伽马变换就是用来图像增强,其提升了暗部细节,简单来说就是通过非线性变换,让图像从暴光强度的线性响应变得更接近人眼感受的响应,即将漂白(相机曝光)或过暗(曝光不足)的图片,进行矫正。
伽马变换的基本形式如下:
大于1时,对图像的灰度分布直方图具有拉伸作用(使灰度向高灰度值延展),而小于1时,对图像的灰度分布直方图具有收缩作用(是使灰度向低灰度值方向靠拢)。
#分道计算每个通道的直方图 img0 = cv2.imread('12.jpg') hist_b = cv2.calcHist([img0],[0],None,[256],[0,256]) hist_g = cv2.calcHist([img0],[1],None,[256],[0,256]) hist_r = cv2.calcHist([img0],[2],None,[256],[0,256]) def gamma_trans(img,gamma): #具体做法先归一化到1,然后gamma作为指数值求出新的像素值再还原 gamma_table = [np.power(x/255.0,gamma)*255.0 for x in range(256)] gamma_table = np.round(np.array(gamma_table)).astype(np.uint8) #实现映射用的是Opencv的查表函数 return cv2.LUT(img0,gamma_table) img0_corrted = gamma_trans(img0, 0.5) cv2.imshow('img0',img0) cv2.imshow('gamma_image',img0_corrted) cv2.imwrite('gamma_image.png',img0_corrted) #分通道计算Gamma校正后的直方图 hist_b_c =cv2.calcHist([img0_corrted],[0],None,[256],[0,256]) hist_g_c =cv2.calcHist([img0_corrted],[1],None,[256],[0,256]) hist_r_c =cv2.calcHist([img0_corrted],[2],None,[256],[0,256]) fig = plt.figure('gamma') pix_hists = [[hist_b, hist_g, hist_r], [hist_b_c, hist_g_c, hist_r_c]] pix_vals = range(256) for sub_plt, pix_hist in zip([121, 122], pix_hists): ax = fig.add_subplot(sub_plt, projection='3d') for c, z, channel_hist in zip(['b', 'g', 'r'], [20, 10, 0], pix_hist): cs = [c] * 256 ax.bar(pix_vals, channel_hist, zs=z, zdir='y', color=cs, alpha=0.618, edgecolor='none', lw=0) ax.set_xlabel('Pixel Values') ax.set_xlim([0, 256]) ax.set_ylabel('Count') ax.set_zlabel('Channels') plt.show() cv2.waitKey()