python - Detect text region in image using Opencv

Question

I have an image and want to detect the text regions in it.

I tried TiRG_RAW_20110219 project but the results are not satisfactory. If the input image is http://imgur.com/yCxOvQS,GD38rCa it is producing http://imgur.com/yCxOvQS,GD38rCa#1 as output.

Can anyone suggest some alternative. I wanted this to improve the output of tesseract by sending it only the text region as input.

Answer 1

import cv2


def captch_ex(file_name):
    img = cv2.imread(file_name)

    img_final = cv2.imread(file_name)
    img2gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    ret, mask = cv2.threshold(img2gray, 180, 255, cv2.THRESH_BINARY)
    image_final = cv2.bitwise_and(img2gray, img2gray, mask=mask)
    ret, new_img = cv2.threshold(image_final, 180, 255, cv2.THRESH_BINARY)  # for black text , cv.THRESH_BINARY_INV
    '''
            line  8 to 12  : Remove noisy portion 
    '''
    kernel = cv2.getStructuringElement(cv2.MORPH_CROSS, (3,
                                                         3))  # to manipulate the orientation of dilution , large x means horizonatally dilating  more, large y means vertically dilating more
    dilated = cv2.dilate(new_img, kernel, iterations=9)  # dilate , more the iteration more the dilation

    # for cv2.x.x

    _, contours, hierarchy = cv2.findContours(dilated, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE)  # findContours returns 3 variables for getting contours

    # for cv3.x.x comment above line and uncomment line below

    #image, contours, hierarchy = cv2.findContours(dilated,cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_NONE)


    for contour in contours:
        # get rectangle bounding contour
        [x, y, w, h] = cv2.boundingRect(contour)

        # Don't plot small false positives that aren't text
        if w < 35 and h < 35:
            continue

        # draw rectangle around contour on original image
        cv2.rectangle(img, (x, y), (x + w, y + h), (255, 0, 255), 2)

        '''
        #you can crop image and send to OCR  , false detected will return no text :)
        cropped = img_final[y :y +  h , x : x + w]

        s = file_name + '/crop_' + str(index) + '.jpg' 
        cv2.imwrite(s , cropped)
        index = index + 1

        '''
    # write original image with added contours to disk
    cv2.imshow('captcha_result', img)
    cv2.waitKey()


file_name = 'your_image.jpg'
captch_ex(file_name)

Answer 2

Since no one has posted a complete solution, here's an approach. Using the observation that the desired text is in white and that words are structured in a horizontal alignment, we can use color segmentation to extract and OCR the letters.

1. Perform color segmentation. We load the image, convert to HSV format, define lower/upper ranges and perform color segmentation using cv2.inRange to obtain a binary mask

2. Dilate to connect text characters. We create a horizontal shaped kernel using cv2.getStructuringElement then dilate using cv2.dilate to combine individual letters into a single contour

3. Remove non-text contours. We find contours with cv2.findContours and filter using aspect ratio to remove non-text characters. Since the text is in a horizontal orientation, if the contour is determined to be less than a predefined aspect ratio threshold then we remove the non-text contour by filling in the contour with cv2.drawContours

4. Perform OCR. We bitwise-and the dilated image with the initial mask to isolate only text characters and invert the image so that the text is in black with the background in white. Finally, we throw the image into Pytesseract OCR

---

Here's a visualization of each step:

Input image

Mask generated from color segmentation

# Load image, convert to HSV format, define lower/upper ranges, and perform
# color segmentation to create a binary mask
image = cv2.imread('1.jpg')
hsv = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)
lower = np.array([0, 0, 218])
upper = np.array([157, 54, 255])
mask = cv2.inRange(hsv, lower, upper)

Dilated image to connect text-contours and removed non-text contours using aspect ratio filtering

# Create horizontal kernel and dilate to connect text characters
kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (5,3))
dilate = cv2.dilate(mask, kernel, iterations=5)

# Find contours and filter using aspect ratio
# Remove non-text contours by filling in the contour
cnts = cv2.findContours(dilate, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
cnts = cnts[0] if len(cnts) == 2 else cnts[1]
for c in cnts:
    x,y,w,h = cv2.boundingRect(c)
    ar = w / float(h)
    if ar < 5:
        cv2.drawContours(dilate, [c], -1, (0,0,0), -1)

Bitwise-and both masks and invert to get result ready for OCR

# Bitwise dilated image with mask, invert, then OCR
result = 255 - cv2.bitwise_and(dilate, mask)
data = pytesseract.image_to_string(result, lang='eng',config='--psm 6')
print(data)

Result from Pytesseract OCR using --psm 6 configuration setting to assume a uniform block of text. Look here for more configuration options

All women become
like their mothers.
That is their tragedy.
No man does.

That's his.

OSCAR WILDE

Full code

import cv2
import numpy as np
import pytesseract

pytesseract.pytesseract.tesseract_cmd = r"C:\Program Files\Tesseract-OCR\tesseract.exe"

# Load image, convert to HSV format, define lower/upper ranges, and perform
# color segmentation to create a binary mask
image = cv2.imread('1.jpg')
hsv = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)
lower = np.array([0, 0, 218])
upper = np.array([157, 54, 255])
mask = cv2.inRange(hsv, lower, upper)

# Create horizontal kernel and dilate to connect text characters
kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (5,3))
dilate = cv2.dilate(mask, kernel, iterations=5)

# Find contours and filter using aspect ratio
# Remove non-text contours by filling in the contour
cnts = cv2.findContours(dilate, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
cnts = cnts[0] if len(cnts) == 2 else cnts[1]
for c in cnts:
    x,y,w,h = cv2.boundingRect(c)
    ar = w / float(h)
    if ar < 5:
        cv2.drawContours(dilate, [c], -1, (0,0,0), -1)

# Bitwise dilated image with mask, invert, then OCR
result = 255 - cv2.bitwise_and(dilate, mask)
data = pytesseract.image_to_string(result, lang='eng',config='--psm 6')
print(data)

cv2.imshow('mask', mask)
cv2.imshow('dilate', dilate)
cv2.imshow('result', result)
cv2.waitKey()

The HSV lower/upper color range was determined using this HSV color thresholder script

import cv2
import numpy as np

def nothing(x):
    pass

# Load image
image = cv2.imread('1.jpg')

# Create a window
cv2.namedWindow('image')

# Create trackbars for color change
# Hue is from 0-179 for Opencv
cv2.createTrackbar('HMin', 'image', 0, 179, nothing)
cv2.createTrackbar('SMin', 'image', 0, 255, nothing)
cv2.createTrackbar('VMin', 'image', 0, 255, nothing)
cv2.createTrackbar('HMax', 'image', 0, 179, nothing)
cv2.createTrackbar('SMax', 'image', 0, 255, nothing)
cv2.createTrackbar('VMax', 'image', 0, 255, nothing)

# Set default value for Max HSV trackbars
cv2.setTrackbarPos('HMax', 'image', 179)
cv2.setTrackbarPos('SMax', 'image', 255)
cv2.setTrackbarPos('VMax', 'image', 255)

# Initialize HSV min/max values
hMin = sMin = vMin = hMax = sMax = vMax = 0
phMin = psMin = pvMin = phMax = psMax = pvMax = 0

while(1):
    # Get current positions of all trackbars
    hMin = cv2.getTrackbarPos('HMin', 'image')
    sMin = cv2.getTrackbarPos('SMin', 'image')
    vMin = cv2.getTrackbarPos('VMin', 'image')
    hMax = cv2.getTrackbarPos('HMax', 'image')
    sMax = cv2.getTrackbarPos('SMax', 'image')
    vMax = cv2.getTrackbarPos('VMax', 'image')

    # Set minimum and maximum HSV values to display
    lower = np.array([hMin, sMin, vMin])
    upper = np.array([hMax, sMax, vMax])

    # Convert to HSV format and color threshold
    hsv = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)
    mask = cv2.inRange(hsv, lower, upper)
    result = cv2.bitwise_and(image, image, mask=mask)

    # Print if there is a change in HSV value
    if((phMin != hMin) | (psMin != sMin) | (pvMin != vMin) | (phMax != hMax) | (psMax != sMax) | (pvMax != vMax) ):
        print("(hMin = %d , sMin = %d, vMin = %d), (hMax = %d , sMax = %d, vMax = %d)" % (hMin , sMin , vMin, hMax, sMax , vMax))
        phMin = hMin
        psMin = sMin
        pvMin = vMin
        phMax = hMax
        psMax = sMax
        pvMax = vMax

    # Display result image
    cv2.imshow('image', result)
    if cv2.waitKey(10) & 0xFF == ord('q'):
        break

cv2.destroyAllWindows()

Answer 3

If you don't mind getting your hands dirty you could try and grow those text regions into one bigger rectangular region, which you feed to tesseract all at once.

I'd also suggest trying to threshold the image several times and feeding each of those to tesseract separately to see if that helps at all. You can compare the output to dictionary words to automatically determine if a particular OCR result is good or not.