摘要
当前的车辆追踪定位系统中,多半处于半自动监控状态,以人工视觉发现为主,系统在抗视觉误差干扰以及实时性方面性能较差;为此提出一种新型车辆传感追踪系统的设计方法,给出了详细的硬件软件设计细节;在系统开发中,运用高清视觉传感器组成的车辆信息采集模块对车辆进行图像采集,运用抗干扰能力较强的DSP芯片硬件配合车辆图像特征离散程度计算的软件算法,求得车辆图像中值与各自的像素均值偏差乘积的平均值,运用压缩比较高的TR100芯片完成信息传递,消除视觉误差干扰;实验证明,该系统能够对车辆进行实时追踪定位,定位误差有了明显降低。
The current vehicle tracking and positioning system, mostly in semi automatic monitoring state to artificial vision found that give priority to, system in resistance to visual error interference and in real time performance is poor. Therefore this paper puts forward a new type of vehicle sensor tracking system design methods, gives the detailed hardware and software design details. In system development, the use of high--definition visual sensor composed of vehicle information acquisition module of vehicle image acquisition, the use of anti--in- terferer/ce ability strong DSP chip hardware fit vehicle image features of the dispersion degree calculation software algorithm, get vehicle im- age value and their respective pixel deviation of mean value of mean the product, using compression higher TR100 chip complete information transfer, eliminate visual error interference. The experimental results prove that the system can real--time tracking of vehicle positioning, positioning error are down significantly.
出处
《计算机测量与控制》
北大核心
2013年第3期800-802,815,共4页
Computer Measurement &Control
关键词
传感器
视觉图像
车辆定位
sensor
visual image
vehicle positioning
