空间划分的目标图像识别与跟踪技术  

Target Image Recognition and Tracking Technology Based on Space Partition

作　　者：张学英[1] 韩广良[2] 

机构地区：[1]河北科技师范学院职教研究院,河北秦皇岛066004 [2]中国科学院长春光学精密机械与物理研究所,吉林长春130000

出　　处：《华侨大学学报（自然科学版）》2017年第2期257-261,共5页Journal of Huaqiao University(Natural Science)

基　　金：国家自然科学基金资助项目(61172111)

摘　　要：针对多目标识别跟踪算法中存在的低效率、高能耗,易产生误检、漏检等问题,以跟踪精确率、能量损耗为评价指标展开研究.对目前定位与跟踪的相关算法进行分析,提出一种基于空间划分的目标图像识别与跟踪算法.利用空间分簇机制,实时收集目标的位置信息,同时,建立目标丢失与恢复机制,显著提高了目标的识别与跟踪精度,减少能量损耗.结果表明:与常规算法相比,所提算法跟踪成功率提高了10%左右,并能有效减少能量消耗,具有一定的实用价值.Considering multi-object tracking algorithm exists some problem such as low efficiency,high energy loss,easy error-detection and loss dectection,we take the tracking accuracy and energy loss as the evaluation index to develop research.Based on the analysis of the current localization and tracking algorithms,a new method of object localization and tracking algorithm based on space partition is proposed,which uses the spatial clustering mechanism to collect the target location information in real time.At the same time,the object loss and restoration mechanism is established.It can improve the tracking accuracy and reduce the energy loss significantly.The simulation results show that compared with the common algorithm,the tracking success rate in our is increased by 10%,and the energy consumption is decreased effectively in this algorithm effective,and it has application value.

关 键 词：空间划分 图像识别 图像定位 跟踪算法 分簇机制 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]