• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.3
• /
• pp.740-753
• /
• 2023

Text recognition in natural scene images is a challenging problem in computer vision. The accurate identification of ship number characters can effectively improve the level of ship traffic management. However, due to the blurring caused by motion and text occlusion, the accuracy of ship number recognition is difficult to meet the actual requirements. To solve these problems, this paper proposes a dual-branch network based on the CRNN identification network. The network couples image restoration and character recognition. The CycleGAN module is used for blur restoration branch, and the Pix2pix module is used for character occlusion branch. The two are coupled to reduce the impact of image blur and occlusion. Input the recovered image into the text recognition branch to improve the recognition accuracy. After a lot of experiments, the model is robust and easy to train. Experiments on CTW datasets and real ship maps illustrate that our method can get more accurate results.

• 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.

• The Journal of the Korea Contents Association
• /
• v.17 no.1
• /
• pp.740-749
• /
• 2017

The epigraphy is a text or a picture engraved on metal or stone. One of advantages of rubbing of ancient inscription has been used in epigraphic field is simple. But the rubbing is not an optimal method in viewpoints of resolution and noise to decode the inscribed characters. In this study, we proposed a new research method that increases the possibility of a reading by reflecting the 3D characteristics of the engraved letters. The proposed techniques apply 3D scanning technology to obtain three-dimensional and high quality data of each of the letters in the epigraphy and use Ambient Occlusion visualization techniques to express the shade according to the 3D form of the letters. Research result enhances the readability of the letters that removed the damaged and worn information from the letters information of surface. This research contributes to narrow the scope of a particular letter and read to the controversial letters on the Pohang Jungsoengri Silla Stone Monument(Korea's national treasure number 318).

• Journal of the Korean Association of Oral and Maxillofacial Surgeons
• /
• v.42 no.3
• /
• pp.166-168
• /
• 2016

It can be challenging to create a safe airway in maxilla facial fracture and some skull surgeries. In this case study, the patient experienced jaw fractures that disturbed the dental occlusion and associated fracture of the base of the skull. Neither nasal nor oral intubation was possible based on the side effects of tracheotomy; therefore, submental intubation was applied successfully. The procedure and results are presented in the text.

• Journal of Sasang Constitutional Medicine
• /
• v.21 no.2
• /
• pp.48-57
• /
• 2009

1. Objective : The object of this thesis is to find the basis of Constitution assignment of Nogeun(蘆根, Phragmites communis Trin.) used to the medication for Taeyangin. Also, it is to recognize what kind of Constitution of Mogeun(茅根, Imperata cylindrica var. koenigii) is needed to be assigned to. 2. Methods "Dongyi Soose Bowon", 'SinchukBon', 'ChobonKwun', other related Sasang Constitutional Medicine(SCM) books, herbal medicine books such as "Boncho Kangmok", and etc. were compared and researched. 3. Results & Conclusions : 1) Nogeun(蘆根) lowers Qi and harmonizes the body. Because of the effect of lowering soar-Qi(氣), it used to treat the symptoms of Taeyangin's vomiting(噎膈反胃, Stomach reflux due to dysphagia-occlusion). Also Yuksoo Nogeun(逆水蘆恨, Reed growing near the river in low scream) is the basis of considering inhale-gathering Qi effect as a major indication of Nogeun. 2) The theory of classifying White Mogeun as a Tae-eumin Medicinal is based on the effect of Radiating while classifying White Mogeun as a Taeyangin Medicinal is based on the effect of treating stomach reflux(反胃). But both basis are not conclusive and thus requires further study. 3) In "Sasang Geumge Bibang", the first text ever to write only about separate medicinals and Classified Foods (食物類) of each constitution, there were examples of Nogeun and Mogeun's usage based on other pre-existing medical texts. 4) Because White Mogeun is classified as Tae-yangin medication in "Dongmu Yugo", but classified as a Taeeumin's medication in "Dongyi Soose Bowon" 'ChobonKwun' and "Sasang Geumge Bibang", there remains a controversy.

• Imaging Science in Dentistry
• /
• v.52 no.4
• /
• pp.333-342
• /
• 2022

Purpose: The goal of this systematic review was to compare the use of cone-beam computed tomography (CBCT) with that of computed tomography (CT) for volumetric evaluations of the tongue and oral cavity. Materials and Methods: A search for articles was conducted in accordance with the Preferred Reporting Items for Systematic Review and Meta-analyses guidelines. The PubMed, Scopus, ScienceDirect, and SAGE Journals databases were searched for articles published between 2011 and 2021. Articles were screened and assessed for eligibility. Screening involved checking for duplication, reading the title and abstract, and reading the full text. Results: The initial search retrieved 25,780 articles. Application of the eligibility criteria yielded 16 articles for qualitative analysis. Multiple uses of CBCT were identified. In several studies, researchers assessed the volumetric correlation between tongue and oral cavity volumes, as well as other parameters. Post-treatment volumetric evaluations of the oral cavity were also reported, and the reliability of CBCT was assessed. The use of CT resembled that of CBCT. Conclusion: CBCT has been used in the evaluation of tongue and oral cavity volumes to assess correlations between those volumes and with the upper airway. It has also been used for volumetric evaluation after surgical and nonsurgical procedures and to assess the relationships between tongue volume, tooth position, occlusion, and body mass index. Participants with obstructive sleep apnea and malocclusion have been evaluated, and the reliability of CBCT has been assessed. In the included studies, CT was utilized for similar purposes as CBCT, but its reliability was not assessed.