• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.18 no.6
• /
• pp.269-276
• /
• 2018

OCR(Optical Character Recognition) using computers has been developed for 20 years and applied to various fields such as parking management based on the recognition of license plates of cars. This technology was also used in the development of our smart OCR-based household ledger. In order to improve filling the purchase history into a smartphone based household account book, we can take pictures of receipts with the smarphone camera and automatically organize the purchase list. In this process, the recognition rate of the characters of the receipt image is not high enough with OCR technology. We could improve the rate by applying the image processing technology and adjusting the contrast of the receipt image. The rate improved from 89% to 92.5%.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2024.01a
• /
• pp.123-125
• /
• 2024

다크웹(Dark Web)은 마약, 불법 촬영물, 해킹, 무기 등 불법 콘텐츠의 공유 및 거래가 이루어지는 인터넷 영역으로서 최근에는 소셜미디어와 연계된 형태로 범죄 양상이 변화하고 있다. 본 논문에서는 최근 국내 외 사회 문제로 대두되고 있는 마약 범죄를 추적하기 위한 다크웹 수사 기술로서 다크웹 사용자를 소셜미디어로 유인하는 마약 정보 수집체계를 개발한다. 먼저 미국 마약단속국에서 공개한 대표적인 마약 용어 3개의 표준어 및 은어를 검색 키워드로 사용하여 마약 관련 다크웹을 수집하고, 수집된 다크웹을 크롤링하여 소셜미디어 계정 링크를 추출한다. 본 논문에서는 다양한 소셜미디어 중, 트위터 및 텔레그램 접속 링크를 수집하였으며 실험 결과, 접속 가능한 총 54개 다크웹 도메인의 9,046개 웹 페이지에서 트위터 유인 링크 567개, 텔레그램 유인 링크 118개를 추출하였다.

• Clinics in Shoulder and Elbow
• /
• v.10 no.2
• /
• pp.199-203
• /
• 2007

Purpose: It is very important to evaluate and fix coronoid process fractures because they are a critical element for a stable, effective elbow function. The lateral view of the elbow joint is used for a radiographic evaluation of the coronoid but an understanding of the fracture pattern is often difficult because of overlap of the radial head and obliquity of the fracture line. We developed the coronoid view, which is a new radiograph for an evaluation of the coronoid process fracture, and discuss its advantages for a postoperative follow-up. Materials and Methods: The coronoid view was designed for an evaluation of the anteromedial fragment of the coronoid process. After the patient sat on his side, the shoulder was abducted $45^{\circ}$ and the elbow was flexed $90^{\circ}$. The X-ray beam was shot perpendicular to the table. Since shoulder was abducted $45^{\circ}$ the fracture line of the coronoid process can be parallel to the X-ray beam, and the radial head can be cleared. Conclusion: The coronoid view can be a good alternative radiograph for an evaluation of a coronoid process fracture because the beam is parallel to the fracture line. The coronoid view can be particularly useful in postoperative patient follow-up where computed tomography is impractical due to metal implants and cost.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.18 no.6
• /
• pp.191-201
• /
• 2019

A fast and accurate integrated traffic object detection framework was proposed and developed, harnessing a computer-vision based deep-learning approach performing automatic object detections, a multi object tracking technology, and video pre-processing tools. The proposed method is capable of detecting traffic object such as autos, buses, trucks and vans from video recordings taken under a various kinds of external conditions such as stability of video, weather conditions, video angles, and counting the objects by tracking them on a real-time basis. By creating plausible experimental scenarios dealing with various conditions that likely affect video quality, it is discovered that the proposed method achieves outstanding performances except for the cases of rain and snow, thereby resulting in 98% ~ 100% of accuracy.

• Journal of the Korean Society of Radiology
• /
• v.6 no.2
• /
• pp.121-127
• /
• 2012

Of this study, it was found that there were 250 mammographic units in total installed and used in the areas for this study, and 36 units were used in general hospitals, 53 units in hospitals and 116 units in clinics. That is, the units in clinics accounted for 50% out of the whole units. As for the image acquisition method, it was found there were 131 units using F/S, 67 units using CR and 7 units using DR respectively. At present, F/S system was mainly used in the areas. As for the materials of target/filter, it was found that Mo/Mo was mostly used (66%), followed by Mo/Rh (25%). As for the size of focus, both 0.1 mm for small focus and 0.3 mm for large focus were mainly used for the units.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.10
• /
• pp.411-420
• /
• 2019

Camera model identification has been a subject of steady study in the field of digital forensics. Among the increasingly sophisticated crimes, crimes such as illegal filming are taking up a high number of crimes because they are hard to detect as cameras become smaller. Therefore, technology that can specify which camera a particular image was taken on could be used as evidence to prove a criminal's suspicion when a criminal denies his or her criminal behavior. This paper proposes a deep learning model to identify the camera model used to acquire the image. The proposed model consists of four convolution layers and two fully connection layers, and a high pass filter is used as a filter for data pre-processing. To verify the performance of the proposed model, Dresden Image Database was used and the dataset was generated by applying the sequential partition method. To show the performance of the proposed model, it is compared with existing studies using 3 layers model or model with GLCM. The proposed model achieves 98% accuracy which is similar to that of the latest technology.

• Journal of the Korean Society of Radiology
• /
• v.82 no.1
• /
• pp.244-249
• /
• 2021

A 53-year-old woman presented with dyspnea. She had undergone extended thymectomy for an invasive thymoma two months prior. CT revealed numerous small nodules in the lung. After that, she deteriorated owing to acute respiratory distress syndrome (ARDS), and the vascular surgeon planned veno-venous extracorporeal membrane oxygenation (ECMO). During percutaneous cannulation through the left femoral vein, a vascular injury was suspected, and the patient's vital signs became unstable. Diagnostic angiography showed a ruptured left common iliac vein, and the bleeding was stopped by placement of a stent-graft. May-Thurner syndrome was diagnosed on abdominal CT. Here, we report a rare case of ECMO-related vascular injury in a patient with an unrecognized anatomical variant, May-Thurner syndrome.

• Journal of Veterinary Clinics
• /
• v.22 no.2
• /
• pp.148-152
• /
• 2005

An intact female, 5-year-old, Pekingese, weighing 3.5kg with a history of a palpated abdominal mass was referred to Veterinary Medical Teaching Hospital, Seoul National University. In laboratory examination, there were no remarkable abnormalities. Radiographic findings included a left mid-abdominal mass with ill-defined margin, serosal detail loss of peritoneal space, non-uniform opacity of retroperitoneal space, and a radiopaque cystic calculus. On abdominal ultrasonography, a heterogeneous parenchymal mass with irregular contour in the left renal region was found. Computed tomographic findings showed a tumor embolus within the caudal vena cava and an invasion into mesentery, small bowel loops, spleen and pancreas around the large left renal mass. Unilateral nephroureterectomy was performed. Histopathologic examination of the resected mass confirmed the diagnosis of renal cell carcinoma. The dog died one day after surgery. Although ultrasonography could give diagnostic information about mass characteristics, computed tomography (CT) can provide key imaging features of mass characteristics.

• Journal of the Korea Computer Graphics Society
• /
• v.30 no.3
• /
• pp.149-158
• /
• 2024

Photo booths have traditionally provided a fun and easy way to capture and print photos to cherish memories. These booths allow individuals to capture their desired poses and props, sharing memories with friends and family. To enable diverse expressions, generative AI-powered photo booths have emerged. However, existing AI photo booths face challenges such as difficulty in taking group photos, inability to accurately reflect user's poses, and the challenge of applying different concepts to individual subjects. To tackle these issues, we present CINEMAPIC, a photo booth system that allows users to freely choose poses, positions, and concepts for their photos. The system workflow includes three main steps: pre-processing, generation, and post-processing to apply individualized concepts. To produce high-quality group photos, the system generates a transparent image for each character and enhances the backdrop-composited image through a small number of denoising steps. The workflow is accelerated by applying an optimized diffusion model and GPU parallelization. The system was implemented as a prototype, and its effectiveness was validated through a user study and a large-scale pilot operation involving approximately 400 users. The results showed a significant preference for the proposed system over existing methods, confirming its potential for real-world photo booth applications. The proposed CINEMAPIC photo booth is expected to lead the way in a more creative and differentiated market, with potential for widespread application in various fields.

• Journal of radiological science and technology
• /
• v.37 no.4
• /
• pp.323-330
• /
• 2014

Image reconstruction is one of the most important processes in CT (Computed tomography) technology. For fast scanning and low dose to the objects, iterative reconstruction is becoming more and more important. In the implementation of iterative reconstruction, projection and backprojection processes are considered to be indispensable parts. However, many approaches for projection and backprojection may result severe image artifacts due to their discrete characteristics and affects the reconstructed image quality. Thus, new approaches for projection and backprojection are highly demanded these days. In this paper, distance-driven approach was evaluated and compared with other conventional methods. The numerical simulator was developed to make the phantoms, and projection and backprojection images were compared using these approaches. As a result, it turned out that there are less artifacts during projection and backprojection in parallel and fan beam geometry.