• The KIPS Transactions:PartB
• /
• v.10B no.4
• /
• pp.429-442
• /
• 2003

In this paper, we describe an address image recognition system for automatic processing of standard- size letter mail. The inputs to the system are gray-level mail piece images and the outputs are delivery point codes with which a delivery sequence of carrier can be generated. The system includes five main modules; destination address block location, text line separation, character segmentation, character recognition and finally address interpretation. The destination address block is extracted on the basis of experimental knowledge and the line separation and character segmentation is done through the analysis of connected components and vortical runs. For recognizing characters, we developed MLP-based recognizers and dynamical programming technique for interpretation. Since each module has been implemented in an independent way, the system has a benefit that the optimization of each module is relatively easy. We have done the experiment with live mail piece images directly sampled from mail sorting machine in Yuseong post office. The experimental results prove the feasibility of our system.

• The KIPS Transactions:PartC
• /
• v.8C no.5
• /
• pp.525-534
• /
• 2001

Generally, compression and encryption procedures are performed independently in lossless image compression and encryption. When compression is followed by encryption, the compressed-stream should have the property of randomness because its entropy is decreased during the compression. However, when full data is compressed using image compression methods and then encrypted by encryption algorithms, real-time processing is unrealistic due to the time delay involved. In this paper, we propose to combine compression and encryption to reduce the overall processing time. It is method decomposing gray-scale image by means of quadtree compression algorithms and encrypting the structural part. Moreover, the lossless compression ratio can be increased using a transform that provides an decorrelated image and homogeneous region, and the encryption security can be improved using a reconstruction of the unencrypted quadtree data at each level. We confirmed the increased compression ratio, improved encryption security, and real-time processing by using computer simulations.

• Journal of Information Processing Systems
• /
• v.9 no.1
• /
• pp.117-140
• /
• 2013

Hindi is the most widely spoken language in India, with more than 300 million speakers. As there is no separation between the characters of texts written in Hindi as there is in English, the Optical Character Recognition (OCR) systems developed for the Hindi language carry a very poor recognition rate. In this paper we propose an OCR for printed Hindi text in Devanagari script, using Artificial Neural Network (ANN), which improves its efficiency. One of the major reasons for the poor recognition rate is error in character segmentation. The presence of touching characters in the scanned documents further complicates the segmentation process, creating a major problem when designing an effective character segmentation technique. Preprocessing, character segmentation, feature extraction, and finally, classification and recognition are the major steps which are followed by a general OCR. The preprocessing tasks considered in the paper are conversion of gray scaled images to binary images, image rectification, and segmentation of the document's textual contents into paragraphs, lines, words, and then at the level of basic symbols. The basic symbols, obtained as the fundamental unit from the segmentation process, are recognized by the neural classifier. In this work, three feature extraction techniques-: histogram of projection based on mean distance, histogram of projection based on pixel value, and vertical zero crossing, have been used to improve the rate of recognition. These feature extraction techniques are powerful enough to extract features of even distorted characters/symbols. For development of the neural classifier, a back-propagation neural network with two hidden layers is used. The classifier is trained and tested for printed Hindi texts. A performance of approximately 90% correct recognition rate is achieved.

• Journal of Fashion Business
• /
• v.5 no.5
• /
• pp.87-96
• /
• 2001

The purposes of this study is to supply information on the most common and generalized purchasing conditions of suits in China by surveying male customers from Beijing, Shanghai, Dalian, Tianjin, Harbin and Guangzhou with ages from 20's to 40's. Statistic data from the questionnaire was analyzed by SPSS 10.0 for windows and the analyzing method was processed by descriptive statistics, cross analysis and dispersed analysis and for verifying similarities, t-test and $\chi^2$-test were used. The results could be summarized as follows; First, Purchasing places for suits were stores specializing in men's wear, department stores and reasons for purchasing suits were 'For business' and 'Change of season'. Two, for purchasing information provider, television worked as the key source of providing purchasing information and outdoor advertisement, for main influence of purchasing, the consumers answered that they are the key decision makers and replied their wife or girlfriend's opinion as the answer. Three, for number of suits in possession, answered three suits followed by more than five suits and for the purchasing period, the consumers chose one suit per year or every 6month. Four, according to the questionnaire on prices, colors, fabrics and styles of suits they buy, the consumers pay between 1,000 and 1,999RMB per suit and for the colors, black, dark gray and deep blue. For the choice of fabrics, said pure wool. For the styles of the suits, answered single three buttons. Five, the result from a question on the level of satisfaction shows 'satisfied', 'average' and 'very satisfied' with the purchase.

• The Korean Journal of Pain
• /
• v.11 no.2
• /
• pp.182-193
• /
• 1998

Background: Ciliary neurotrophic factor (CNTF), identified as a survival factor for developing peripheral neurons is upregulated by reactive astrocytes in the traumatized tissue and in areas of terminal degeneration after a brain lesion. But in the spinal cord, CNTF is expressed in the non-astrocytic phenotypic, maybe oligodendrocytes. The present study was undertaken to determine the upregulation of CNTF expression in reactive astrocytes following spinal cord lesion in the rat. Methods: Unilateral incision of the dorsal funiculus at the thoracic level was performed and rats were sacrificed on days 3, 7, 14 and 28 postlesion. Western blot analysis, immunocytochemical analysis and double immunofluorescence for CNTF and glial fibrillary acidic protein (GFAP) were performed after spinal cord lesion. Results: A major band with 24 kDa and additional band of higher molecular weight form were detectable, and the intensity of the 24 kDa immunoreactive band increased up to 14 days postlesion and decreased toward laminectomized control values. CNTF immunoreactivity was markedly upregulated in the injured dorsal funiculus and adjacent gray matter. The time course of CNTF expression is coincident with the appearance of reactive astrocytes in the injured spinal cord. Moreover, double immunofluorescence for CNTF and glial fibrillary acidic protein (GFAP) revealed that CNTF immunoreactivity was in GFAP immunoreactive astrocytes. Conclusions: These results show that CNTF upregulation occurred in reactive astrocytes following spinal cord lesion, and suggest a role for CNTF in the regulation of astrocytic responses after spinal cord injury.

• Journal of Korea Multimedia Society
• /
• v.7 no.4
• /
• pp.451-458
• /
• 2004

A leather quality inspection by naked eyes has known as unreliable because of its biological characteristics like accumulated fatigue caused from an optical illusion and biological phenomenon. Therefore it is necessary to automate the leather quality inspection by computer vision technique. In this paper, we present automatic leather qua1ity classification system get information from leather surface. Leather is usually graded by its information such as texture density, types and distribution of defects. The presented algorithm explain how we analyze leather information like texture density and defects from the gray-level images obtained by digital camera. The density data is computed by its ratio of distribution area, width, and height of Fourier spectrum magnitude. And the defect information of leather surface can be obtained by histogram distribution of pixels which is Windowed from preprocessed images. The information for entire leather could be a standard for grading leather quality. The proposed leather inspection system using machine vision can also be applied to another field to substitute human eye inspection.

• Journal of Korea Multimedia Society
• /
• v.17 no.3
• /
• pp.277-286
• /
• 2014

In this paper, we propose a texture feature-based language identification by fusion of Gabor, MDLC (multi-lag directional local correlation), and co-occurrence features. In the proposed method, for a test image, Gabor magnitude images are first obtained by Gabor transform followed by magnitude operator. Moments for the Gabor magniude images are then computed and vectorized. MDLC images are then obtained by MDLC operator and their moments are computed and vectorized. GLCM (gray-level co-occurrence matrix) is next calculated from the test image and co-occurrence features are computed using the GLCM, and the features are also vectorized. The three vectors of the Gabor, MDLC, and co-occurrence features are fused into a feature vector. In classification, the WPCA (whitened principal component analysis) classifier, which is usually adopted in the face identification, searches the training feature vector most similar to the test feature vector. We evaluate the performance of our method by examining averaged identification rates for a test document image DB obtained by scanning of documents with 15 languages. Experimental results show that the proposed method yields excellent language identification with rather low feature dimension for the test DB.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.4
• /
• pp.1112-1119
• /
• 1999

Digital watermarking is to embed imperceptible mark into image, video, audio and text data to prevent the illegal copy of multimedia data, arbitrary modification, and also illegal sales of the copes without agreement of copyright ownership. The DCT(discrete Cosine Transforms) transforms of original image is conducted in this research and these DCT coefficients are expanded by Fourier series expansion algorithm. In order to embed the imperceptible and robust watermark, the Fourier coefficients(lower frequency coefficients) can be calculated using sine and cosine function which have a complete orthogonal basis function, and the watermark is embedded into these coefficients, In the experiment, we can show robustness with respect to image distortion such as JPEG compression, bluring and adding uniform noise. The correlation coefficient are in the range from 0.5467 to 0.9507.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.4
• /
• pp.75-86
• /
• 2004

This paper proposes a method to speed up the document image binarization using a water flow model. The proposed method extracts the region of interest (ROI) around characters from a document image and restricts pouring water onto a 3-dimensional terrain surface of an image only within the ROI. The amount of water to be filed into a local valley is determined automatically depending on its depth and slope. The proposed method accumulates weighted water not only on the locally lowest position but also on its neighbors. Therefore, a valley is filed enough with only one try of pouring water onto the terrain surface of the ROI. Finally, the depth of each pond is adaptively thresholded for robust character segmentation, because the depth of a pond formed at a valley varies widely according to the gray-level difference between characters and backgrounds. In our experiments on real document images, the Proposed method has attained good binarization performance as well as remarkably reduced processing time compared with that of the existing method based on a water flow model.

• Journal of Korean Academy of Oral and Maxillofacial Radiology
• /
• v.20 no.2
• /
• pp.187-198
• /
• 1990

Examinations of the temporomandibular joints were performed on a 1.5 Tesla magnetic resonance (MR) system. An MR surface receiver coil 3 inch in diameter was placed on plastic frame, the patient's head being placed in the frame so that the coil was pressed against the temporal region. In taking advantage of the magnetic resonance imaging that has been studied briskly till now, author obtained the images of parasagittal and paracoronal planes about the temporomandibular joint by using MPGR (Multi-Planar Gradient Recalled), GRASS (Gradient Recalled Acquisition in the Steady State), and CSMEMP (Contiguous Slice Multiple Echo, Multi-Planar), that differ from the Spin Echo pulse sequence which the previous authors used. Five subjects with no symptoms of temporomandibular joint pain and dysfunction were studied. The plane images obtained by these methods were compared with those by Spin Echo pulse sequence. The results were as follows: 1. The optimal repetition times (TR) and echo times (TE) for T.M.J. image were; a. 400 msec and 18 msec in PMGR pulse sequence. b. 40 msec and 12 msec in GRASS pulse sequence. c. 700 msec and 30 msec in CSMEMP pulse sequence. d. 500 msec and 20 msec in Spin Echo pulse sequence. 2. When the MPGR pulse sequence was using, T2-weighted image was obtained in very short time. On the image of the paracoronal plane by GRASS pulse sequence, meniscus showed the moderate signal intensity, and the meniscus and its anteromedial, posterolateral attachments were observed definitely with gray color. 4. The signal intensity of Spin Echo pulse sequence was equal to that of CSMEMP pulse sequence, but the image by CSMEMP pulse sequence showed relatively lower level in its resolution.