Journal of Computer Applications ›› 2015, Vol. 35 ›› Issue (8): 2274-2279.DOI: 10.11772/j.issn.1001-9081.2015.08.2274

Previous Articles     Next Articles

Uniform SILTP based background modeling and its implementation on Intel HD graphics

LIN Zecheng, ZHU Jianqing, LIAO Shengcai, LI Stan Z.   

  1. 1. Institute of Automation, Chinese Academy of Sciences, Beijing 100080, China;
    2. University of Chinese Academy of Sciences, Beijing 100080, China
  • Received:2015-02-12 Revised:2015-04-05 Online:2015-08-10 Published:2015-08-14

基于均匀尺度不变局部三元模式的背景建模及其在Intel HD显卡上的并行实现

林泽诚, 朱建清, 廖胜才, 李子青   

  1. 1. 中国科学院 自动化研究所, 北京 100080;
    2. 中国科学院大学, 北京 100080
  • 通讯作者: 林泽诚(1989-),男,山东威海人,硕士研究生,主要研究方向:图像处理、视频监控,linzecheng1989@126.com
  • 作者简介:朱建清(1987-),男,福建莆田人,博士研究生,主要研究方向:智能视频监控、图像理解; 廖胜才(1982-),男,广东梅州人,副研究员,博士,CCF会员,主要研究方向:人脸检测与识别、物体检测与识别、度量学习、行人再识别、智能视频分析与监控; 李子青(1958-),男,湖南株洲人,研究员,博士,主要研究方向:统计模式识别、机器学习、图像处理、计算机视觉方法、人脸识别、智能视频分析。
  • 基金资助:

    国家自然基金项目资助项目(61105023,61103156,61105037,61203267,61375037);国家科技支撑计划项目(2013BAK02B01);中国科学院重点部署项目(KGZD-EW-102-2)。

Abstract:

Since Scale Invariant Local Ternary Pattern (SILTP) background modeling algorithm is of high complexity and slow computing speed, which is not suitable for real-time video processing, a new method named Uniform Scale Invariant Local Ternary Pattern (USILTP) background modeling algorithm was proposed. Firstly, the feature of USILTP was extracted by regulating the frequency of SILTP coding jump in order to reduce the feature dimension of SILTP. Secondly, a USILTP background modeling parallel algorithm based on Intel core graphics (Intel HD) and Open Computing Language technology (OpenCL) was designed and implemented to further accelerate USILTP background modeling algorithm. Finally, the foreground result of USILTP background modeling algorithm was optimized by combing multiple color channel models. The experimental result shows that the proposed algorithm can be applied to process 320×240 resolution video at a rate of 98 frame/s on the Intel HD 4600, which is 4 times faster than that of SILTP background modeling algorithm. In terms of foreground detection, the performance of the proposed algorithm is improved by 2.1% compared with SILTP background modeling algorithm on the public dataset.

Key words: Scale Invariant Local Ternary Pattern (SILTP), Intel core graphics, Open Computing Language (OpenCL), background modeling

摘要:

针对尺度不变局部三元模式(SILTP)背景建模算法复杂度较高、计算速度较慢,不利于高速视频处理的问题,提出了一种新的均匀尺度不变局部三元模式(USILTP)背景建模并行算法。首先,通过规范SILTP编码的跳变次数得到USILTP纹理特征,实现了SILTP特征降维;其次,设计并实现基于英特尔核芯显卡(Intel HD)及开放式计算语言(OpenCL)的USILTP背景建模并行算法,进一步加速了USILTP背景建模算法;最后,通过融合多颜色通道模型结果,优化了USILTP背景建模算法的前景效果。实验结果表明,在Intel HD 4600上处理320×240分辨率的视频,该算法将速度提升至98 frame/s,比SILTP背景建模算法快4倍;在前景检测方面,该算法在背景公开数据库上的性能比SILTP背景建模算法提升了2.1%。

关键词: 尺度不变局部三元模式, 英特尔核芯显卡, 开放式计算语言, 背景建模

CLC Number: