电子设计工程
電子設計工程
전자설계공정
ELECTRONIC DESIGN ENGINEERING
2015年
8期
140-142,146
,共4页
车辆跟踪%卡尔曼滤波%特征加权融合%空间区域划分
車輛跟蹤%卡爾曼濾波%特徵加權融閤%空間區域劃分
차량근종%잡이만려파%특정가권융합%공간구역화분
vehicle tracking%Kalman filter%characteristics weighted fusion%space regional division
针对行驶中的车辆易受阴影及相似车辆的影响问题,提出了一种基于卡尔曼滤波的区域特征融合车辆跟踪算法。算法首先借鉴空间金字塔匹配模型将车辆在空间上进行划分,然后将颜色特征、局部三值模式纹理特征和Hu不变矩特征进行加权融合,提取车辆各区域信息,最后引入卡尔曼滤波对目标车辆的运动趋势进行预测。实验结果显示,该算法在目标车辆受阴影及特征相似车辆干扰时,能够有效区分目标与背景,实现正确跟踪。
針對行駛中的車輛易受陰影及相似車輛的影響問題,提齣瞭一種基于卡爾曼濾波的區域特徵融閤車輛跟蹤算法。算法首先藉鑒空間金字塔匹配模型將車輛在空間上進行劃分,然後將顏色特徵、跼部三值模式紋理特徵和Hu不變矩特徵進行加權融閤,提取車輛各區域信息,最後引入卡爾曼濾波對目標車輛的運動趨勢進行預測。實驗結果顯示,該算法在目標車輛受陰影及特徵相似車輛榦擾時,能夠有效區分目標與揹景,實現正確跟蹤。
침대행사중적차량역수음영급상사차량적영향문제,제출료일충기우잡이만려파적구역특정융합차량근종산법。산법수선차감공간금자탑필배모형장차량재공간상진행화분,연후장안색특정、국부삼치모식문리특정화Hu불변구특정진행가권융합,제취차량각구역신식,최후인입잡이만려파대목표차량적운동추세진행예측。실험결과현시,해산법재목표차량수음영급특정상사차량간우시,능구유효구분목표여배경,실현정학근종。
In order to solve the moving vehicles are affected by shadow and similar vehicles. Aregional characteristics fusion of vehicle tracking base on Kalman filter is proposed. The approach uses the Space pyramid matching model for reference to divide the target in space. Then extract color features, LTP features and Hu invariant moments from each sub-region of the target. Finally, it usesKalman filter to predict the trend of the movement of the target vehicle. The experimental results show thatthe approach can correctly distinguish between target and backgroundwhen the target is affected by shadow and similar vehicles.