ブランケット法を用いた文字列抽出に適した情景画像の2値化法 [in Japanese] A Binarization Method for the Character Extraction from Scene Image by the Blanket Method [in Japanese]
-
- 吉田 大海 YOSHIDA Hiromi
- 神戸大学大学院海事科学研究科 Graduate School of Maritime Science, Kobe University
-
- 田中 直樹 TANAKA Naoki
- 神戸大学大学院海事科学研究科 Graduate School of Maritime Science, Kobe University
Access this Article
Search this Article
Author(s)
-
- 吉田 大海 YOSHIDA Hiromi
- 神戸大学大学院海事科学研究科 Graduate School of Maritime Science, Kobe University
-
- 田中 直樹 TANAKA Naoki
- 神戸大学大学院海事科学研究科 Graduate School of Maritime Science, Kobe University
Abstract
本稿では，情景画像を対象とする文字列抽出に適した大局的2値化方式を提案する．領域抽出を目的として2値化を行う場合，対象領域と背景とをできるだけ分離することが重要となる．しかしながら，情景画像のような画像ではテクスチャやベタ，文字などの多様な領域で構成されているため，従来の大局的2値化方式のように濃淡情報に基づくしきい値決定法では良質な2値化画像を得ることが困難である．本提案方式は，2値化画像の良さを評価することによって2値化しきい値を選択できる方式である．すなわち，ブランケット法を用いることにより2値化画像の領域の密集度・しきい値の変化に対する領域の安定性に関する評価を行っている．また，提案方式を用いた情景画像の2値化実験を行い，目視による2値化画像と比較してほぼ遜色のない2値化画像が得られることを示した．更に，従来の代表的な大局手法である判別分析法との比較でも本方式が文字領域の2値化に関して高性能であることを示している．
We propose a new global binarization method suited for character strings extraction from a scene image. To obtain a high-quality segmentation result, binarization is a significant step in character extraction process. Since character strings region of sign board has the wide variety of layout and color, it is hard to binarize this category of images by conventional global methods, which are based on statistical information. The proposed method can select a threshold by evaluating the fineness of the binarized image by using the fractal dimension. By evaluating the compactness and stability of the regions of binarized image with the fractal dimension calculated by blanket method, the proposed method achieves fine binarzation results, which are no way inferior to manually and/or using one of promising global methods known as Otsu binarizing method.
Journal
-
- The Journal of the Institute of Image Electronics Engineers of Japan
-
The Journal of the Institute of Image Electronics Engineers of Japan 39(4), 463-472, 2010-07-25
The Institute of Image Electronics Engineers of Japan
References: 23
-
1
- Dynamic thresholding of grey-level images
-
BERNSE J.
Proc. 8th Int. Conf. on Pattern Recognition, 1986, 1251-1255, 1986
Cited by (1)
-
2
- A fast adaptive method for binarization of document images
-
EIKVIL L.
Proc. 1st Int. Conf. on Document Analysis and Recognition, 1991, 435-443, 1991
Cited by (1)
-
3
- <no title>
-
NIBLACK W.
An Introduction to Digital Image Processing, 115-116, 1986
Cited by (1)
-
4
- A Binarization method for Crack Detection in a Road Surface Image with the Fractal Dimension
-
YOSHIDA H.
IAPR Conf. on Machine Vision Applications, Yokohama (2009-5), 70-73, 2009
Cited by (1)
-
5
- A Binarization Method for a Scenery Image with the Fractal Dimension
-
YOSHIDA H.
Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications, 29-36, 2009
Cited by (1)
-
6
- A New Binarization Method for a Sign Board Image with the Blanket Method
-
YOSHIDA H.
the First Joint Workshop on Pattern Recognition, IEEE, CJKPR2009, 845-848, 2009
Cited by (1)
-
7
- Text String Extraction from Scene Image Based on Edge Feature and Morphology
-
WANG Y.
Proc. the 8th IAPR Int. Workshop on Document Analysis Systems, Nara (2008-9), 323-328, 2008
Cited by (1)
-
8
- 情景中文字の検出のための動的2値化処理法 [in Japanese]
-
SHIO Akio
The Transactions of the Institute of Electronics,Information and Communication Engineers 71(5), p863-873, 1988-05
Cited by (49)
-
9
- Goal-Directed Evaluation of Binarization Methods
-
OIVIND DT
IEEE 17(12), 1191-1201, 1996
Cited by (15)
-
10
- 判別および最小2乗規準に基づく自動しきい値選定法 [in Japanese]
-
大津 展之
The Transactions of the Institute of Electronics and Communication Engineers of Japan 63(4), p349-356, 1980-04
Cited by (370)
-
11
- Automatic thresholding of gray-level pictures using two-dimensional entropy
-
ABUTALEB AS
Comput. Vis. Graph. Image Process. 47, 22-32, 1989
Cited by (3)
-
12
- A New Method for Gray-Level Picture Thresholding Using the Entropy of the Histgram
-
KAPUR J.
CVGIP 29, 273-285, 1985
Cited by (15)
-
13
- Minimum Error Thresholding
-
KITTLER J.
Pattern Recognition 19, 41-47, 1986
DOI Cited by (49)
-
14
- A New Method for Image Segmentation
-
YANOWITZ S. D.
CVGIP 46, 82-95, 1989
Cited by (9)
-
15
- Gray level thresholding in badly illuminated images
-
PARKER J. R.
IEEE Pattern Anal. Machine Intell. 13(8), 813-819, 1991
Cited by (5)
-
16
- Improvement of "Integrated Function Algorithm" for binarization of document images
-
TRIER Oivind Due
Pattern Recognition Letters 16, 277-283, 1995
Cited by (4)
-
17
- A threshold selection method based on the mean of adjacent number in binary images [in Japanese]
-
SASAKAWA K. , 黒田 伸一 , 池端 重樹
The Transactions of the Institute of Electronics,Information and Communication Engineers. 73(3), p360-366, 1990-03
Cited by (24)
-
18
- 2値画像の複雑さと多値画像の閾値処理に関する考察 [in Japanese]
-
谷口 倫一郎 , 河口 英二
The Transactions of the Institute of Electronics,Information and Communication Engineers 70(1), p164-173, 1987-01
Cited by (25)
-
19
- A Thresholding Technique based on Maximum Frequency of Edge Strength [in Japanese]
-
SAITOH Fumihiko
The Journal of the Institute of Image Electronics Engineers of Japan 31(3), 378-387, 2002-05-25
J-STAGE References (13) Cited by (1)
-
20
- Multiple Resolution Texture Analysis and Classification
-
PELEG S.
IEEE Trans. Patt. Anal. Machine Intell. 6(6), 518-523, 1984
Cited by (38)
-
21
- A Gray-Level Image Thresholding Method [in Japanese]
-
GUO Sumei , OZAWA Shinji
The Transactions of the Institute of Electronics,Information and Communication Engineers. 00080(00001), 183-189, 1997-01-25
References (5) Cited by (11)
-
22
- Optimum Estimation of Local Fractal Dimension Based on the Blanket Method
-
Sonny Novianto , Yukinori Suzuki , Junji Maeda
Transactions of Information Processing Society of Japan 43(3), 825-828, 2002-03-15
IPSJ References (12) Cited by (13)
-
23
- Contour Extraction using Texture and Non-Texture Distinction [in Japanese]
-
IGUCHI Shigeru , YAMANA Hayato
IEICE technical report 105(414), 13-18, 2005-11-17
References (11) Cited by (1)
Cited by: 1
-
1
- An Improvement of Compression Algorithm of Weather Chart Suited for Transmission by Voice Band of Fisheries Radio [in Japanese]
-
Yoshida Hiromi , Uemura Hiroyuki , Tanaka Naoki , Inoue Takeshi
The Journal of Japan Institute of Navigation 124(0), 345-353, 2011
J-STAGE References (11)