基于遗传算法的太阳影子定位研究

Research on the Sun Shadow Location Based on Genetic Algorithm

确定视频的拍摄日期和地点是视频数据分析的重要研究方向,太阳影子定位技术就是通过视频中物体的影子变化确定视频拍摄的时间和地点的一种方法.针对太阳影子定位问题,在合理的假设下,以太阳高度角为桥梁,建立了杆高、影长、日期、时间、经度、纬度之间的关系式,然后结合视频中能够读取的时间和影长的序列,考虑杆高已知和杆高未知的情况,建立了最小方差优化模型,并设计了粗劣搜索算法+GA算法对视频的拍摄地点(经纬度)进行精确求解.

It is an important research direction to determine the date and location of video data analysis.The sun shadow positioning technology is a method to determine the time and place of video shot by the shadow of the object.In order to solve the problem of the sun shadow position,under reasonable assumptions,with solar alititude angle as a bridge,this paper establishes the relationship of the height of rod,the length of shadow,the date,the time,the longitude and the latitude.Then,lombining with the readable time and the sequence of shodow,the minimum variance optimizational model is establshed with considering both the known and unknown rod length.Besides,this paper designs the coarse search algorithm and GA algorithm in order to solve the video filming location（longitude and latitude）exactly.