A Discriminative Cascade CNN Model for Offline Handwritten Digit Recognition

被引：0

作者：

Pan, Shulan ^{[1
]}

Wang, Yanwei

Liu, Changsong

Ding, Xiaoqing

机构：

[1] Tsinghua Univ, Dept Elect Engn, State Key Lab Intelligent Technol, Beijing, Peoples R China

来源：

2015 14TH IAPR INTERNATIONAL CONFERENCE ON MACHINE VISION APPLICATIONS (MVA) | 2015年

关键词：

D O I：

暂无

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

This paper presents a high-performance two-stage cascade CNN model. The main idea behind the cascade CNN model is complementary classification objectives between Stage I and Stage II. Discriminative learning is introduced to train Stage II by feeding back poorly recognized training samples. Experiments have been conducted on the competitive MNIST handwritten digit database. The cascade model achieved the best state-of-the-art performance with an error rate of 0.18%.

引用

页码：501 / 504

页数：4

共 50 条

[21] Improving Offline Handwritten Digit Recognition Using Concavity-Based Features
Karic, Miran
Martinovic, Goran
INTERNATIONAL JOURNAL OF COMPUTERS COMMUNICATIONS & CONTROL, 2013, 8 (02) : 220 - 234
[22] Building efficient CNN architecture for offline handwritten Chinese character recognition
Li, Zhiyuan
Teng, Nanjun
Jin, Min
Lu, Huaxiang
INTERNATIONAL JOURNAL ON DOCUMENT ANALYSIS AND RECOGNITION, 2018, 21 (04) : 233 - 240
[23] Discriminative training of tied mixture density HMMs for online handwritten digit recognition
Nopsuwanchai, R
Biem, A
2003 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOL II, PROCEEDINGS: SPEECH II; INDUSTRY TECHNOLOGY TRACKS; DESIGN & IMPLEMENTATION OF SIGNAL PROCESSING SYSTEMS; NEURAL NETWORKS FOR SIGNAL PROCESSING, 2003, : 817 - 820
[24] Bangla Handwritten Digit Recognition Using Deep CNN for Large and Unbiased Dataset
Shawon, Ashadullah
Rahman, Md. Jamil-Ur
Mahmud, Firoz
Zaman, M. M. Arefin
2018 INTERNATIONAL CONFERENCE ON BANGLA SPEECH AND LANGUAGE PROCESSING (ICBSLP), 2018,
[25] Ensemble Learning Network for Handwritten Digit Recognition Based on Fusion Optimized CNN
Cui, Li
Chen, Ting-Xuan
Xia, Ying-Qing
Cao, Xia
Wu, Ling
Journal of Computers (Taiwan), 2023, 34 (03) : 137 - 150
[26] Offline Handwritten Text Recognition Using Hybrid CNN-BLSTM Network
Namdeo, Rahul Kumar
Gupta, Chetan
Shrivastava, Ritu
Proceedings - 2022 IEEE 11th International Conference on Communication Systems and Network Technologies, CSNT 2022, 2022, : 318 - 323
[27] Offline Handwritten Devanagari Word Recognition Using CNN-RNN-CTC
Bisht M.
Gupta R.
SN Computer Science, 4 (1)
[28] Arabic handwritten digit recognition
Sherif Abdleazeem
Ezzat El-Sherif
International Journal of Document Analysis and Recognition (IJDAR), 2008, 11 : 127 - 141
[29] Arabic handwritten digit recognition
Abdleazeem, Sherif
El-Sherif, Ezzat
INTERNATIONAL JOURNAL ON DOCUMENT ANALYSIS AND RECOGNITION, 2008, 11 (03) : 127 - 141
[30] FIRMLP for Handwritten Digit Recognition
Codrescu, Cristinel
PROCEEDINGS OF 2016 15TH INTERNATIONAL CONFERENCE ON FRONTIERS IN HANDWRITING RECOGNITION (ICFHR), 2016, : 483 - 488

← 1 2 3 4 5 →