• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.7
• /
• pp.2390-2406
• /
• 2022

Scene text recognition has important application value and attracted the interest of plenty of researchers. At present, many methods have achieved good results, but most of the existing approaches attempt to improve the performance of scene text recognition from the image level. They have a good effect on reading regular scene texts. However, there are still many obstacles to recognizing text on low-quality images such as curved, occlusion, and blur. This exacerbates the difficulty of feature extraction because the image quality is uneven. In addition, the results of model testing are highly dependent on training data, so there is still room for improvement in scene text recognition methods. In this work, we present a natural scene text recognizer to improve the recognition performance from the feature level, which contains feature representation and feature enhancement. In terms of feature representation, we propose an efficient feature extractor combined with Representative Batch Normalization and ResNet. It reduces the dependence of the model on training data and improves the feature representation ability of different instances. In terms of feature enhancement, we use a feature enhancement network to expand the receptive field of feature maps, so that feature maps contain rich feature information. Enhanced feature representation capability helps to improve the recognition performance of the model. We conducted experiments on 7 benchmarks, which shows that this method is highly competitive in recognizing both regular and irregular texts. The method achieved top1 recognition accuracy on four benchmarks of IC03, IC13, IC15, and SVTP.

• Physical Therapy Korea
• /
• v.31 no.2
• /
• pp.91-103
• /
• 2024

Stroke is one of the most common disabilities experienced by the elderly in the community. However, stroke progresses to a chronic level, patients are discharged from medical institutions and eventually no longer receive therapeutic interventions at home. In this systematic review, we compared home-based rehabilitation (HBR) with comparison for patients with stroke. Literature published in Cumulative Index for Nursing and Allied Health Literature (CINAHL), Embase, Physiotherapy Evidence Database (PEDro), PubMed, and Google Scholar were reviewed. A total of 1,158 studies were initially retrieved. After reading the full texts, 11 articles were included in the systematic review. Quality assessment of the included studies was conducted using Risk of Bias (RoB) 2.0, and Egger's regression test was used to evaluate publication bias. Data analysis was performed using the R studio software (R Studio). According to the quality assessment using RoB 2.0, three studies were evaluated as low risk, two as of some concern, and three as high risk. The overall effect size was moderate (0.309). The value of the balance function was a small effect size (0.201), while the value of the gait function was a moderate effect size (0.353). The values were small and moderate effect (0.154, 0.411) for the chronic and subacute conditions, respectively. According to the Egger's regression test, no publication bias was observed. The findings of this study indicate that HBR resulted in the greatest improvement in gait function in patients with subacute stroke compared to those with chronic stroke. Therefore, the application of this intervention to patients with stroke in the community is recommended.

• The Journal of the Korea Contents Association
• /
• v.5 no.4
• /
• pp.150-159
• /
• 2005

With the increase of interest in ubiquitous computing, image code is attracting attention in various areas. Image code is important in ubiquitous computing in that it can complement or replace RFID (radio frequency identification) in quite a few areas as well as it is more economical. However, because of the difficulty in reading precise colors due to the severe distortion of colors, its application is quite restricted by far. In this paper, we present an efficient method of image code recognition including automatically locating the image code using the hue and saturation values. In our experiments, we use an image code whose design seems most practical among currently commercialized ones. This image code uses six safe colors, i.e., R, G, B, C, M, and Y. We tested for 72 true-color field images with the size of $2464{\times}1632$ pixels. With the color calibration based on the histogram, the localization accuracy was about 96%, and the accuracy of color classification for localized codes was about 91.28%. It took approximately 5 seconds to locate and recognize the image code on a PC with 2 GHz P4 CPU.

• Journal of the Korean Society of Food Culture
• /
• v.32 no.3
• /
• pp.185-203
• /
• 2017

This research analyzed consumer culture and usage of sugar in modern times based on 12 modern popular Korean cooking books with sugar recipes. Procedures were formed via textual analysis. The outcomes of the study can be summarized in brief statements. According to "Banchandeungsok", "Booinpilj", and "Chosunmoossangsinsikyorijebeob", sugar was utilized in 34 out of 663 or 5.1% of cooked foods during the 1910s to 1920s. According to books such as "Ganpyounchosunyorijebeob", "Ililhwalyongsinyoungyangyoribeob", "Chosun's cooking of the four seasons", "Halpaengyoungoo", "Chosunyorijebeob", and "Required reading for housewife", sugar was added to 165 out of 998 or 16.5% of cooked foods during the 1930s. According to the books like "Chosunyorihak", "Chosunyoribeob", and "Woorieumsik", sugar was an ingredient in 241 out of 756 or 31.9% of cooked foods during the 1940s. Sugar depicted within the 12 modern popular Korean cooking books primarily functioned as an alternative sweetener, starch, sweet enhancer, preservative, and seasoning. Similar to illustrated sugar from modern popular Korean cooking books, sugar has continually been favored by Korean cooks starting from the 1910s with 5.1% usage, the 1930s with 16.5% usage, and the 1940s with 31.9% usage. Despite its short history, sugar's culinary importance in Korea has been on the rise ever since the early 1900s. Although sugar is an exotic spice in Korea, it has gained social, cultural, and symbolic recognition as well as practicality within Korean food culture. Thus, it has become more internalized and familiarized as an inseparable sweetness that characterizes current Korean food.

• The Journal of the Korea Contents Association
• /
• v.13 no.3
• /
• pp.485-494
• /
• 2013

In this paper, both content analysis and Romey analysis were employed to analyze 'Basic Information Technology' textbooks in technical high schools. Not only was the content analysis employed to study the organization and contents of the textbooks, but also the Romey analysis was applied so as to determine inquisitive tendency on four factors - text, data, activity and evaluation. The results from content analysis showed that the total number of pages, reading material, the number of concepts introduced in each section, and the use of application software demonstrated a discrepancy among textbooks. Also, Romey analysis showed that while the section 'text' was written with relatively low inquiry tendency for its authoritative tone, two textbooks showed high inquisitive tendency for section 'activity'. Finally, based on the obtained results, the ways to improve 'Basic Information Technology' textbooks were proposed.

• Journal of Korea Game Society
• /
• v.14 no.5
• /
• pp.15-24
• /
• 2014

Semiotics studies the structures and systems of all signs related to human life, thus being capable of analyzing games. "Playing games" can be deemed as an act of reading or interpreting games semiotically, which makes game producers "senders," games "texts," gamers "receivers," and gamers playing games "contexts." Since most games are in the multi-variable narrative format, however, it is not a frequent case that gamers interpret games in the ways intended by producers. This study thus set out to analyze and interpret the console game (2001) remembered as the same evaluation by many gamers in the way intended by the producer. For analysis, the study defined its story program by analyzing the plot and sequence. For semantic analysis, the study applied the Actor Model and the Semiotic Square Model to interpret . The process identified such codes as confrontation, assistance, collaboration, and control and confirmed that Ico and Yorda, non-subject characters, were transforming into subject ones. That is, tells a story of the main characters that used to lead a non-subject life earning lives of their own.

• The KIPS Transactions:PartD
• /
• v.9D no.3
• /
• pp.409-416
• /
• 2002

The reuse by using framework is proposed to improve productivity. It is a set of usable and expandable classes and their connectivity. But frameworks are described with programming languages, it is hard for developers to learn the collaborative patterns of a framework by reading it. Patterns are one approach to improving the documentation. But this should be redesigned to expand and redefine the framework. The necessity of the formal description of architecture information is being proposed to relate to programming language. This paper support the following points. First of all, it has been proposed the description of the needed elements when developing a framework using ADL. Secondly, the current development process has been refined to be suitable for developing the domain framework. Thirdly, it has been proposed the development of a application using a framework implemented by an architecture information described with ADL. Overalls, the main contents of this research is defining the ideas of a description and development of an architecture framework using ADL.

• Publications of The Korean Astronomical Society
• /
• v.34 no.1
• /
• pp.1-16
• /
• 2019

"Hon-Chon-Jeon-Do" is a woodcut star map with the size of $79.4cm{\times}127.5cm$, and was widely disseminated as it was made into a set with Kim, Jung Ho's "Yeoji-Jeon-Do". This study confirmed that Yixiang kaocheng xubian ("의상고성속편") star catalogue was used as a source to produce the star map, and the stereographic projection was applied with the projection center being the mid-point (Q) between the celestial and ecliptic north poles. The 'mid-circle' around the Q is arisen between the equator and the ecliptic, and on this circle, the hour angle and the ecliptic longitude of a star can be marked using the same scale. This means that the hour of the day and the season of the year can be read on the same dial of the mid-circle, and the application of this character in the practical use was the key point of the star map production. By observing either transits or positions of the 28 xiu (宿), it is easy to find the corresponding season and time by simply reading the dial on the mid-circle. This is just the function of a portable almanac and thus by disseminating it widely, the convenience of the people would have been promoted. For this reason, it can be stated that "Hon-Chon-Jeon-Do" was a practical astronomical tool which was produced by the western astronomical projection method and was used to find time and season. Choi, Han Ki and Kim, Jung Ho are strong candidates for the makers of this star map. The time of production is estimated to be 1848 ~ 1857, and "Hon-Chon-Jeon-Do" could be regarded as a good contributor to popularization of astronomy in the late Joseon Dynasty.

• Fisheries and Aquatic Sciences
• /
• v.27 no.6
• /
• pp.346-355
• /
• 2024

Groupers are economically important species in the fishery and aquaculture industries in Asian countries. Various species of grouper, including hybrids, have been brought to market even without precise species identification. In this study, we analyzed the structure and expression profile of the gene encoding prolactin (PRL) in the red-spotted grouper Epinephelus akaara based on genomic DNA and cDNA templates. The results showed that the PRL gene consists of five exons encoding an open reading frame of 212 amino acids, including a putative signal peptide of 24 amino acids and a mature structural protein of 188 amino acids. It showed amino acid identities of 99% with Epinephelus coioides, 83% with Amphiprion melanopus, 82% with Acanthopagrus schlegelii, 75% with Oreochromis niloticus, 70% with Coregonus autumnalis, and 67% with Oncorhynchus mykiss, indicating its closer similarity to E. coioides and other groupers but marked distinction from non-teleost PRLs. PRL mRNA expression was detected mostly in the brain, including the pituitary gland, with little expression in other tissues. While the 5-exon structure of the PRL gene of red-spotted grouper and the exon sizes were conserved, the sizes of the introns, particularly the first intron, were markedly different among the grouper species. To examine whether these differences can be used to distinguish groupers of similar phenotypes, exon-primed intron-crossing analysis was carried out for various commercially important grouper species. The results showed clear differences in size of the amplified fragment encompassing the first intron of the PRL gene, indicating that this method could be used to develop species-specific markers capable of discriminating between grouper species and their hybrids at the molecular level.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.21 no.1
• /
• pp.105-122
• /
• 2022

According to the statistics about the fatal crashes that have occurred on the expressways for the last 5 years, those who died on the shoulders of the road has been as 3 times high as the others who died on the expressways. It suggests that the crashes on the shoulders of the road should be fatal, and that it would be important to prevent the traffic crashes by cracking down on the vehicles intruding the shoulders of the road. Therefore, this study proposed a method to detect a vehicle that violates the shoulder lane by using the Faster R-CNN. The vehicle was detected based on the Faster R-CNN, and an additional reading module was configured to determine whether there was a shoulder violation. For experiments and evaluations, GTAV, a simulation game that can reproduce situations similar to the real world, was used. 1,800 images of training data and 800 evaluation data were processed and generated, and the performance according to the change of the threshold value was measured in ZFNet and VGG16. As a result, the detection rate of ZFNet was 99.2% based on Threshold 0.8 and VGG16 93.9% based on Threshold 0.7, and the average detection speed for each model was 0.0468 seconds for ZFNet and 0.16 seconds for VGG16, so the detection rate of ZFNet was about 7% higher. The speed was also confirmed to be about 3.4 times faster. These results show that even in a relatively uncomplicated network, it is possible to detect a vehicle that violates the shoulder lane at a high speed without pre-processing the input image. It suggests that this algorithm can be used to detect violations of designated lanes if sufficient training datasets based on actual video data are obtained.