如何在 Python 中使用 cv2 获取指纹？

Question

What would you recommend me in order to get a better fingerprints extraction? I doesn't look so well. Thank you. Here's my code:

import cv2
import numpy as np

img = cv2.imread("huella.jpg")
img = cv2.resize(img, None, fx=0.7, fy=1.0, interpolation=cv2.INTER_AREA)
w, h = img.shape[:2]
fp = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

sharp = np.array([[-1, -1, -1, -1, -1], [-1, 2, 2, 2, -1], [-1, 2, 8, 2, -1], [-1, 2, 2, 2, -1], [-1, -1, -1, -1, -1]]) / 8
fp = cv2.filter2D(fp, -1, sharp)

fp = cv2.Canny(fp, 45, 45)

cv2.imshow("Original", img)
cv2.imshow("Huella", fp)

cv2.waitKey(0)
cv2.destroyAllWindows()

Images

Answer 1

You need to use morphological operation.

First. Try to use cv2.dilate() and then cv2.erode() . This should remove all small and far object.

You can see full documentation here.

Morphological Transformations

Eroding and Dilating

New Edit:

The image will lost the information upon dilate and erode, so here is a script to remove small connected component. You should change the minSize as your need.

import cv2
import numpy as np


def remove_small_pixel(img, minSize=50):
    nlabels, labels, stats, centroids = cv2.connectedComponentsWithStats(img, None, None, None, 8, cv2.CV_32S)
    sizes = stats[1:, -1]  # get CC_STAT_AREA component
    img2 = np.zeros(labels.shape, np.uint8)

    for i in range(0, nlabels - 1):
        if sizes[i] >= minSize:  # filter small dotted regions
            img2[labels == i + 1] = 255

    return img2

Note: This script only available for grayscale image.