第3章 Python 数字图像处理(DIP) - 灰度变换与空间滤波7 - 直方图处理 - 直方图、归一化直方图

直方图处理

令$r_k,k=0,1,2,\dots ,L-1$表于一幅$L$级灰度数字图像$f(x,y)$的灰度。$f$的非归一化直方图定义为：

$$h(r_k)=n_k,k=0,1,2,\dots ,L-1\text{\hspace{1em}(3.6)}$$

$n_k是f$中灰度为$r_k$的像素的数量，并且细分的灰度级称为直方图容器。

归一化直方图

$$p(r_k)=\frac{h(r_k)}{MN}=\frac{n_k}{MN}\text{\hspace{1em}(3.7)}$$

def my_hist(img, bins=256, normalized=False):
    """
    create a hist of uint8 image value range[0, 255]
    param: input img: grayscale image range[0, 255]
    param: input bins: bins for the image, range[0, 255]
    return hist and bins for the image, hist -> counts for all the values
    """
    # initializ a list for values and counts
    list1 = list([x, y] for x in np.arange(bins) for y in np.arange(bins) if y == 0)
    data = img.flatten()
    for i in range(img.size):
        list1[data[i]][1] += 1  
    dst = np.array(list1)
    bins, hist= dst[:, 0], dst[:, 1]
    if normalized:
        hist = hist / img.size  #nomalized hist
    return hist, bins

# 直方图
img_1st = cv2.imread('DIP_Figures/DIP3E_Original_Images_CH03/Fig0316(1)(top_left).tif', 0)
img_2nd = cv2.imread('DIP_Figures/DIP3E_Original_Images_CH03/Fig0316(2)(2nd_from_top).tif', 0)
img_3rd = cv2.imread('DIP_Figures/DIP3E_Original_Images_CH03/Fig0316(3)(third_from_top).tif', 0)
img_4th = cv2.imread('DIP_Figures/DIP3E_Original_Images_CH03/Fig0316(4)(bottom_left).tif', 0)

img_list =['img_1st', 'img_2nd', 'img_3rd', 'img_4th']

fig = plt.figure(figsize=(20, 15))
for i in range(len(img_list)):
    ax = fig.add_subplot(3, 4, i+1)
    ax.imshow(eval(img_list[i]), cmap='gray', vmin=0, vmax=255)
    ax1 = fig.add_subplot(3, 4, i+5)
    hist, bins = my_hist(eval(img_list[i]), bins=256)
    ax1.bar(bins, hist), ax1.set_title('My Hist')
    ax2 = fig.add_subplot(3, 4, i+9)
    hist, bins = np.histogram(eval(img_list[i]), bins=256, range=[0, 256])
    ax2.bar(bins[:-1], hist), ax2.set_title('Numpy Hist')  # numpy 的返回bins 比hist多一点数，需要去掉最后一个
#     ax2.hist(eval(img_list[i]).flatten(), bins=256, range=[0, 255]), ax2.set_title('Matplotlib Hist')
plt.tight_layout()
plt.show()