【摘要】 当天气出现轻度雾霾时,户外光线受到反射遮挡等物理作用的影响,使得图像的能见度降低,图像出现过多冗余噪声,降低图像清晰度.本文采用三维重建技术中的双目视觉系统技术,采集关于同一景物的两幅图像,通过自适应的立体匹配方法得到初步粗糙的视差图像,然后依据Retinex原理对粗糙视差图像进行滤波的平滑处理,将实验结果叠加于双目视觉系统得到的左侧图上,即得到最终地清晰图像.仿真实验论证证明,双目视觉系统是一种有效的图像信息采集方式,其运用于图像清晰化领域确实可行,有效降低了低频有用信息的损耗,提高了图像清晰度.更多还原

【Abstract】 Collected images are often degraded in terms of contrast and visibility by scattering caused by atmospheric particles. An algorithm based on binocular stereo vision is proposed,in order to collect two pictures on different visual angle. A rough scene parallax images could be obtained through the adaptive stereo matching method,and then it will be smoothed according to the Retinex principle. The result obtained is superimposed over the original picture in left. Through the simulation experiment,the result shows the practicability of this algorithm working on the image sharpening field,and the Binocular vision system is also good at collecting image information.The study demonstrates that the algorithm can reduce the low frequency information loss,and improve the robustness and reliability of it.更多还原