锐化图像中每个 object 的边框像素

Question

I have an image with some random components... border of each component has some blurred pixels ie...see screenshot

So using OpenCv and python, i want this image to be really sharp...ie no blurry pixels on the edge... as shown in this image below

Here is the full image for your refrence.....

Answer 1

To a certain tolerance for errors this behaviour can be achived by using k-means algorithm. This algorithm can be used to build clusters of pixels having similar color. The OpenCV k-means implementation provides a lot of parameters to tweak with to get the desired results. You may use the following code snippet to start with:

import cv2
import numpy as np

img = cv2.imread("/path/to/img.png")
Z = img.reshape((-1, 3))
Z = np.float32(Z)
criteria = (cv2.TERM_CRITERIA_EPS + cv2.TERM_CRITERIA_MAX_ITER, 10, 1.0)
K = 2
ret, label, center = cv2.kmeans(Z, K, None, criteria, 10, cv2.KMEANS_RANDOM_CENTERS)

center = np.uint8(center)
res = center[label.flatten()]
res2 = res.reshape((img.shape))

cv2.imwrite("./output.png", res2)

The above code will generate the following results for the sample image:

But this code may not work very will with the larger image(even after passing K=5), this is due to the fact that, we are depending upon the K-means to pick the seeds by random sampling. We have an option of passing in the seed colors to look for which will be BGR values for yellow, light greem, dark green, brown and blue in your code. After supplying these BGR values as seed, you can get better results.