基于改进高低帽变换的高速铁路弓网红外图像增强技术

High-speed Train Pantograph-catenary Infrared Image Enhancement Based on Improved Top-hat and Bottom-hat Transformation

针对现有高速铁路接触网运行状态检测系统中弓网红外图像对比度低、信噪比低、边缘细节不突出而影响图像视觉效果,降低检测精度的问题,本文提出了多尺度高帽低帽变换的弓网红外图像增强优化算法。算法为保留图像细节,通过改进高低帽变换算法提取原始图像的不同尺度亮、暗区域;其次利用最大值获取每一尺度上的亮、暗区域;然后操作其最大值以构建明亮和暗淡的图像区域;最后为了有效增强对比度,引入了基于加权的处理算法。实验结果表明:该方法既能提高图像对比度和信噪比,又能有效突出弓网红外图像线索边缘进而保留细节信息,为后续弓网状态监测奠定了必要的基础。

Aiming at the problem that it is difficult to enhance the infrared image of pantograph catenary with low contrast, low signal-to-noise ratio and outstanding edge details, a new pantograph catenary infrared image enhancement optimization algorithm based on the morphological top-hat transform and bottom-hat transform of the multi-scale tophat and bottom-hat transformation is proposed. In order to preserve the details of the image, the different bright and dark areas of the original image are extracted by the algorithm. then the maximum value extract from the multi-scale bright and dim image regions is manipulated to establish the bright and dark areas of the image, which will further go through an weighted processing, thereby realizing image enhancement. Finally, in order to effectively enhance contrast, a weighted based processing algorithm is introduced. Experimental results show that the method can not only improve the image contrast and the signal to noise ratio, but also effectively highlight the edge of the infrared image of the pantograph catenary and retain the details, which lays the necessary foundation for the monitoring of the pantograph catenary.