Blob detection in Python?

Question

I'm trying to detect a blob from the following image. I used the skimage and used the 3 different methods explained in the manual, but it's not able to detect the grey blob. Here is the original image:

So I tried the following code:

from math import sqrt
import cv2
from skimage.feature import blob_dog, blob_log, blob_doh
from skimage.color import rgb2gray
import matplotlib.pyplot as plt

image = cv2.imread("blob800_cropped.png")
image_gray = rgb2gray(image)

blobs_log = blob_log(image_gray, max_sigma=30, num_sigma=10, threshold=.05)

# Compute radii in the 3rd column.
blobs_log[:, 2] = blobs_log[:, 2] * sqrt(2)

blobs_dog = blob_dog(image_gray, max_sigma=30, threshold=.05)
blobs_dog[:, 2] = blobs_dog[:, 2] * sqrt(2)

blobs_doh = blob_doh(image_gray, max_sigma=30, threshold=.01)

blobs_list = [blobs_log, blobs_dog, blobs_doh]
colors = ['yellow', 'lime', 'red']
titles = ['Laplacian of Gaussian', 'Difference of Gaussian',
          'Determinant of Hessian']
sequence = zip(blobs_list, colors, titles)

fig, axes = plt.subplots(1, 3, figsize=(9, 3), sharex=True, sharey=True)
ax = axes.ravel()

for idx, (blobs, color, title) in enumerate(sequence):
    ax[idx].set_title(title)
    ax[idx].imshow(image)
    for blob in blobs:
        y, x, r = blob
        c = plt.Circle((x, y), r, color=color, linewidth=2, fill=False)
        ax[idx].add_patch(c)
    ax[idx].set_axis_off()

plt.tight_layout()
plt.show()

However, the blob which I'm looking for is not detected:

This is the output I'm expecting:

fmw42 · Accepted Answer · 2020-04-26 17:51:28Z

8

Here is how I would do that in Python/OpenCV.

Read input
Convert to gray
Apply extreme adaptive thresholding
Apply morphology open and close to remove small regions
Get contours and save the largest
Draw the largest contour on the input
Save the results

Input:

import cv2
import numpy as np

# read image
img = cv2.imread("doco3.jpg")

# convert img to grayscale
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

# do adaptive threshold on gray image
thresh = cv2.adaptiveThreshold(gray, 255, cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY, 101, 3)

# apply morphology open then close
kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (5,5))
blob = cv2.morphologyEx(thresh, cv2.MORPH_OPEN, kernel)
kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (9,9))
blob = cv2.morphologyEx(blob, cv2.MORPH_CLOSE, kernel)

# invert blob
blob = (255 - blob)

# Get contours
cnts = cv2.findContours(blob, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
cnts = cnts[0] if len(cnts) == 2 else cnts[1]
big_contour = max(cnts, key=cv2.contourArea)

# test blob size
blob_area_thresh = 1000
blob_area = cv2.contourArea(big_contour)
if blob_area < blob_area_thresh:
    print("Blob Is Too Small")

# draw contour
result = img.copy()
cv2.drawContours(result, [big_contour], -1, (0,0,255), 1)

# write results to disk
cv2.imwrite("doco3_threshold.jpg", thresh)
cv2.imwrite("doco3_blob.jpg", blob)
cv2.imwrite("doco3_contour.jpg", result)

# display it
cv2.imshow("IMAGE", img)
cv2.imshow("THRESHOLD", thresh)
cv2.imshow("BLOB", blob)
cv2.imshow("RESULT", result)
cv2.waitKey(0)

Threshold image:

Morphology cleaned image of blob:

Resulting contour on input:

edited Apr 26, 2020 at 17:51

answered Apr 25, 2020 at 21:42

fmw42

54.1k10 gold badges80 silver badges95 bronze badges

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

3 Comments

Mohammad Over a year ago

this is great, thanks. One quick question, how should I set a limit size for the blob? For example if the max blob size is smaller than some value, I want to be able to return "No Blob". Thanks for your help @fmw42

fmw42 Over a year ago

Add after big_contour=... blob_area = cv2.contourArea(big_contour) and then if blob_area < blob_area_thresh: print("Too Small") where blob_area_thresh is your area limitation. See my change in my answer

Mohammad Over a year ago

best response ever! Thanks

Collectives™ on Stack Overflow

Blob detection in Python?

1 Answer 1

3 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

3 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related