Crop RGBImages from https://sourceforge.net/p/videosubfinder to only include subtitles identified in TXTImages for use with Google Vision API

crop.py

import cv2,os,numpy,math,glob

path = '/some/place/'
frames = [os.path.basename(x) for x in glob.glob(path + 'TXTImages/*.jpeg')]
#add padding around sub
pad = 40

for frame in frames:
    print(frame)
    txtpath = path + 'TXTImages/' + frame
    croppath = path + 'RGBCropped/' + frame
    rgbpath = path + 'RGBImages/' + frame
    #load txt image and find bounding box of contents
    im = cv2.imread(txtpath)
    imgray = cv2.cvtColor(im,cv2.COLOR_BGR2GRAY)
    imgray = cv2.blur(imgray,(15,15))
    ret,thresh = cv2.threshold(imgray,math.floor(numpy.average(imgray)),255,cv2.THRESH_BINARY_INV)
    dilated=cv2.morphologyEx(thresh, cv2.MORPH_OPEN, cv2.getStructuringElement(cv2.MORPH_ELLIPSE,(10,10)))
    contours,_ = cv2.findContours(dilated,cv2.RETR_LIST,cv2.CHAIN_APPROX_SIMPLE)

    best_box=[-1,-1,-1,-1]
    for c in contours:
        x,y,w,h = cv2.boundingRect(c)
        if best_box[0] < 0:
            best_box=[x,y,x+w,y+h]
        else:
            if x<best_box[0]:
                best_box[0]=x
            if y<best_box[1]:
                best_box[1]=y
            if x+w>best_box[2]:
                best_box[2]=x+w
            if y+h>best_box[3]:
                best_box[3]=y+h
    x1,y1,x2,y2 = best_box
    #add padding but don't exceed edge of image then crop and save
    x1 = max(x1-pad, 0)
    y1 = max(y1-pad, 0)
    x2 = min(x2+pad, im.shape[1])
    y2 = min(y2+pad, im.shape[0])
    rgb = cv2.imread(rgbpath)
    cropped = rgb[y1:y2, x1:x2]
    cv2.imwrite(croppath, cropped)