自适应HSV空间Retinex煤矿监控图像增强算法

针对煤矿井下的监控图像由于粉尘、煤尘、低照度或点光源等光照的影响,而整体阴暗模糊,对比度低,背景噪声强,视觉效果不理想的问题,提出一种改进的HSV(Hue,Saturation,Value)空间的颜色可恢复的多尺度Retinex(Muhi-Scale Ret-inex,MSR)的图像增强算法.该算法先将图像从RGB空间转换到HSV空间,以确保后续图像增强处理不会影响图像的色彩效果;然后利用提出的自适应的高斯核函数分离亮度分量V的照度分量和反射分量,再利用自适应的增益系数对反射分量进行增强处理,获得不受光照影响的、增强的反射图像;最后将反射图像逆变换回RGB空间,再利用优化的颜色恢复函数对增强的图像进行颜色修正,从而改善图像的全局视觉效果和局部对比度.实验表明该方法能有效提高煤矿井下监控图像的对比度和亮度,抑制背景噪声,从而大大改善煤矿井下监控图像的视觉效果,且很好地实现了参数自适应,减少了人为因素对结果的影响.

Adaptive Retinex algorithm at HSV space for coal mine monitoring image enhancement

To solve the problem that the coal mine monitoring images are easily influenced by illumination and coal dust and cause the low contrast and the bad visual effect,an improved image enhancement method based on the Multi Scale Retinex(MSR) algorithm at HSV space was proposed.Firstly,the image is converted from the RGB space to the HSV space to ensure that the subsequent processing for image enhancement will not affect the color.Then,according to the Retinex theory,the illumination and reflection of the V component are separated by the adaptive gauss kernel function.Next,the reflection component of V is enhanced by the adaptive gain coefficient to obtain an enhanced reflection image which is not affected by illumination.Finally,the enhanced reflection image is converted to RGB space,and an optimized color correction function is used to adjust the image color,so that the overall visual effect and local contrast of image is improved.The experiments show that the contrast and brightness of the coal mine monitoring images are effectively increased and background noise also is suppressed by this method,thus the visual effect of the coal mine monitoring images are greatly improved.So as to,a better adaptive parameters are achieved so that the impact of human factors on the result may be reduced.