Crop rectangle in OpenCV Python

Question

I have an image like this

and I want to crop each book from the shelf. I started it with this code.

thresh = cv2.adaptiveThreshold(blur, 255, 1, 1, 11, 2)

cv2.imshow("Gray", gray)
cv2.waitKey(0)

cv2.imshow("Blurred", blur)
cv2.waitKey(0)

# detect edges in the image
edged = cv2.Canny(img, 10, 250)
cv2.imshow("Edged", edged)
cv2.waitKey(0)

# construct and apply a closing kernel to 'close' gaps between 'white'
# pixels
kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (1, 6))
closed = cv2.morphologyEx(edged, cv2.MORPH_CLOSE, kernel)
cv2.imshow("Closed", closed)
cv2.waitKey(0)

# loop over the contours
for contour in contours:

    # peri = cv2.arcLength(contour, True)
    # approx = cv2.approxPolyDP(contour, 0.02 * peri, True)
    # r = cv2.boundingRect(contour)
    if len(contour) >= 4:
        index += 1
        x, y, w, h = cv2.boundingRect(contour)
        roi = img[y:y+h, x:x+w]
        # cv2.imwrite("a/" + "book - " + str(index) + '.jpg', roi)
        draw_contour = cv2.drawContours(img, [contour], -1, (255, 140, 240), 2)
        total += 1

print contour

cv2.imshow("Drawed Contour", img)
cv2.waitKey(0)

I created a bounding box in each of the books from the shelf, but unfortunately this gives me the output. I want only to draw a bounding box in the side/corner of the books and then crop it from the bounding box.

I would use hough transform to detect rectangles and then filter by orientation and size. I dont think a simple edge detector will work in your case. — Mario Dekena
– Mario Dekena, Commented Feb 1, 2017 at 10:06

Nilay · Accepted Answer · 2017-02-01 11:59:09Z

5

I don't think you can explicitly identify only books with this code but one quick improvement you can do in code is to draw contours which have area greater than some value. Following code snippet

 if len(contour) >= 4:
    index += 1
    x, y, w, h = cv2.boundingRect(contour)
    roi = img[y:y+h, x:x+w]
    # cv2.imwrite("a/" + "book - " + str(index) + '.jpg', roi)
    if cv2.contourArea(contour) > 200:
        draw_contour = cv2.drawContours(img, [contour], -1, (255, 140, 240), 2)
        total += 1

edited Feb 1, 2017 at 11:59

answered Feb 1, 2017 at 9:56

Nilay

1245 bronze badges

Sign up to request clarification or add additional context in comments.

3 Comments

rbo13 Over a year ago

I will apply your suggestion. Thanks for the heads up

rbo13 Over a year ago

Thanks @Nilay its working. But I did some adjustments for larger book spines, like for example a dictionary. Again, thanks!

Thegerdfather Over a year ago

I'm facing a similar problem. If you still have the code, can you please share the adjustments you made? (Perhaps as an edit to the original question)

Collectives™ on Stack Overflow

Crop rectangle in OpenCV Python

1 Answer 1

3 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

3 Comments

Your Answer

Sign up or log in

Post as a guest

Related