Class Boundary Exemplar Selection Based Incremental Learning for Automatic Target Recognition

When adding new tasks/classes in an incremental learning scenario, the previous recognition capabilities trained on the previous training data can be lost. In the real-life application of automatic target recognition (ATR), part of the previous samples may be able to be used. Most incremental learning methods have not considered how to save the previous key samples. In this article, the class boundary exemplar selection-based incremental learning (CBesIL) is proposed to save the previous recognition capabilities in the form of the class boundary exemplars. For exemplar selection, the class boundary selection method based on local geometrical and statistical information is proposed. And when adding new classes continually, a class-boundary-based data reconstruction method is introduced to update the exemplar set. Thus, when adding new classes, the previous class boundaries could be kept complete. Experimental results demonstrate that the proposed CBesIL outperforms the other state of the art on the accuracy of multiclass recognition and class-incremental recognition.

[1]  Matthieu Guillaumin,et al.  Incremental Learning of Random Forests for Large-Scale Image Classification , 2016, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Xin Yao,et al.  Resampling-Based Ensemble Methods for Online Class Imbalance Learning , 2015, IEEE Transactions on Knowledge and Data Engineering.

[3]  Heiko Wersing,et al.  Incremental on-line learning: A review and comparison of state of the art algorithms , 2018, Neurocomputing.

[4]  Q. M. Jonathan Wu,et al.  Incremental Learning in Human Action Recognition Based on Snippets , 2012, IEEE Transactions on Circuits and Systems for Video Technology.

[5]  Haipeng Wang,et al.  Target Classification Using the Deep Convolutional Networks for SAR Images , 2016, IEEE Transactions on Geoscience and Remote Sensing.

[6]  Lei Zhang,et al.  Fine-Tuning Convolutional Neural Networks for Biomedical Image Analysis: Actively and Incrementally , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[7]  Byoung-Tak Zhang,et al.  Overcoming Catastrophic Forgetting by Incremental Moment Matching , 2017, NIPS.

[8]  Jian Yang,et al.  On Selecting Effective Patterns for Fast Support Vector Regression Training , 2018, IEEE Transactions on Neural Networks and Learning Systems.

[9]  Feng Huang,et al.  Robust Prototype-Based Learning on Data Streams , 2018, IEEE Transactions on Knowledge and Data Engineering.

[10]  Yiming Pi,et al.  Open Set Incremental Learning for Automatic Target Recognition , 2019, IEEE Transactions on Geoscience and Remote Sensing.

[11]  Christoph H. Lampert,et al.  iCaRL: Incremental Classifier and Representation Learning , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[12]  Yuan Yan Tang,et al.  New Incremental Learning Algorithm With Support Vector Machines , 2019, IEEE Transactions on Systems, Man, and Cybernetics: Systems.

[13]  Peter Rockett,et al.  The training of neural classifiers with condensed datasets , 2002, IEEE Trans. Syst. Man Cybern. Part B.

[14]  Yuhua Li,et al.  Selecting Critical Patterns Based on Local Geometrical and Statistical Information , 2011, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[15]  Jianyu Yang,et al.  SAR Automatic Target Recognition Based on Multiview Deep Learning Framework , 2018, IEEE Transactions on Geoscience and Remote Sensing.

[16]  Jiwon Kim,et al.  Continual Learning with Deep Generative Replay , 2017, NIPS.

[17]  Jian Yang,et al.  An Improved Attributed Scattering Model Optimized by Incremental Sparse Bayesian Learning , 2016, IEEE Transactions on Geoscience and Remote Sensing.

[18]  Hamid Beigy,et al.  Detection of evolving concepts in non-stationary data streams: A multiple kernel learning approach , 2018, Expert Syst. Appl..

[19]  Hamid Beigy,et al.  Novel class detection in data streams using local patterns and neighborhood graph , 2015, Neurocomputing.

[20]  Eric R. Keydel,et al.  MSTAR extended operating conditions: a tutorial , 1996, Defense, Security, and Sensing.

[21]  Gerhard Tröster,et al.  Incremental kNN Classifier Exploiting Correct-Error Teacher for Activity Recognition , 2010, 2010 Ninth International Conference on Machine Learning and Applications.

[22]  Philip H. S. Torr,et al.  Riemannian Walk for Incremental Learning: Understanding Forgetting and Intransigence , 2018, ECCV.

[23]  Hamid Beigy,et al.  Concept-evolution detection in non-stationary data streams: a fuzzy clustering approach , 2018, Knowledge and Information Systems.

[24]  Gabriela Csurka,et al.  Distance-Based Image Classification: Generalizing to New Classes at Near-Zero Cost , 2013, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[25]  Shiguang Shan,et al.  Exemplar-Supported Generative Reproduction for Class Incremental Learning , 2018, BMVC.

[26]  Zongjie Cao,et al.  SAR target recognition via incremental nonnegative matrix factorization with Lp sparse constraint , 2017, 2017 IEEE Radar Conference (RadarConf).

[27]  Frank Rudzicz,et al.  Fast incremental LDA feature extraction , 2015, Pattern Recognit..

[28]  Xiaorun Li,et al.  Hyperspectral Unmixing Based on Incremental Kernel Nonnegative Matrix Factorization , 2018, IEEE Transactions on Geoscience and Remote Sensing.

[29]  Derek Hoiem,et al.  Learning without Forgetting , 2016, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[30]  João Gama,et al.  A survey on concept drift adaptation , 2014, ACM Comput. Surv..

[31]  Zongjie Cao,et al.  SAR Unlabeled Target Recognition Based on Updating CNN With Assistant Decision , 2018, IEEE Geoscience and Remote Sensing Letters.

[32]  Hamid Beigy,et al.  Exploiting Structural Information of Data in Active Learning , 2014, ICAISC.

[33]  Liyuan Xu,et al.  Automatic target recognition with joint sparse representation of heterogeneous multi-view SAR images over a locally adaptive dictionary , 2016, Signal Process..

[34]  Cordelia Schmid,et al.  End-to-End Incremental Learning , 2018, ECCV.

[35]  Barbara Hammer,et al.  Incremental learning algorithms and applications , 2016, ESANN.

[36]  Jianyu Yang,et al.  Neighborhood Geometric Center Scaling Embedding for SAR ATR , 2014, IEEE Transactions on Aerospace and Electronic Systems.

[37]  Terrance E. Boult,et al.  The Extreme Value Machine , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[38]  Zhang Yi,et al.  An Efficient Representation-Based Method for Boundary Point and Outlier Detection , 2018, IEEE Transactions on Neural Networks and Learning Systems.

[39]  Ribana Roscher,et al.  Incremental Import Vector Machines for Classifying Hyperspectral Data , 2012, IEEE Transactions on Geoscience and Remote Sensing.

[40]  Ales Leonardis,et al.  Incremental PCA for on-line visual learning and recognition , 2002, Object recognition supported by user interaction for service robots.

[41]  Francisco Herrera,et al.  SMOTE-IPF: Addressing the noisy and borderline examples problem in imbalanced classification by a re-sampling method with filtering , 2015, Inf. Sci..

[42]  Huangang Wang,et al.  Parameter Selection of Gaussian Kernel for One-Class SVM , 2015, IEEE Transactions on Cybernetics.

[43]  Rui Min,et al.  SAR target recognition via incremental nonnegative matrix factorization with L p sparse constraint , 2017 .