基于相邻格间相似度的边缘检测算法

共计 1695 个字符，预计需要花费 5 分钟才能阅读完成。

提醒：本文最后更新于 2024-08-30 15:08，文中所关联的信息可能已发生改变，请知悉！

相似度分布图

省略 100 以后的数据，得到的放大图

效果示例

代码示例

# utf-8
# python3.8

import cv2 as cv
import numpy as np

def get_unit_similarity(a_4, b_4, k):
    a_4 = sorted(a_4)
    b_4 = sorted(b_4)
    result = 0
    for i in range(4):
        result = result + (abs(int(a_4[i]) - int(b_4[i]))) ** k
    result = result ** (1/k)
    return int(result)

def get_channels_similarity(channel, k):
    row, col = channel.shape
    result = np.zeros((row - 1, col - 1))
    for i in range(row - 2):
        for j in range(col - 2):
            array = [channel[i][j], channel[i][j + 1], channel[i + 1][j + 1], channel[i + 1][j]]
            r_array = [channel[i][j + 1], channel[i + 1][j + 1], channel[i][j + 2], channel[i + 1][j + 2]]
            d_array = [channel[i + 1][j + 1], channel[i + 1][j], channel[i + 2][j + 1], channel[i + 2][j]]
            r_similarity = get_unit_similarity(array, r_array, k)
            d_similarity = get_unit_similarity(array, d_array, k)
            result[i][j] = (r_similarity + d_similarity) / 2
    return result

def get_img_similarity(img, k):
    p = cv.imread(img, 1)
    b, g, r = cv.split(p)
    b_similarity = get_channels_similarity(b, k)
    g_similarity = get_channels_similarity(g, k)
    r_similarity = get_channels_similarity(r, k)
    cnt = np.zeros(100)
    for i in range(len(b_similarity)):
        if b_similarity[i] < 100:
            cnt[b_similarity[i]] += 1
    for i in range(len(g_similarity)):
        if g_similarity[i] < 100:
            cnt[g_similarity[i]] += 1
    for i in range(len(r_similarity)):
        if r_similarity[i] < 100:
            cnt[r_similarity[i]] += 1

def mark(channel, flag):
    row, col = flag.shape
    for i in range(row):
        for j in range(col):
            if flag[i][j] > 30:
                channel[i][j] = 0
    return channel

def mark_similarity(img, k):
    p = cv.imread(img, 1)
    b, g, r = cv.split(p)
    b_similarity = get_channels_similarity(b, k)
    b = mark(b, b_similarity)
    g_similarity = get_channels_similarity(g, k)
    g = mark(g, g_similarity)
    r_similarity = get_channels_similarity(r, k)
    r = mark(r, r_similarity)
    p = cv.merge((b, g, r))
    cv.imwrite('img_mark.png', p)

if __name__ == '__main__':
    img_addr = 'img.png'
    mark_similarity(img_addr, 1)