一种基于K-means改进聚类的图像增强算法

Image enhancement algorithm based on improved K-means clustering

在低光照环境下,CMOS成像器件无法拍摄出清晰的图像。为了提升低照度条件成像器件输出图像的质量,根据低照度图像的特点,提出一种基于K-means聚类的图像增强算法。通过改进的K-means算法将图像分块,并根据每一块图像的信息量分别进行直方图均衡。该方法与CMOS成像器件进行实验,可以在保留约98.6%图像细节(信息熵)的前提下,将图像的对比度提升至原图像的17倍,平均梯度提升至原图像的4倍。

In the case of low-light conditions,the quality of the output image is not satisfactory provided by complementary metal oxide semiconductor（CMOS）imaging device.An image enhancement algorithm based on K-means clustering was proposed according to the characteristics of low illumination image,and the analysiscomparison of various algorithms.The clustering center k can be automatically determined by the histogram feature of original image,and the sub images can be enhanced by histogram equalization according to the information content of each sub image after pixel sets are divided into several non-overlapping subsets by clustering.The experiment was carried out in CMOS imaging device using this method,the results shows that this algorithm can enhance the contrast of the image to 17 times of the original image,and the average gradient can be increased to 4times under the condition of preserving the details of the image（information entropy）to about 98.6%.