光电工程
光電工程
광전공정
OPTO-ELECTRONIC ENGINEERING
2014年
2期
53-62
,共10页
孙锐%侯能干%陈军
孫銳%侯能榦%陳軍
손예%후능간%진군
行人检测%直方图交叉核支持向量机(HIKSVM)%多层次导向边缘能量特征%特征融合%主元分析(PCA)
行人檢測%直方圖交扠覈支持嚮量機(HIKSVM)%多層次導嚮邊緣能量特徵%特徵融閤%主元分析(PCA)
행인검측%직방도교차핵지지향량궤(HIKSVM)%다층차도향변연능량특정%특정융합%주원분석(PCA)
pedestrian detection%histogram intersection kernel support vector machine%multi-level oriented edge energy feature%feature fusion%principal components analysis
行人检测是目标识别领域的一大难点。现阶段用于行人检测的特征维数都比较高，为克服高维特征对实时性的影响，本文运用主元分析(PCA)对特征进行降维，加快检测速度。单一特征的信息有限，本文运用基于线性鉴别分析(LDA)的线性权重融合原则对一些底层特征(颜色、梯度、直方图)和多层次导向边缘能量特征进行特征融合使特征具有多源信息。且上述特征可采用积分图技术进行快速计算，所以行人检测系统的鲁棒性和实时性得到加强。在目标识别领域直方图交叉核支持向量机(HIKSVM)具有分类快，且准确率高的优点，采用其进行分类，系统实时性更进一步提升。实验表明本文方法检测速度和检测率优于经典的HOG+SVM算法。
行人檢測是目標識彆領域的一大難點。現階段用于行人檢測的特徵維數都比較高，為剋服高維特徵對實時性的影響，本文運用主元分析(PCA)對特徵進行降維，加快檢測速度。單一特徵的信息有限，本文運用基于線性鑒彆分析(LDA)的線性權重融閤原則對一些底層特徵(顏色、梯度、直方圖)和多層次導嚮邊緣能量特徵進行特徵融閤使特徵具有多源信息。且上述特徵可採用積分圖技術進行快速計算，所以行人檢測繫統的魯棒性和實時性得到加彊。在目標識彆領域直方圖交扠覈支持嚮量機(HIKSVM)具有分類快，且準確率高的優點，採用其進行分類，繫統實時性更進一步提升。實驗錶明本文方法檢測速度和檢測率優于經典的HOG+SVM算法。
행인검측시목표식별영역적일대난점。현계단용우행인검측적특정유수도비교고，위극복고유특정대실시성적영향，본문운용주원분석(PCA)대특정진행강유，가쾌검측속도。단일특정적신식유한，본문운용기우선성감별분석(LDA)적선성권중융합원칙대일사저층특정(안색、제도、직방도)화다층차도향변연능량특정진행특정융합사특정구유다원신식。차상술특정가채용적분도기술진행쾌속계산，소이행인검측계통적로봉성화실시성득도가강。재목표식별영역직방도교차핵지지향량궤(HIKSVM)구유분류쾌，차준학솔고적우점，채용기진행분류，계통실시성경진일보제승。실험표명본문방법검측속도화검측솔우우경전적HOG+SVM산법。
Pedestrian detection is a major difficulty in object recognition. Features used for pedestrian detection are high in dimension. We use principal component analysis to reduce the dimension of features, and make the detection algorithm run faster. It overcomes the influence of the high dimensional features which reduce the real-time of pedestrian detection. The information content of single feature is limited. To make use of multi-source information feature, we fusion some low-level features (color、gradient、histogram) and multi-level oriented edge energy feature based on the linear discriminant analysis of linear weighted fusion strategy. Features can be calculated fast by integral image technique. The robustness and real-time performance of pedestrian detection system have been strengthened. Histogram intersection kernel support vector machine have the advantage of fast classification and high accuracy in object recognition. It can be used for further enhancing the system real-time performance. The experiments show that the proposed algorithm has faster detection speed and higher precision than the classical algorithm HOG+SVM.
