摘要
在彩色图像视觉跟踪过程中,存在定位精度低、伺服周期长、可靠性差等问题.提出了一种改进的目标物体检测方法,同时运用已有的双目运动机构建立了视觉跟踪系统.在保证定位精度和可靠性不变的前提下,减小了图像处理过程中的计算量,缩短了整个系统的计算周期.同时通过大量的试验进行验证.结果表明,方法原理正确有效,对于具有不同颜色、不同形状、不同尺寸的运动目标物体都能够使双目运动机构跟随目标物体进行,实现了立体视觉平台的旋转运动.
In the process of visual tracking based on color image, there are some problems of lowly oriented precision, slowly servoing speed, and low reliability. An improved arithmetic of searching object was proposed. At the same time, visual tracking system was established using binocular stereo vision institution. This method had less calculation quantity in digital image processing and shorter calculation period in whole system, under insuring invariable oriented precision and reliability. It was proved by large numbers of examination. The result showed that this method was right and availability. Stereovision platform revolved following moving object which had different color, different shape or different size. It realized rotating movement for Stereovision platform.
出处
《哈尔滨工业大学学报》
EI
CAS
CSCD
北大核心
2007年第6期932-935,共4页
Journal of Harbin Institute of Technology
基金
国家高技术研究发展计划资助项目(2006AA04Z245)
关键词
图像分割
视觉跟踪
目标检测
立体视觉平台
image segmentation
visual tracking
object detection
stereovision platform