多个图片亮度和色度归一化处理

在准备数据时，由于相机的原因采集到的图片亮度、色度会出现问题

正常：

进行数据处理（代码看不懂自己一步步研究研究）

• 读取图片

image = cv2.imread(image_path, cv2.IMREAD_COLOR)

• 根据灰度、Gamma归一化亮度

python 

m_gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
Gamma = np.log(128.0 / 255.0) / np.log(cv2.mean(m_gray)[0] / 255.0)
lookUpTable = np.empty((1, 256), np.uint8)
for i in range(256):
    lookUpTable[0, i] = np.clip(pow(i / 255.0, Gamma) * 255.0, 0, 255)
image = cv2.LUT(image, lookUpTable)

c++ 

cv::Mat m_gray;
cv::cvtColor(image, m_gray, cv::COLOR_BGR2GRAY);
float Gamma = log(128.0 / 255.0) / log(cv::mean(m_gray)[0] / 255.0);
cv::Mat lookUpTable = cv::Mat::zeros(cv::Size(1, 256), CV_8UC1);
for (int i = 0; i < 256; i++)
	lookUpTable.at<uchar>(0, i) = pow(i / 255.0, Gamma) * 255.0;
cv::LUT(image, lookUpTable, image);

 

• 转到LAB，统一色度

python 

m_lab = cv2.cvtColor(image, cv2.COLOR_BGR2LAB)
split_lab = cv2.split(m_lab)
cv2.convertScaleAbs(split_lab[1], split_lab[1], 128.0 / cv2.mean(split_lab[1])[0], 0.0)
cv2.convertScaleAbs(split_lab[2], split_lab[2], 128.0 / cv2.mean(split_lab[2])[0], 0.0)
result_lab = cv2.merge(split_lab)
image = cv2.cvtColor(result_lab, cv2.COLOR_LAB2BGR)

c++ 

cv::Mat m_lab;
cv::cvtColor(image, m_lab, cv::COLOR_BGR2Lab);
std::vector<cv::Mat> split_lab;
cv::split(m_lab, split_lab);
cv::convertScaleAbs(split_lab[1], split_lab[1], 128.0 / cv::mean(split_lab[1])[0], 0.0);
cv::convertScaleAbs(split_lab[2], split_lab[2], 128.0 / cv::mean(split_lab[2])[0], 0.0);
cv::merge(split_lab, result);
cv::cvtColor(result, result, cv::COLOR_Lab2BGR);

 

修正前后的效果对比图

• 偏暗

• 偏绿

• 偏蓝

OK，完美，觉得可以记得点赞