基于高通滤波和直方图均衡的钢轨裂纹红外图像增强技术

针对钢轨裂纹红外图像采集过程中产生的噪声大、对比度低的问题,以及传统直方图均衡化方法在对红外图像增强时灰度级减少和细节信息丢失等不足,提出频域增强与空域增强相结合的红外图像增强算法,以弥补传统算法在红外图像增强时的不足。在空间域上采用直方图均衡化对红外图像进行处理,提高图像对比度降低图像噪声;在频率域上采用高通滤波对红外图像进行锐化处理,增强图像的边缘和细节信息。仿真结果表明:该算法在对钢轨裂纹红外图像增强时很好地抑制了背景噪声,而且有效地凸显了钢轨裂纹内部微弱的细节纹理,保护图像的细节信息,为更好地提取钢轨裂纹的缺陷信息提供条件。

Infrared Image Enhancement Technology Based on High Pass Filter and Histogram Equalization for Rail Crack Detection

Low contrast and noise of infrared image acquisition process of rail crack in the traditional histogram equalization method may lead to infrared image edge blur and information loss in infrared image enhancement. In this paper,an infrared image enhancement algorithm based on the combination of frequency domain and spatial domain is proposed to make up for the inefficiencies of the traditional algorithm in infrared image enhancement. Histogram equalization is used in the spatial domain to deal with infrared image,which improves the gray contrast of the image. While,in the frequency domain,high pass filter is employed to sharpen the infrared image and enhance the edge of the image and details.Simulation results show that the algorithm in rail crack infrared image enhancement can better suppress background noise, effectively highlight rail crack faint details, protect the details of the image information,facilitate extraction of rail crack defect information.