• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.8
• /
• pp.1941-1947
• /
• 2014

Document image binarization is largely used as previous stage of document recognition. And the result of document recognition is much affected from the result of document image binarization. There were many studies to binarize document images. The results of previous studies for document image binarization is varied according to the state of document images. In this paper, we propose a technique for document image binarization using MSER that is applied to extract objects from an image. At first, raw MSER objects are extracted from a document image. Because the raw MSER objects cannot be used for document image binarization, the extracted raw MSER objects are modified. Then the final MSER objects are used for document image binarization with the contrast image that is extracted from the document image. Experimental results show that the proposed technique is useful for document image binarization.

• The Journal of Information Technology
• /
• v.2 no.2
• /
• pp.89-105
• /
• 1999

In this paper, we present a countermeasure for a various trouble occurred in secure communication of document image. We Propose a security system for transmission of document image using mixing algorithm that the third party cannot conceive secure transmission of information instead of existing scheme which depend on crypto-degree of security algorithm, itself. For this, RM, DM and RDM algorithm for mixing of secure bits are proposed and applied to digital signature for mixing for secure document and mixing for non-secure document by secure document. Security system for document image involves not only security scheme for document image transmission itself, but also digital signature scheme. The transmitter embeds secretly the signatures onto secure document, embeds it to non-secure document and transfers it to the receiver. The receiver makes a check of any forgery on the signature and the document. Because the total amount of transmitted data and the image quality are about the same to those of the original document image, respectively, the third party cannot notice the fact that signatures and secure document are embedded on the document image. Thus, the probability of attack will be reduced.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.8
• /
• pp.1869-1876
• /
• 2009

Scanner which is a kind of optical devices is used to convert paper documents into document image files. The assessment of scanned document image is performed to check if there are any modification on document image files in scanning process. In assessment of scanned documents, user checks the degree of skew, noise, folded state and etc This paper proposed to how to measure geometrical modifications of document image in scanning process. In this study, we check the degree of modification in document image file by image processing and we compare the evaluation value which means the degree of modification in each items with OCR success ratio in a document image file. To analyse the correlation between OCR success ratio and the evaluation value which means the degree of modification in each items, we apply Pearson Correlation Coefficient and calculate weight value for each items to score total evaluation value of image modification degrees on a image file. The document image which has high rating score by proposed method also has high OCR success ratio.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.1
• /
• pp.123-132
• /
• 1993

This paper presents the segmentation of sentence area and diagram area from docwnent image. For extracting the sentence area, we perform the Dilation, basic operation of Morphology, to the document image and obtain the smeared document image. After the smeared docwnent image is blocked, we determine the writing form by the vertical and horizontal characteristics of the document image and calculate the skew from it. And then, we relocate the document image and extract the chatacter string from the relocated docwnent. 11 document images of three classes are considered and the character string has been well extracting from 11 document images.

• The KIPS Transactions:PartB
• /
• v.17B no.2
• /
• pp.125-138
• /
• 2010

The purpose of this research is to enhance document recognition system which is essential for developing full-text retrieval system of the document image data stored in the digital library of a public institution. To achieve this purpose, the main tasks of this research are: 1) analyzing the document image data and then developing its image preprocessing technology and document structure analysis one, 2) building its specialized knowledge base consisting of document layout and property, character model and word dictionary, respectively. In addition, developing the management tool of this knowledge base, the document recognition system is able to handle the various types of the document image data. Currently, we developed the prototype system of document recognition which is combined with the specialized knowledge base and the library of document structure analysis, respectively, adapted for the document image data housed in National Archives of Korea. With the results of this research, we plan to build up the test-bed and estimate the performance of document recognition system to maximize the utilization of full-text retrieval system.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2013.05a
• /
• pp.379-381
• /
• 2013

In this paper, we propose a method to detect the gradient of the spaces between sentences and to deskew in the document image. First, gradient is measured by pixels for spaces between sentences that has been done an edge extraction in document image and then skewed image is corrected by using the value of the gradient which has been measured. Since document image is divided into several areas, it shows a robust processing result by handling the margin, images, and multistage form in the document. Because the proposed method does not use pixel of the character region but use the blank area, degraded document image as well as vivid document image is effectively corrected than conventional method.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.1
• /
• pp.201-207
• /
• 2015

The smartphone with camera can easily generate an image instead of a scanner. However the document image through a smartphone can have distortions related rotation or perspective. In this paper, we proposed a method to generate the document image in that distortions are reduced from the captured document image through a smartphone. For this, the original document image through a smartphone is preprocessed using the MSER-b technique to reduce the light effect. Then, the text area contour is extracted using the characteristics of the document image. Lastly, rotation or perspective distortions are reduced using the extracted text area contour. For experiments, the proposed method is compared two other products. Through experiments, we show that the distortions within the captured document image through smartphone can be effectively reduced.

• Journal of Information Processing Systems
• /
• v.10 no.1
• /
• pp.55-68
• /
• 2014

Automatic segmentation of foreground text from the background in degraded document images is very much essential for the smooth reading of the document content and recognition tasks by machine. In this paper, we present a novel approach to the binarization of degraded document images. The proposed method uses a new local contrast feature extracted based on the stroke width of text. First, a pre-processing method is carried out for noise removal. Text boundary detection is then performed on the image constructed from the contrast feature. Then local estimation follows to extract text from the background. Finally, a refinement procedure is applied to the binarized image as a post-processing step to improve the quality of the final results. Experiments and comparisons of extracting text from degraded handwriting and machine-printed document image against some well-known binarization algorithms demonstrate the effectiveness of the proposed method.

• Journal of the Institute of Convergence Signal Processing
• /
• v.11 no.3
• /
• pp.215-220
• /
• 2010

In this paper, we propose a new texture-based page segmentation and classification method in which table region, background region, image region and text region in a given document image are automatically identified. The proposed method for document images consists of two stages, document segmentation and contents classification. In the first stage, we segment the document image, and then, we classify contents of document in the second stage. The proposed classification method is based on a texture analysis. Each contents in the document are considered as regions with different textures. Thus the problem of classification contents of document can be posed as a texture segmentation and analysis problem. Two-dimensional Gabor filters are used to extract texture features for each of these regions. Our method does not assume any a priori knowledge about content or language of the document. As we can see experiment results, our method gives good performance in document segmentation and contents classification. The proposed system is expected to apply such as multimedia data searching, real-time image processing.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.5
• /
• pp.1778-1797
• /
• 2021

Document image binarization is an important pre-processing step in document analysis and archiving. The state-of-the-art models for document image binarization are variants of encoder-decoder architectures, such as FCN (fully convolutional network) and U-Net. Despite their success, they still suffer from three limitations: (1) reduced feature map resolution due to consecutive strided pooling or convolutions, (2) multiple scales of target objects, and (3) reduced localization accuracy due to the built-in invariance of deep convolutional neural networks (DCNNs). To overcome these three challenges, we propose an improved semantic segmentation model, referred to as DP-LinkNet, which adopts the D-LinkNet architecture as its backbone, with the proposed hybrid dilated convolution (HDC) and spatial pyramid pooling (SPP) modules between the encoder and the decoder. Extensive experiments are conducted on recent document image binarization competition (DIBCO) and handwritten document image binarization competition (H-DIBCO) benchmark datasets. Results show that our proposed DP-LinkNet outperforms other state-of-the-art techniques by a large margin. Our implementation and the pre-trained models are available at https://github.com/beargolden/DP-LinkNet.