A New Method of Glare Protection on Highway Real-time Monitoring During Nighttime

摘要

When car run in highway during night, need switch on its headlight, for intensity of the light is very high, it will cause glare vision, the camera of monitoring system will be saturation, and all the other detail (For example: License plate) can't be shown on the screen except headlight and its glare. In this paper, introduce a new system and method of glare protection on highway nighttime monitoring, it can decrease the intensity of headlight and the glare vision in real-time, so the monitoring system could gain more details of highway from camera. The whole system's hardware is made up of outer camera-lens, inner camera-lens, spatial light modulator array and image sensor. The outer camera-lens images the objects (cars on highway) on the plane of the spatial light modulator array's panel. So, the spatial light modulator array can modulate part of the image on its panel, such as decrease or increase transmission ratio (or reflect ratio). The inner camera-lens images the spatial light modulator array's panel to the image sensor. So, the image on the image sensor is the image of object image after modulated. After the image sensor gained the image, we will use our software to analyze the image, use real-time processing to get the saturation and glared region. Firstly, we use threshold arithmetic to get the saturation and high intensive pixels of the image; secondly, we use filter to get ride of the noise made from threshold arithmetic, so we can get the region of saturation and glare region of the original image; thirdly, we do expand arithmetic at the direction of car's moving, and feedback the image to spatial light modulator. If we don't use expand arithmetic at the direction of car's moving, and feedback the image gained after step 2, for the car in highway is moving very fast, after the time of image processing, the car has moved to a new location with it's direction, so the feedback image will not decrease the light intensity of the car's headlight and it's glare region properly. For these three processes are decreasing the intensity of the image region, we called it negative feedback. At same time, we do positive feedback to increase the transmission ratio (or reflect ratio) at the low light intensity region, so that more details of the object will be imaging on image sensor. In fact, the image feedback on the spatial light modulator array is the composite of the negative feedback and positive feedback. In experiment, the spatial light modulator array is liquid crystal on silicon (LCoS), all image processing is on computer, the feedback image transfers via DVI bus. For the limit of operation system, the feedback time is about 100ms.