• Journal of Intelligence and Information Systems
• /
• v.8 no.1
• /
• pp.27-39
• /
• 2002

A locating text technique capable of locating and extracting text blocks in various Web images is presented here. Until now this area of work has been ignored by researchers even if this sort of text may be meaningful for internet users. The algorithms associated with the technique work without prior knowledge of the text orientation, size or font. In the work presented in this research, our text extraction algorithm utilizes useful edge detection followed by histogram analysis on the genuine characteristics of letters defined by text clustering region, to properly perform extraction of the text region that does not depend on font styles and sizes. By a number of experiments we have showed impressively acceptable results.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.5 no.4
• /
• pp.65-74
• /
• 2002

The purpose of this study is to develop an XML based GUI that can search for satellite image information which is converted to XML data format and stored in the database server on the web, and modify and reuse data. In order to implement these functions efficiently, we used a DOM interface of XML that increases the efficiency of accessing the document structure. We used HTML and Java script programming to facilitate this interface. The system was applied to the management system of satellite images in the Research Institute of SFC at Keio University. Our results confirmed the technical functionalities.

• Journal of Korea Society of Industrial Information Systems
• /
• v.6 no.2
• /
• pp.16-22
• /
• 2001

In this paper we mose a new scheme for recognition of deformed letters by extracting feature vectors based on Gibbs distributions which are well suited for representing the spatial continuity. The extracted feature vectors are comprised of 2-D conditional moments which are invariant under translation, rotation, and scale of an image. The Algorithm for pattern recognition of deformed letters contains two parts: the extraction of feature vector and the recognition process. (i) We extract feature vector which consists of an improved 2-D conditional moments on the basis of estimated conditional Gibbs distribution for an image. (ii) In the recognition phase, the minimization of the discrimination cost function for a deformed letters determines the corresponding template pattern. In order to evaluate the performance of the proposed scheme, recognition experiments with a generated document was conducted. on Workstation. Experiment results reveal that the proposed scheme has high recognition rate over 96%.

• Ocean Systems Engineering
• /
• v.13 no.2
• /
• pp.195-224
• /
• 2023

This paper proposes a method for the acoustic imaging wherein the traditional requirement of the relative movement between the transmitter and target is overcome. This is facilitated through the beamforming acoustic array in the transmitter, in which the target is illuminated by the array at various azimuth and elevation angles without the physical movement of the acoustic array. The concept of beam steering of the acoustic array facilitates the formation of the beam at desired angular positions of azimuth and elevation angles. This paper substantiates that the combination of illumination of the target from different azimuth and elevation angles with respect to the transmitter (through the beam steering of beam forming acoustic array) and the beam steering at multiple frequencies (through SF) results in enhanced reconstruction of images of the target in the underwater scenario. This paper also demonstrates the possibility of reconstruction of the image of a target in underwater without invoking the traditional algorithms of Digital Image Processing (DIP). This paper comprehensively and succinctly presents all the empirical formulae required for modelling the acoustic medium and the target to facilitate the reader with a comprehensive summary document incorporating the various parameters of multi-disciplinary nature.

• International Journal of Contents
• /
• v.14 no.1
• /
• pp.1-6
• /
• 2018

Text, as one of the most influential inventions of humanity, has played an important role in human life since ancient times. The rich and precise information embodied in text is very useful in a wide range of vision-based applications such as the text data extracted from images that can provide information for automatic annotation, indexing, language translation, and the assistance systems for impaired persons. Therefore, natural-scene text detection with active research topics regarding computer vision and document analysis is very important. Previous methods have poor performances due to numerous false-positive and true-negative regions. In this paper, a fully-convolutional-network (FCN)-based method that uses supervised architecture is used to localize textual regions. The model was trained directly using images wherein pixel values were used as inputs and binary ground truth was used as label. The method was evaluated using ICDAR-2013 dataset and proved to be comparable to other feature-based methods. It could expedite research on text detection using deep-learning based approach in the future.

• Journal of Veterinary Clinics
• /
• v.18 no.1
• /
• pp.35-43
• /
• 2001

Present study was undertaken in order to document early renal ultrasonographic changes of gentamicin nephrotoxicosis and to show the value of renal ultrasonography as a contributory means of early diagnosis of acute renal failure in dogs. The experimental design was a randomized complete block design with six treatments in two blocks (gentamicin-treated & saline-treated). Acute renal failure was induced by toxic dosage of gentamicin (30 mg/kg) and saline solution sham equivalent in volume to that of the toxic dosage of gentamicin (1.5-3ml). Subjective visualization of increased renal cortex was visible as homogenous echoes that were hypoechoic relative to the surrounding tissues, whereas the renal medulla was anechoic to slightly hypoechoic. After treatment, the renal cortex was hyperechoic relative to the surrounding tissue. Increased renal cortex echogenicity was associated with significant nephrotoxicosis and was superior to serum creatinine elevation in nephrotoxicosis detection. Urine GGT was superior to other clinicopathological data utilized in the diagnosis of nephrotoxicosis. Based on the above results, increased renal cortex echogenicity seemed to be of use in detecting of acute renal failure.

• The Journal of the Korea Contents Association
• /
• v.3 no.1
• /
• pp.14-20
• /
• 2003

In this paper efficient skew estimation and correction approaches are proposed. To detect the skew of text images, Hough transform using the perpendicular angle view property and shape moments are peformed. The resultant primary text skew angle is used to align the original text. The performance evaluations of the proposed methods with respect to running time are shown.

• The Journal of Society for e-Business Studies
• /
• v.1 no.1
• /
• pp.159-171
• /
• 1996

Nowadays, multimedia and Hypermedia become hot topics in information industry. Due to high capacity of media storage and fast communication network, it is possible to exchange text data as well as image, moving picture and voice. Especially to apply hypermedia under CALS standard environment, the relation between international standard and CALS standard needs to be considered. This study introduces conceptual background and processing model of HyTime (Hypermedia Time-based Structuring Language) which is a specification of hypermedia exchange, Hyper ODA (Hyper Open Document Architecture) which is a major multimedia communication basis, MMCF (Multimedia Communication Forum), AHM(Amsterdam Hypermedia Model), and DSRM(DAVIC System Reference Model) reference model which helps determination of hypermedia communication specification Although they are international standard, provisional standard or non-standard, it discusses the Possibility of adopting them as CALS standard. Hence, this paper chooses the best recommend for CALS among these models.

• Journal of the Korean Graphic Arts Communication Society
• /
• v.22 no.1
• /
• pp.65-74
• /
• 2004

This study analyzed the main bodies, sizes of litters, spaces between letters, Words, lines, length of lines, sizes of main body and margin, Position of the main bodies of PDF(Portable Document Format) in high schools and "the criteria of reading system for the textbooks" which was suggested by the Department of Education and Provided some suggestions for the improvement. First, it is judged that spaces of the main body of textbooks should be shrunken to make the ratio of 50:50 between the main body and space margin. Second, it is judged that the main body should be placed in the center of text page. Third, the results of the above findings showed that it would be difficult to deal with the task of deciding space and position or the main body and marginal space in a form of single project, because there are many variables involved and psychologically complicated elements which can not be measured objectively.

• IEIE Transactions on Smart Processing and Computing
• /
• v.6 no.1
• /
• pp.53-59
• /
• 2017

Optical character recognition (OCR) automatically recognizes text in an image. OCR is still a challenging problem in computer vision. A successful solution to OCR has important device applications, such as text-to-speech conversion and automatic document classification. In this work, we analyze character recognition performance using the current state-of-the-art deep-learning structures. One is the AlexNet structure, another is the LeNet structure, and the other one is the SPNet structure. For this, we have built our own dataset that contains digits and upper- and lower-case characters. We experiment in the presence of salt-and-pepper noise or Gaussian noise, and report the performance comparison in terms of recognition error. Experimental results indicate by five-fold cross-validation that the SPNet structure (our approach) outperforms AlexNet and LeNet in recognition error.