python图像处理：直方图均衡化

理论参考链接：【图像处理算法】直方图均衡化

算法实现

# 对单通道的处理函数
def channel_equalize(img):
    od = img.reshape(-1)
    # 对像素计数
    tmp = {}
    for num in od:
        tmp[num] = tmp.get(num, 0)+1
    tmp_list = list(tmp.items())
    tmp_list = sorted(tmp_list, key=lambda x:x[0], reverse=False)
    n = {}
    for data in tmp_list:
        n[data[0]] = data[1]

    # 像素分布频率
    for key, _ in n.items():
        n[key] /= len(od)

    # 像素累计分布频率
    pre = 0
    for key, _ in n.items():
        pre += n[key]
        n[key] = pre

    # 均衡化
    for key, _ in n.items():
        n[key] = int(n[key]*(len(n)-1)+1)
    for i in range(img.shape[0]):
        for j in range(img.shape[1]):
            img[i, j] = n[img[i, j]]
    return img

原图三通道的直方图

均衡化后三通道的直方图

原图和均衡化图对比