Semisynthetic Versus Real-World Sonar Training Data for the Classification of Mine-Like Objects
作者： Christopher Barngrover , Ryan Kastner , Serge Belongie
关键词: Boosting (machine learning) 、 Engineering 、 Artificial intelligence 、 Test data 、 Machine learning 、 Sonar 、 Subject-matter expert 、 Local binary patterns 、 Data set 、 Training set 、 Training data sets
摘要: The detection of mine-like objects (MLOs) in sidescan sonar (SSS) imagery continues to be a challenging task. In practice, subject matter experts tediously analyze images searching for MLOs. the literature, there are many attempts at automated target recognition (ATR) detect This paper focuses on classifiers that use computer vision and machine learning approaches. These techniques require large amounts data, which is often prohibitive. For this reason, synthetic semisynthetic data sets training testing commonplace. shows how simple creation scheme can used pretest these data-hungry algorithms determine what features value. provides real-world addition sets. considers Haar-like local binary pattern (LBP) with boosting, showing improvements performance real over as set size increases.
harvard.edu
sci-hub.se 参考文章(22)
Paul Hollesen, Warren A. Connors, Thomas Trappenberg, Comparison of learned versus engineered features for classification of mine like objects from raw sonar images canadian conference on artificial intelligence. pp. 174- 185 ,(2011) , 10.1007/978-3-642-21043-3_21
C.M. Ciany, W.C. Zurawski, G.J. Dobeck, D.R. Weilert, Real-time performance of fusion algorithms for computer aided detection and classification of bottom mines in the littoral environment oceans conference. ,vol. 2, pp. 1119- 1125 ,(2003) , 10.1109/OCEANS.2003.178500
Gerald J. Dobeck, John C. Hyland, Le'Derick Smedley, Automated detection and classification of sea mines in sonar imagery international conference on multimedia information networking and security. ,vol. 3079, pp. 90- 110 ,(1997) , 10.1117/12.280846
Timo Ojala, Matti Pietikäinen, David Harwood, A comparative study of texture measures with classification based on featured distributions Pattern Recognition. ,vol. 29, pp. 51- 59 ,(1996) , 10.1016/0031-3203(95)00067-4
E. Coiras, P.-Y. Mignotte, Y. Petillot, J. Bell, K. Lebart, Supervised target detection and classification by training on augmented reality data Iet Radar Sonar and Navigation. ,vol. 1, pp. 83- 90 ,(2007) , 10.1049/IET-RSN:20060098
Gerald J. Dobeck, Fusing sonar images for mine detection and classification international conference on multimedia information networking and security. ,vol. 3710, pp. 602- 614 ,(1999) , 10.1117/12.357082
Tom Aridgides, Manuel Fernández, Image-based ATR utilizing adaptive clutter filter detection, LLRT classification, and Volterra fusion with application to side-looking sonar international conference on multimedia information networking and security. ,vol. 7664, ,(2010) , 10.1117/12.846596
M. An, J. Cobb, B. Shenefelt, R. Tolimieri, Advances in Group Filter Applications to Sea Mine Detection oceans conference. pp. 1- 5 ,(2006) , 10.1109/OCEANS.2006.306951
S. Reed, Y. Petillot, J. Bell, An automatic approach to the detection and extraction of mine features in sidescan sonar IEEE Journal of Oceanic Engineering. ,vol. 28, pp. 90- 105 ,(2003) , 10.1109/JOE.2002.808199
Payam Saisan, Shubha Kadambe, Shape normalized subspace analysis for underwater mine detection international conference on image processing. pp. 1892- 1895 ,(2008) , 10.1109/ICIP.2008.4712149