Pornographic images recognition based on spatial pyramid partition and multi-instance ensemble learning

Spatial pyramid partition based multi-instance modeling method is proposed.Fuzzy histograms fusion based metadata extraction method is proposed.Extreme learning machine and classifier ensemble based MIL algorithm is proposed.The ELMCE-MIL is robust and comparable to other state-of-the-art MIL algorithms. For tackling the problem of pornographic image recognition, a novel multi-instance learning (MIL) algorithm is proposed by using extreme learning machine (ELM) and classifiers ensemble. Firstly, a spatial pyramid partition-based (SPP) multi-instance modeling technique has been deployed to transform the pornographic images recognition problem into a typical MIL problem. The method has deployed a bag corresponding to an image and an instance corresponding to each partitioned sub-block described by low-level visual features (i.e. color, texture and shape). Secondly, a collection of visual word (VW) has been generated by using hierarchical k-mean clustering method, and then based on the fuzzy membership function between instance and VW, a fuzzy histogram fusion-based metadata calculation method has been proposed to convert each bag to a single sample, which allows the MIL problem to be solved directly by a standard single instance learning (SIL) machine. Finally, by using ELM, a group of base classifiers with different number of hidden nodes have been constructed, and their weights bas been dynamically determined by using performance weighting rule. Therefore, the strategy of classifiers ensemble is used to improve the overall adaptability of proposed ELMCE-MIL algorithm. Experimental results have shown that the method is robust, and its performance is superior to other similar algorithms.

[1]  Jing Wang,et al.  Multiple kernel-based multi-instance learning algorithm for image classification , 2014, J. Vis. Commun. Image Represent..

[2]  Bill Triggs,et al.  Histograms of oriented gradients for human detection , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[3]  Thomas Gärtner,et al.  Multi-Instance Kernels , 2002, ICML.

[4]  Zhan Li,et al.  LSA based multi-instance learning algorithm for image retrieval , 2011, Signal Process..

[5]  James T. Kwok,et al.  Marginalized Multi-Instance Kernels , 2007, IJCAI.

[6]  Wu-Jun Li,et al.  MILD: Multiple-Instance Learning via Disambiguation , 2010, IEEE Transactions on Knowledge and Data Engineering.

[7]  Matti Pietikäinen,et al.  Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[8]  David W. Opitz,et al.  Generating Accurate and Diverse Members of a Neural-Network Ensemble , 1995, NIPS.

[9]  Guoliang Fan,et al.  Automatic Target Recognition in Infrared Imagery Using Dense HOG Features and Relevance Grouping of Vocabulary , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops.

[10]  Jun Zhou,et al.  MILIS: Multiple Instance Learning with Instance Selection , 2011, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Thomas G. Dietterich,et al.  Solving the Multiple Instance Problem with Axis-Parallel Rectangles , 1997, Artif. Intell..

[12]  Gao Wen.  Detecting Pornographic Images with Visual Words , 2008 .

[13]  James R. Foulds,et al.  A review of multi-instance learning assumptions , 2010, The Knowledge Engineering Review.

[14]  Zhiqiong Wang,et al.  Elastic extreme learning machine for big data classification , 2015, Neurocomputing.

[15]  Peter V. Gehler,et al.  Deterministic Annealing for Multiple-Instance Learning , 2007, AISTATS.

[16]  Qi Zhang,et al.  Condensation-based multi-person detection and tracking with HOG and LBP , 2014, 2014 IEEE International Conference on Information and Automation (ICIA).

[17]  David A. Forsyth,et al.  Identifying nude pictures , 1996, Proceedings Third IEEE Workshop on Applications of Computer Vision. WACV'96.

[18]  Yixin Chen,et al.  MILES: Multiple-Instance Learning via Embedded Instance Selection , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[19]  Oded Maron,et al.  Multiple-Instance Learning for Natural Scene Classification , 1998, ICML.

[20]  B. S. Manjunath,et al.  Unsupervised Segmentation of Color-Texture Regions in Images and Video , 2001, IEEE Trans. Pattern Anal. Mach. Intell..

[21]  Wu-Jun Li,et al.  Localized content-based image retrieval through evidence region identification , 2009, 2009 IEEE Conference on Computer Vision and Pattern Recognition.

[22]  Thomas Hofmann,et al.  Multiple instance learning with generalized support vector machines , 2002, AAAI/IAAI.

[23]  Li Zhuo,et al.  Compressed domain based pornographic image recognition using multi-cost sensitive decision trees , 2013, Signal Process..

[24]  Congyan Lang,et al.  Hierarchical sparse representation based Multi-Instance Semi-Supervised Learning with application to image categorization , 2014, Signal Process..

[25]  Chee Kheong Siew,et al.  Extreme learning machine: Theory and applications , 2006, Neurocomputing.

[26]  Kunfeng Wang,et al.  License plate recognition using MSER and HOG based on ELM , 2014, Proceedings of 2014 IEEE International Conference on Service Operations and Logistics, and Informatics.

[27]  R. Balasubramanian,et al.  Expert system design using wavelet and color vocabulary trees for image retrieval , 2012, Expert Syst. Appl..

[28]  Yixin Chen,et al.  Image Categorization by Learning and Reasoning with Regions , 2004, J. Mach. Learn. Res..

[29]  James Ze Wang,et al.  System for screening objectionable images , 1998, Comput. Commun..

[30]  Sheng Tang,et al.  Pornprobe: an LDA-SVM based pornography detection system , 2009, ACM Multimedia.

[31]  David A. Forsyth,et al.  Automatic Detection of Human Nudes , 1999, International Journal of Computer Vision.

[32]  Sally A. Goldman,et al.  MISSL: multiple-instance semi-supervised learning , 2006, ICML.

[33]  Arnaldo de Albuquerque Araújo,et al.  A bag-of-features approach based on Hue-SIFT descriptor for nude detection , 2009, 2009 17th European Signal Processing Conference.

[34]  Feiniu Yuan,et al.  A double mapping framework for extraction of shape-invariant features based on multi-scale partitions with AdaBoost for video smoke detection , 2012, Pattern Recognit..

[35]  Zhi-cheng Zhao.  Combining SVM and CHMM classifiers for porno video recognition , 2012 .

[36]  Markus Koskela,et al.  Skeleton-based action recognition with extreme learning machines , 2015, Neurocomputing.

[37]  Zhi-Hua Zhou,et al.  Multi-instance learning by treating instances as non-I.I.D. samples , 2008, ICML '09.

[38]  F. Xavier Roca,et al.  Compact and adaptive spatial pyramids for scene recognition , 2012, Image Vis. Comput..

[39]  G. G. Stokes.  "J." , 1890, The New Yale Book of Quotations.

[40]  Chenggang Clarence Yan,et al.  Extracting salient region for pornographic image detection , 2014, J. Vis. Commun. Image Represent..

[41]  Zhouyu Fu,et al.  Recognition of Pornographic Web Pages by Classifying Texts and Images , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[42]  Ali Ahmadi,et al.  Intelligent classification of web pages using contextual and visual features , 2011, Appl. Soft Comput..

[43]  R. Polikar,et al.  Ensemble based systems in decision making , 2006, IEEE Circuits and Systems Magazine.