现代计算机(普及版)
現代計算機(普及版)
현대계산궤(보급판)
MODERN COMPUTER
2014年
5期
76-80
,共5页
视频分析%LPB算法%PCA算法%岭回归算法%人脸识别
視頻分析%LPB算法%PCA算法%嶺迴歸算法%人臉識彆
시빈분석%LPB산법%PCA산법%령회귀산법%인검식별
Video Analysis%Local Binary Patterns(LPB)%Principal Components Analysis(PCA)%Ridge Regression%Face Recognition
智能视频监控随着大数据视频分析技术的应用得到迅速发展,通过实时监控,视频分析系统即时反应,识别出视频监控采集到的目标人脸。在基于ARM11的S3C6410嵌入式硬件平台,以Linux操作系统为软件平台,用LBP算法对人脸图像进行预处理,减少光照对识别的干扰;用PCA算法对经一致模式统计后的均衡直方图进行特征提取和降维,降低数据的计算量;对降维后的数据运用岭回归算法进行分类,设计并优化LBP+PCA算法,结合 Ridge Regression分类技术作为人脸识别视频分析的解决方案。
智能視頻鑑控隨著大數據視頻分析技術的應用得到迅速髮展,通過實時鑑控,視頻分析繫統即時反應,識彆齣視頻鑑控採集到的目標人臉。在基于ARM11的S3C6410嵌入式硬件平檯,以Linux操作繫統為軟件平檯,用LBP算法對人臉圖像進行預處理,減少光照對識彆的榦擾;用PCA算法對經一緻模式統計後的均衡直方圖進行特徵提取和降維,降低數據的計算量;對降維後的數據運用嶺迴歸算法進行分類,設計併優化LBP+PCA算法,結閤 Ridge Regression分類技術作為人臉識彆視頻分析的解決方案。
지능시빈감공수착대수거시빈분석기술적응용득도신속발전,통과실시감공,시빈분석계통즉시반응,식별출시빈감공채집도적목표인검。재기우ARM11적S3C6410감입식경건평태,이Linux조작계통위연건평태,용LBP산법대인검도상진행예처리,감소광조대식별적간우;용PCA산법대경일치모식통계후적균형직방도진행특정제취화강유,강저수거적계산량;대강유후적수거운용령회귀산법진행분류,설계병우화LBP+PCA산법,결합 Ridge Regression분류기술작위인검식별시빈분석적해결방안。
Intelligent video surveillance system has been developing rapidly with the application of large data video analysis technology. Through re-al-time monitoring, real time analysis system of reaction, to identify the target face. The video analysis takes the ARM11 processor as the hardware platform, uses ordinary camera as facial image capture device, 7-inch touch-screen LCD display as an input. Software based on embedded Linux as operating system, combined with LBP local binary pattern, PCA principal component analysis algorithm, Ridge Regression algorithm for face recognition algorithm solution, breaks through the problem face size, distance, light interference.