驾驶员疲劳检测中的彩色图像增强算法  

An algorithm of color image enhancement for driver fatigue detection

作　　者：张万枝[1] 王增才[1] 李云霞[1] 王保平[1] 

机构地区：[1]山东大学机械工程学院,山东济南250061

出　　处：《光电子．激光》2012年第8期1545-1549,共5页Journal of Optoelectronics·Laser

基　　金：国家自然科学基金(51174126)资助项目

摘　　要：为解决因驾驶室光照不均引起的图像质量下降问题,提出一种新的基于Retinex的驾驶员视频图像增强算法。首先将图像从RGB空间转换到YCbCr空间,保持Cb和Cr分量不变,对Y分量进行高、低频分解,低频分量部分包含图像全部光照信息;利用多尺度Retinex(MSR)算法对低频部分进行光照补偿,然后对Y分量进行拉普拉斯锐化和高斯平滑滤波,增强图像细节和消除噪声干扰;最后将处理后的Y分量与原来的Cb、Cr分量进行逆变换生成新图像。实验表明,与传统算法相比,本文算法能对驾驶员视频图像起到增强效果,同时也能很好地增强图像细节信息,有利于系统对驾驶员状态进行疲劳检测。To solve the reduction of image quality caused by uneven illumination of cab, a method which can enhance driver video image quality has been proposed. Firstly, an image is transformed from RGB space into YCbCr space,the Cb and Cr elements remain unchanged and the Y element is divided into high-frequency part and low-frequency part which contains the whole illumination information. The lowfrequency part is compensated by using Retinex theory and then by using Laplace sharpness and Gaussi- an smoothing on the Y elements to enhance image details and eliminate noise interference. After the alove treatment, the Y element and the original Cb and Cr elements form a new image. Results show that compared with other traditional algorithms, the method can enhance images and details, which is benefit for detecting fatigue state.

关 键 词：彩色图像增强 光照补偿 颜色恒常性 RETINEX理论 多尺度 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]