现代电子技术
現代電子技術
현대전자기술
MODERN ELECTRONICS TECHNIQUE
2015年
3期
57-58
,共2页
压缩感知%实时目标追踪%贪婪算法%稀疏重建
壓縮感知%實時目標追蹤%貪婪算法%稀疏重建
압축감지%실시목표추종%탐람산법%희소중건
compressive sensing%real-time object tracking%greedy algorithm%sparse reconstruction
提出了一种高效的基于压缩感知的实时目标追踪算法,该算法将空域数据转换到小波域,然后利用变密度采样矩阵对小波域数据进行压缩,从而极大地降低了数据量。在稀疏重建上,将St?OMP算法代替OMP算法以提高稀疏重建的速度。在多种具有挑战性的视频序列上进行实验,结果表明该算法提高了追踪准确度和速度。
提齣瞭一種高效的基于壓縮感知的實時目標追蹤算法,該算法將空域數據轉換到小波域,然後利用變密度採樣矩陣對小波域數據進行壓縮,從而極大地降低瞭數據量。在稀疏重建上,將St?OMP算法代替OMP算法以提高稀疏重建的速度。在多種具有挑戰性的視頻序列上進行實驗,結果錶明該算法提高瞭追蹤準確度和速度。
제출료일충고효적기우압축감지적실시목표추종산법,해산법장공역수거전환도소파역,연후이용변밀도채양구진대소파역수거진행압축,종이겁대지강저료수거량。재희소중건상,장St?OMP산법대체OMP산법이제고희소중건적속도。재다충구유도전성적시빈서렬상진행실험,결과표명해산법제고료추종준학도화속도。
An efficient object tracking algorithm based on compressive sensing is proposed to improve the implementation speed and efficiency of ?1 tracker and the OMP tracker. It transforms the spatial data to the wavelet domain,and compresses wavelet signal by means of variable density sampling matrix to reduce the data size greatly. Besides,OMP algorithm is replaced by St?OMP algorithm to improve the speed of sparse reconstruction in the process of sparse reconstruction. Experiments in several challenging video sequences show that the proposed algorithm achieves better tracking result both in accuracy and speed.