectangle(box['x'], box['y'], box['w'], box['h'])
        ocrResult = api.GetUTF8Text()
        conf = api.MeanTextConf()
        print (u"Box[{0}]: x={x}, y={y}, w={w}, h={h}, "
               "confidence: {1}, text: {2}").format(i, conf, ocrResult, **box)

You can also access other API methods such as GetHOCRText and GetBoxText among others.

However, right now it only supports *nix systems although a user successfully compiled it on Windows and provided binaries if you'd like to give it a try.

Disclaimer: tesserocr author here.

回答2:

You may want to call GetHOCRText method instead, if it's supported by the Python wrapper.

来源：https://stackoverflow.com/questions/18715742/how-to-get-the-letter-coordinate-retrieved-by-tesseract-ocr

标签

python

ocr

tesseract