Search code examples
pythonopencvocrdigits

OpenCV: Recognizing discontinuous digits - multiple contour boxes problem

I am following this tutorial to recognize six digits from the following image enter image description here

The threshold seems (to me) to be very good enter image description here However, when I reach the contour definition, digits 7, 1, 0 (and possibly more) are always split in two or more boxes. enter image description here By definition, a contour is a boundary of a continuous entity, which means that these digits separated by a small ligature cannot be classified as such. What to do in this case? My first instinct is to try and merge these small boxes? I have already tried to play around with the height and width of the contour with no success. The code is written below.

# https://pyimagesearch.com/2017/02/13/recognizing-digits-with-opencv-and-python/
# import the necessary packages
from imutils.perspective import four_point_transform
from imutils import contours
import imutils
import cv2
# define the dictionary of digit segments so we can identify
# each digit on the thermostat
DIGITS_LOOKUP = {
    (1, 1, 1, 0, 1, 1, 1): 0,
    (0, 0, 1, 0, 0, 1, 0): 1,
    (1, 0, 1, 1, 1, 1, 0): 2,
    (1, 0, 1, 1, 0, 1, 1): 3,
    (0, 1, 1, 1, 0, 1, 0): 4,
    (1, 1, 0, 1, 0, 1, 1): 5,
    (1, 1, 0, 1, 1, 1, 1): 6,
    (1, 0, 1, 0, 0, 1, 0): 7,
    (1, 1, 1, 1, 1, 1, 1): 8,
    (1, 1, 1, 1, 0, 1, 1): 9
}

# load the example image
image = cv2.imread('DSC_01922.JPG', 1)
# pre-process the image by resizing it, converting it to
# graycale, blurring it, and computing an edge map
# image = imutils.resize(image, height=500)
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
blurred = cv2.GaussianBlur(gray, (5, 5), 0)
# edged = cv2.Canny(blurred, 50, 200, 255)

# threshold the warped image, then apply a series of morphological
# operations to cleanup the thresholded image
thresh = cv2.threshold(blurred, 0, 255, cv2.THRESH_BINARY_INV | cv2.THRESH_OTSU)[1]
kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (1, 5))
thresh = cv2.morphologyEx(thresh, cv2.MORPH_OPEN, kernel)

# cv2.imshow('thresh', thresh)
# cv2.waitKey(0)
# cv2.destroyAllWindows()

# find contours in the thresholded image, then initialize the
# digit contours lists
cnts = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
cnts = imutils.grab_contours(cnts)
digitCnts = []
# loop over the digit area candidates
for c in cnts:
    # compute the bounding box of the contour
    (x, y, w, h) = cv2.boundingRect(c)
    # if the contour is sufficiently large, it must be a digit
    if (h >= 90 and h <= 300):
            digitCnts.append(c)
            cv2.rectangle(image, (x, y), (x + w, y + h), (0, 0, 255), 2)

cv2.imshow('image', image)
cv2.waitKey(0)
cv2.destroyAllWindows()

Update 1

Using MORPH_CLOSE instead of OPEN and enlarging the kernel as suggested by @Croolman improves the results as can be seen below

kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (1, 7))
thresh = cv2.morphologyEx(thresh, cv2.MORPH_CLOSE, kernel)

enter image description here

Note that I am doing this as a hobby and I am not familiar with/doing research on existent tools of OpenCV/python. Thank you in advance.

Update 2

This solution works.

kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (1, 15))
thresh = cv2.morphologyEx(thresh, cv2.MORPH_CLOSE, kernel)

threshold enter image description here contour enter image description here

Solution

  • This is the complete answer. It required tweeking the kernel values + using morph_close

    # https://pyimagesearch.com/2017/02/13/recognizing-digits-with-opencv-and-python/
# import the necessary packages
from imutils.perspective import four_point_transform
from imutils import contours
import imutils
import cv2
# define the dictionary of digit segments so we can identify
# each digit on the thermostat
DIGITS_LOOKUP = {
    (1, 1, 1, 0, 1, 1, 1): 0,
    (0, 0, 1, 0, 0, 1, 0): 1,
    (1, 0, 1, 1, 1, 1, 0): 2,
    (1, 0, 1, 1, 0, 1, 1): 3,
    (0, 1, 1, 1, 0, 1, 0): 4,
    (1, 1, 0, 1, 0, 1, 1): 5,
    (1, 1, 0, 1, 1, 1, 1): 6,
    (1, 0, 1, 0, 0, 1, 0): 7,
    (1, 1, 1, 1, 1, 1, 1): 8,
    (1, 1, 1, 1, 0, 1, 1): 9
}

# load the example image
image = cv2.imread('DSC_01922.JPG', 1)
# pre-process the image by resizing it, converting it to
# graycale, blurring it, and computing an edge map
# image = imutils.resize(image, height=500)
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
blurred = cv2.GaussianBlur(gray, (5, 5), 0)
# edged = cv2.Canny(blurred, 50, 200, 255)

# threshold the warped image, then apply a series of morphological
# operations to cleanup the thresholded image
thresh = cv2.threshold(blurred, 0, 255, cv2.THRESH_BINARY_INV | cv2.THRESH_OTSU)[1]
kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (1, 15))
thresh = cv2.morphologyEx(thresh, cv2.MORPH_CLOSE, kernel)

cv2.imshow('thresh', thresh)
cv2.waitKey(0)
cv2.destroyAllWindows()

# find contours in the thresholded image, then initialize the
# digit contours lists
cnts = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
cnts = imutils.grab_contours(cnts)
digitCnts = []
# loop over the digit area candidates
for c in cnts:
    # compute the bounding box of the contour
    (x, y, w, h) = cv2.boundingRect(c)
    # if the contour is sufficiently large, it must be a digit
    if (h >= 90 and h <= 300):
            digitCnts.append(c)
            cv2.rectangle(image, (x, y), (x + w, y + h), (0, 0, 255), 2)
            
cv2.imshow('image', image)
cv2.waitKey(0)
cv2.destroyAllWindows()