【图像处理】直方图均衡化基于python实现

读图像、显示图像采用opencv接口，算法主体用python编写
算法步骤在代码中详细体现。
代码如下：

#!/usr/bin/python3
# -*- coding: utf-8 -*-

'直方图均衡化函数，0::28 2019/06/20'

__author__ = 'mint'

import numpy as np
import cv2 as cv

#创建空列表(数组)
def createEmptyList(size):
    newList = []
    for eachNum in range(0, size):
        newList.append(0)
    return newList

#创建空图像矩阵
def createEmptyImage(rows, cols,type):
    img = np.zeros((rows, cols), dtype=type)
    return img

#直方图均衡化
def histequaLize(src, dst):

    #step 1 校验参数#
    assert(type(src)==np.ndarray)
    assert(src.dtype == np.uint8)
    assert (type(dst) == np.ndarray)
    assert (dst.dtype == np.uint8)

    # step 2 直方图统计#
    hist = createEmptyList(256)
    rows, cols = src.shape
    for r in range(rows):
        for c in range(cols):
            hist[src[r,c]] += 1

    # step 3 直方图归一化#
    for each in range(256):
        hist[each] /= rows*cols
        
    # step 4 直方图累加#
    for each in range(1,256):
        hist[each] = hist[each-1] + hist[each]

    # step 4 均衡#
    for each in range(256):
        hist[each] = (np.uint8)(255*hist[each] + 0.5)

    for r in range(rows):
        for c in range(cols):
            dst[r,c] = hist[src[r,c]]

def histMain():
    img = cv.imread('lena.jpg')
    gray = cv.cvtColor(img, cv.COLOR_BGR2GRAY)
    dst = createEmptyImage(img.shape[0], img.shape[1], np.uint8)
    histequaLize(gray, dst)
    cv.imshow('histEnhance.jpg', dst)
    cv.waitKey(0)

if __name__ == '__main__':
    histMain()

输出如下：