基于双极偏好的滑动窗口参数优化方法

计算机科学 ›› 2014, Vol. 41 ›› Issue (3): 297-301.

基于双极偏好的滑动窗口参数优化方法

邱飞岳,金锋涛,王丽萍,张维泽

浙江工业大学现代教育技术研究所杭州310023;浙江工业大学信息工程学院杭州310023;浙江工业大学智能信息处理研究所杭州310023;浙江工业大学现代教育技术研究所杭州310023

出版日期:2018-11-14 发布日期:2018-11-14
基金资助:
本文受国家自然科学基金(61379077,61070135),国家社会科学基金(10GBL095),浙江省自然科学基金(LY12F02032)资助

Sliding Window Parameter Optimization Method Based on Bipolar Preferences

QIU Fei-yue,JIN Feng-tao,WANG Li-ping and ZHANG Wei-ze

Online:2018-11-14 Published:2018-11-14

摘要/Abstract

摘要： 滑动窗口是形状匹配中的常用检测方法,可以检测图像中不同尺度不同位置的多个物体。检测效果采用检测率和误检率来衡量。在传统的滑动窗口检测方法中,通常基于经验选取滑动步长和图像缩放规模这两个参数值,来获得较高的检验率和较低的误检率。然而这是典型的两目标优化问题,传统方法未考虑决策者对检验率与误检率的不同偏好。根据实际情况,考虑到决策者的正偏好(高检验率与低误检率)及负偏好(低检验率和高误检率),引入双极偏好控制策略,提出基于双极偏好的多目标粒子群算法(BPMOPSO)的滑动窗口参数优化方法。通过Leeds Cows图像数据集上图像的检测实验结果表明,与传统算法相比,该算法显著改善了滑动窗口检测中的检验率和误检率,且大大提高了运行效率。

关键词: 滑动窗口,多目标算法,双极偏好,参数优化中图法分类号TP301.6文献标识码A

Abstract: One of the commonly-used detection methods in shape matching is the sliding window,in which multiple objects,different in size and position,can be detected．The detection performance is generally measured by detection rate and false positive rate．The two parameters in sliding window detection method,sliding step and the scale step are empirically selected for high detection rate and low false positive rate．However,those two factors can be formulated as a typical two-objective optimization problem,while the empirical selection shows no consideration over decision-makers’ different preferences regarding detection rate and false positive rate．Given the fact that decision makers’ positive pre-ferences are represented by high detection rate and low false positive rate,and negative preferences,by low detection rate and high false positive rate,the paper introduced the bipolar control strategy and then proposed a new way to optimize the sliding window parameters based on Bipolar Preference Multi-objective Particle Swarm Optimization (BPMOPSO)．The new method was applied in the detection experiment on Leeds Cows image datasets．The experiment results show that the performance of the new method is largely improved,i.e.,the false positive rate declines sharply and the detection rate improves significantly,and that the efficiency of the algorithm ameliorates considerably as well.

Key words: Sliding window,Multi-objective algorithm,Bipolar preferences,Parameter optimization

邱飞岳,金锋涛,王丽萍,张维泽. 基于双极偏好的滑动窗口参数优化方法[J]. 计算机科学, 2014, 41(3): 297-301. https://doi.org/

QIU Fei-yue,JIN Feng-tao,WANG Li-ping and ZHANG Wei-ze. Sliding Window Parameter Optimization Method Based on Bipolar Preferences[J]. Computer Science, 2014, 41(3): 297-301. https://doi.org/

参考文献

[1] Shotton J,Blake A,Cipolla R.Multi-scale categorical object re-cognition using contour fragments[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2008,0(7):1270-1281
[2] Viola P,Jones M．Rapid object detection using a booste dcascade of simple features[C]∥Proceeding of IEEE Conference on Computer Vision and Pattern Recognition．Hawaii,USA:IEEE,2001:511-518
[3] Agarwal S,Awan A,Roth D.Learning to detect objects in images via a sparse,part-based representation[J].IEEE Transaction on Pattern Analysis and Machine Intelligence,2004,6(11):1475-1490
[4] Fawcett T.An introduction to ROC analysis[J]．Pattern Recognition Letters,2006,7:861-874
[5] Perez C B,Olague G.Learning invariant region descriptor operators with genetic programming and the f-measure[C]∥Procee-dings of the Nineteenth International Conference on Pattern Re-cognition．Florida,USA:IEEE,2008:3201-3204
[6] Airola A,Pahikkala T,Waegeman W,et al.An experimentalcomparison of cross-validation techniques for estimation the areaunder the ROC curve[J]．Computational Statistics and Data Analysis,2011,5(4):1828-1844
[7] Everson R M,Fieldsend J E.Multi-class ROC analysis from amulti-objective optimisation perspective[J].Patten Recognition Letters,2006,7:918-927
[8] Chatelain C,Adam S,Lecourtier Y,et al.A multi-model selec-tion framework for unknown and/or evolutive misclassification cost problems[J]．Pattern Recognition,2010,3:815-823
[9] Verschae R,Solar J R,Koppen M,et al.Improvement of a face detection system by evolutionary multi-objective optimization [C]∥Proceedings of the Fifth International Conference on Hybrid Intelligent Systems.Rio de Janeiro,Brazil:IEEE,2005:361-366
[10] 王丽萍,江波,蔡建湖,等.基于双极偏好控制的多目标粒子群优化算法研究[J].信息与控制,2009,8(6):711-717
[11] Deb K.Multi-Objective Optimization Using Evolutionary Algorithms[M].Chichester:John Wiley & Sons,2001
[12] 公茂果,李焦成,杨咚咚,等.进化多目标优化算法研究[J].软件学报,2009,0(2):271-289
[13] Kennedy J,Eberhart R.Particle swarm optimiza-tion [C]∥Proceedings of the 1995IEEE International Conference on Neural Networks Proceedings.Piscataway,NJ,USA:IEEE,1995:1942-1948
[14] Hwang C L,Yoon K S.Multiple attribute decision making:methods and applications:A state-of-the-art survey[M]．Berlin,Germany:Springer-Verlag,1981
[15] Gong Dun-wei,Zhang Yong,Zhang Jian-hua．Multi-objectiveparticle swarm optimization based on minimal particle angle[J]．Lecture Note in Computer Science,2005,4:571-580
[16] Deb K,Pratap A,Agarwal S,et al.A fast and elitist multiobjective genetic algorithm:NSGA-II [J].IEEE Transactions on Evo-lutionary Computation,2002,6(2):182-197
[17] Zhang Wei-ze,Vézien J M,Tong Ruo-feng,et al.Hough transformed contour line segments for object detection[C]∥IADIS International Conferences Computer Graphics,Visualization,Computer Vision and Image Processing.Freiburg,Germany:2010:235-242

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed