摘要
目前通过图像传感器采集到的图像颜色与物体本身真实的颜色存在偏差,这为人们通过图像获取相关信息带来不便。针对图像色偏问题,提出了一种色彩还原方法。该方法首先确认当前光照条件下的理想RGB值,然后通过迭代方法求出色彩矫正矩阵集合,最后通过计算每个色彩还原矩阵的平均方差选出最优颜色矫正矩阵,通过该矩阵对原始图像进行矫正。结果表明,该方法能够有效的矫正图像的颜色偏差,使图像传感器采集到的图像更接近物体本身颜色。
Currently, there is color deviation between the image color collected by the image sensor and the real color of the object itself. This makes people to get information through the image inconvenience. Aiming at the problem of image color shift, a color restoration method is proposed. This method first confirms the ideal RGB value under the current illu- mination condition, then obtains the set of color correction matrix by the iterative method, and finally selects the optimal color correction matrix by calculating the average variance of each color reproduction matrix, and the original image is processed by the optimal correction matrix. The results show that the method can effectively correct the image color deviation, so that the image sensor to collect the image closer to the object itself color.
出处
《国外电子测量技术》
2016年第11期24-26,41,共4页
Foreign Electronic Measurement Technology
关键词
色偏
色彩还原
颜色矫正
RGB颜色空间
color cast
color reproduction
color correction
RGB color space
