• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.7 no.2
• /
• pp.91-97
• /
• 1974

Two scale characters, the width ana circuli counts of the first-year band, were used in a discriminant function analysis to see how effectively the two scale characters would separate geographical chum stocks from the western North Pacific. A total of 476 scale samples were taken from spawning adults which ascended to rivers of Hokkaido, Japan, in 1956, and Kamchatka, the U.S.S.R., in 1957. The scale characters were examined for conformity to the statistical requirements of a discriminant function. As a result of the examinations the two characters were verified to be able to be used in a discriminant function analysis that would classify chum taken on the high seas to most Probable origin. A discriminant function computed using the two characters correctly classified 78.5 percent of the Hokkaido and Kamchatka chum fish. Of the two characters the number of the circuli could alone classify fish to its origin with nearly the same probability of correct classification as the discriminant function based on the two characters can.

• Korean Journal of Plant Resources
• /
• v.8 no.3
• /
• pp.223-235
• /
• 1995

Comparative studies on external morphology and anatomical characters were conducted understand the interspecific relationships and possibly to identify the position of Hanabusaya asiatica as an endemic genus among 4 genera and 9 species in H. asiatica and its allied groups. Such morphological characteristics as root shape, leaf size, presence or absence of wings in petiole, corolla and calyx shape, and inflorescence and stigma shape were useful qualitative characters. The morphology of stem, peduncle, calyx lobe, leaf, petiole, and ovary were also good characters to be included in this identification study. In the anatomical observations in these organs, projecting degree of epidermis and number of projecting area rather than cross-sectioned internal structures were found to be valuable index for the classification among species. External morphology and anatomical characters support the position of H. asiatica as an endemic genus very well. Known as the most important external characters to taxonomists(Fedorov, 1957: Nakai, 1952: Hong, 1983), leaf arrangement, inflorescence, and corolla shape were also preven to be very useful characters to identify the genera and species in the H. asiatica and its allied groups.

• Korean Journal of Plant Taxonomy
• /
• v.43 no.3
• /
• pp.196-207
• /
• 2013

As a recent infrageneric classification of Juncus (Juncaceae), Kirschner suggested two subgenera which are distinguished by the type of inflorescence and the absence/presence of bracteole along with 10 sections which are distinguished by the characters of the leaves, the position of the inflorescence, and types of stamens and seeds. However, an intensive morphological investigation on Korean Juncus has yet to be performed. We surveyed the morphological characters of 14 taxa distributed throughout the Korean peninsula, including one North Korean taxon. The quantitative and qualitative characters of rhizomes, leaves, inflorescences, flowers, fruits, and seeds were investigated on each taxon using about 950 specimens of Korean Juncus on loan from major Korean herbaria. As a result, we provide diagnostic characters with illustrations and a key to Korean Juncus.

• Journal of the Semiconductor & Display Technology
• /
• v.20 no.3
• /
• pp.6-10
• /
• 2021

Recognition of characters printed or marked on the PCB components from images captured using cameras is an important task in PCB components inspection systems. Previous optical character recognition (OCR) of PCB components typically consists of two stages: character segmentation and classification of each segmented character. However, character segmentation often fails due to corrupted characters, low image contrast, etc. Thus, OCR without character segmentation is desirable and increasingly used via deep neural networks. Typical implementation based on deep neural nets without character segmentation includes convolutional neural network followed by recurrent neural network (RNN). However, one disadvantage of this approach is slow execution due to RNN layers. LPRNet is a segmentation-free character recognition network with excellent accuracy proved in license plate recognition. LPRNet uses a wide convolution instead of RNN, thus enabling fast inference. In this paper, LPRNet was adapted for recognizing characters printed on PCB components with fast execution and high accuracy. Initial training with synthetic images followed by fine-tuning on real text images yielded accurate recognition. This net can be further optimized on Intel CPU using OpenVINO tool kit. The optimized version of the network can be run in real-time faster than even GPU.

• Cartoon and Animation Studies
• /
• s.24
• /
• pp.155-172
• /
• 2011

The purpose of this study is to analyze the gesture features on the personalities of Navi characters in . In order to analyze the personality type of characters, the study applied the classification of Enneagram based on script of . The character features are classified according to character types, then the metaphorical character of the expression is obtained through gesture analysis in . Thus, it is possible to set up characters that fit its personalities in Contents of digital image. Also this study suggests creation of attractive characters and expression methods with gesture based personality.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2006.05a
• /
• pp.784-787
• /
• 2006

In this paper, we propose a method for nonlinear grapheme segmentation in Korean printed character type classification. The characters are subdivided into six types based on character type information. The feature vector is consist of mesh features, vertical projection features and horizontal projection features which are extracted from gray-level images. We classify characters into 6 types using Back propagation. Character segmentation regions are determined based on character type information. Then, an optimal nonlinear grapheme segmentation path is found using multi-stage graph search algorithm. As the result, a proposed methodology is proper to classify character type and to find nonlinear char segmentation paths.

• The Journal of Korean Medicine
• /
• v.16 no.2 s.30
• /
• pp.149-176
• /
• 1995

The results were as follow: 1. In classification of the virulence of medicines, it is the virulent animal that have a deadly poison and the rest is the animal of weak nor non-toxic. 2. In classification of the channel distribution, the most is the medicine that belongs to liver channel, the next are the stomach, lung, kidney and spleen channel. 3. In classification of four characters, the most parts are cool, common and warm medicine and there is a few that is hot and cooling. 4. In classification of five tastes, the most numerous tastes are sweet and salty and the next are acrid, bitter and sour tastes. 5. In classification of the medical action, there are few medicine of invigorating vital energy, tonic therapy and astringent and a great part of the medicine are regulating vital energy and blood, removing blood stasis and mass, clearing away heat-evil and eliminating sputum, calming the river to inhibit the wind-evil and pain control. 6. In classification of the application of cancer, the most numerous disease is the liver cancer and the next are stomach cancer, esophageal cancer, lung cancer, leukemia, uterine cancer,mastitis, brain tumor.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.33 no.4 s.111
• /
• pp.1-10
• /
• 2005

Recently, as urban landscape is growing in importance, urban landscape planning is being actively performed. for this purpose, classification of the urban landscape is definitely required. Therefore, this research focuses on classifying urban landscape in Daejeon metropolis by dividing the urban land use pattern. This results are as follows. 1. Urban land use pattern is divided into 20 classes. The residential, commercial and industrial areas, the old market and the bus terminal are evaluated negatively, whereas the areas of school, water reservoir, neighborhood park and train station are appreciated as being positive in landscape characters. 2. As a result of a cluster analysis, urban landscape has five different landscape types. These are: landscapes of medium diversity lacking green area, landscapes of high diversity lacking green area, landscapes rich in green area and with medium diversity, landscapes rich in green area and with high diversity, and landscapes rich in green area and with low diversity. 3. In landscape characters of beauty and harmony, landscapes rich in green area and with medium diversity are more positively evaluated than those rich in green area and with low diversity. This point should be taken into account for planning the urban landscape.

• Korean Journal of Cognitive Science
• /
• v.2 no.1
• /
• pp.33-50
• /
• 1990

Recently, neural networks have been proposed as computaional models for hard prlblems that the brain appears to solve easily. This paper proposes a hierarchical network which practically recognizes printed Hangul characters based on the various psychological stueies. This system is composed of a type classification netwotk and six recognition networks. The former clessifier input character images into one of the six thper by their overall sturcture, and the latter further classify them into character code. Extperiments with most frequently used 990 printed hangul characters conform the superiority of the propsed system. After all, neural nework approach turns out to be very reasonable through a comparison with statistical classifier and an analysis of mis-classification and generalization capability.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1992.06a
• /
• pp.71-75
• /
• 1992

This paper proposes algorithm in order to classify Korean consonant phonemes same as polosives, fricatives affricates into la sounds, glottalized sounds, aspirated sounds. This three kinds of sounds are one of distinctive characters of the Korean language which don't eist in language same as English. This is thesis on classfication of 14 Korean consonants(k, t, p, s, c, k', t', p', s', c', kh, ph, ch) as a previous stage for Korean phone recognition. As feature sets for classification, LPC cepstral analysis. The eperiments are two stages. First, using short-time speech signal analysis and Mahalanobis distance, consonant segments are detected from original speech signal, then the consonants are classified by fuzzy inference. As the results of computer simulations, the classification rate of the speech data was come to 93.75%.