This paper deals with a vision improvement system (VIS) developed using real-time image processing technique by CSIR-Central Institute of Mining and Fuel Research, Dhanbad aiming at providing comparatively clear vision to the operators of HEMM (Heavy Earth Moving Machinery) in foggy weather condition in opencast mines. It can also be used in deteriorated visibility situation due to mist, haze, smog, etc. while driving other vehicles also. In adverse weather, mining processes cannot be carried out during one-third of a day due to problem of perceptibility which reduces production rate of opencast mines. The developed system amplifies colour, sets the brightness and contrast to the input video to make objects more visible by reducing the effect of fog, mist, haze and smog. In order to enable quick real time operation, a fast CPU with dedicated graphics processing unit is used which is well suited for intensive data processing; because of its high computational efficiency and robustness, the developed system is able to operate at high frame rate and it is suitable for real-time driving assistance in adverse weather condition.

本文涉及由CSIR-中央矿业和燃料研究所（Dhanbad）使用实时图像处理技术开发的视力改善系统（VIS），旨在为雾中的HEMM（重型地球运动机械）的操作人员提供相对清晰的视力露天矿的天气状况。当驾驶其他车辆时，由于雾气，雾霾，烟雾等原因，也可用于能见度较低的情况。在恶劣天气下，由于可感知性问题降低了露天矿的生产率，因此一天的三分之一无法进行采矿过程。开发的系统可放大色彩，设置输入视频的亮度和对比度，从而减少雾，雾，雾霾和烟雾的影响，从而使物体更清晰可见。为了实现快速的实时操作，使用带有专用图形处理单元的快速CPU，它非常适合密集数据处理；由于其高计算效率和鲁棒性，因此开发的系统能够以高帧速率运行，并且适合在恶劣天气条件下进行实时驾驶辅助。