Contractive Autoencoder and SVM for Recognition of Handwritten Devanagari Numerals

被引：0
|
作者：
Kabra, Ruhi R. [1 ]
机构：
[1] MSSs Coll Engn & Technol, Dept Comp Sci & Engn, Jalna, India
来源：
2017 1ST INTERNATIONAL CONFERENCE ON INTELLIGENT SYSTEMS AND INFORMATION MANAGEMENT (ICISIM) | 2017年
关键词：
Contractive autoencoder; SVM; Handwritten numerals recognition;
D O I：
暂无
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
Representation of data is very important in case of machine learning. Better the representation, the classifiers will give better results. Contractive autoencoders are used to learn the representation of data which are robust to small changes in the input. This paper uses contractive autoencoder and SVM classifier for handwritten Devanagari numerals recognition. The accuracy obtained using CAE+SVM is 96 %.
引用
收藏
页码：26 / 29
页数：4
相关论文
共 50 条
  • [41] Similar Handwritten Devanagari Character Recognition by Critical Region Estimation
    Jangid, Mahesh
    Srivastava, Sumit
    2016 INTERNATIONAL CONFERENCE ON ADVANCES IN COMPUTING, COMMUNICATIONS AND INFORMATICS (ICACCI), 2016, : 1936 - 1939
  • [42] A Preprocessing Technique for Recognition of Online Handwritten Gurmukhi Numerals
    Bawa, Rajesh Kumar
    Rani, Rekha
    HIGH PERFORMANCE ARCHITECTURE AND GRID COMPUTING, 2011, 169 : 275 - 281
  • [43] Deep Learning Approach for Recognition of Handwritten Kannada Numerals
    Ganesh, Anirudh
    Jadhav, Ashwin R.
    Pragadeesh, K. A. Cibi
    PROCEEDINGS OF THE EIGHTH INTERNATIONAL CONFERENCE ON SOFT COMPUTING AND PATTERN RECOGNITION (SOCPAR 2016), 2018, 614 : 294 - 303
  • [44] Input fuzzy modeling for the recognition of handwritten Hindi numerals
    Hanmandlu, M.
    Grover, J.
    Madasu, V. K.
    Vasikarla, S.
    INTERNATIONAL CONFERENCE ON INFORMATION TECHNOLOGY, PROCEEDINGS, 2007, : 208 - +
  • [45] Handwritten Optical Character Recognition System for Sindhi Numerals
    Sanjrani, Anwar Ali
    Baber, Junaid
    Bakhtyar, Maheen
    Noor, Waheed
    Khalid, Muhammad
    2016 INTERNATIONAL CONFERENCE ON COMPUTING, ELECTRONIC AND ELECTRICAL ENGINEERING (ICE CUBE), 2016, : 262 - 267
  • [46] Wavelet feature extraction for the recognition and verification of handwritten numerals
    Zhang, P
    Bui, TD
    Suen, CY
    Wavelet Analysis and Active Media Technology Vols 1-3, 2005, : 3 - 9
  • [47] Handwritten numerals recognition based on segmented contour feature
    Lou, Zhen
    Hu, Zhongshan
    Yang, Jingyu
    Liu, Ke
    Suen, C.Y.
    Jisuanji Xuebao/Chinese Journal of Computers, 1999, 22 (10): : 1065 - 1073
  • [48] Handwritten numerals recognition based on segmented contour feature
    Jisuanji Xuebao, 10 (1065-1073):
  • [49] Recognition of handwritten Hindu numerals using structural descriptors
    Elnagar, A
    Harous, S
    JOURNAL OF EXPERIMENTAL & THEORETICAL ARTIFICIAL INTELLIGENCE, 2003, 15 (03) : 299 - 314
  • [50] Multivariate structural Bernoulli mixtures for recognition of handwritten numerals
    Grim, J
    Pudil, P
    Somol, P
    15TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, VOL 2, PROCEEDINGS: PATTERN RECOGNITION AND NEURAL NETWORKS, 2000, : 585 - 589
12345