Is quantity better than quality?: a study case on a simple classifier applied to digit character recognition (情報論的学習理論と機械学習)
スポンサーリンク
概要
- 論文の詳細を見る
This paper deals with a learning method for digit character recognition. We propose to expand character image databases, by automatically generating deformations, noises and new fonts, and then to use a simple classifier based on k nearest neighbor technique. The results of first experiments show that the classifier is more robust on the resulting larger database. Moreover, performance on the MNIST digit character database is very promising.
- 社団法人電子情報通信学会の論文
- 2010-08-29
著者
-
Omachi Shinichiro
Graduate School Of Engineering Tohoku University
-
KISE Koichi
Department of Computer Science and Intelligent Systems, Osaka Prefecture University
-
IWAMURA Masakazu
Department of Computer Science and Intelligent Systems, Osaka Prefecture University
-
Kise Koichi
Department Of Computer Science And Intelligent Systems Osaka Prefecture University
-
Kise Koichi
Graduate School Of Engineering Osaka Prefecture University
-
Iwamura Masakazu
Graduate School Of Engineering Osaka Prefecture University
-
Iwamura Masakazu
Department Of Computer Science And Intelligent Systems Osaka Prefecture University
-
Barrat Sabine
Graduate School Of Engineering Osaka Prefecture University:japan Society For The Promotion Of Scienc
-
Uchida Seiichi
Faculty Of Information Science And Electrical Engineering Kyushu University
-
ALARY Geoffrey
Graduate School of Engineering, Osaka Prefecture University
-
Alary Geoffrey
Graduate School Of Engineering Osaka Prefecture University:eisti Computer Science And Mathematics En
関連論文
- An Approximation Method of the Quadratic Discriminant Function and Its Application to Estimation of High-Dimensional Distribution(Image Recognition and Understanding)
- An Attempt of CUDA Implementation of PCA-SIFT(Internationa Session 6)
- Position detection for a camera pen using LLAH and dot patterns (パターン認識・メディア理解)
- Pattern Recognition with Gaussian Mixture Models of Marginal Distributions
- Comparative Study on Different Color Space Models for Skin Color Segmentation
- A Discriminant Function Based on Feature Transformation Considering Normality Improvement of Distribution
- Segmenting Shape Using Deformation Information
- Object detection in images with cluttered background by using local features and their configuration (パターン認識・メディア理解)
- Position detection for a camera pen using LLAH and dot patterns (ヒューマン情報処理)
- Is quantity better than quality?: a study case on a simple classifier applied to digit character recognition (パターン認識・メディア理解)
- Is quantity better than quality?: a study case on a simple classifier applied to digit character recognition (情報論的学習理論と機械学習)
- A Survey of Elastic Matching Techniques for Handwritten Character Recognition(Character Recognition, Document Image Understanding and Digital Documents)
- Nonuniform Slant Correction for Handwritten Word Recognition(Image Recognition, Computer Vision)
- Novel Confidence Feature Extraction Algorithm Based on Latent Topic Similarity
- Representative Graph Generation for Graph-Based Character Recognition
- Construction of Multi-Template Dictionary for Pattern Recognition by Decreasing Region Overlaps
- Decorative Character Recognition by Graph Matching(Image Recognition, Computer Vision)
- Better Decision Boundary for Pattern Recognition with Supplementary Information(Pattem Recognition)
- Extraction of Structure of Silhouette Images by Weighted Minimum Common Supergraph(Internationa Session 2)
- Comparative Study on Color Components for PCA-Based Face Recognition
- Cartoon Character Recognition Using Concentric Multi-Region Histograms of Oriented Gradients
- Cartoon Character Recognition Using Concentric Multi-Region Histograms of Oriented Gradients
- A Discriminative Metric Learning Algorithm for Face Recognition