• Proceedings of the Korea Information Processing Society Conference
• /
• 2011.04a
• /
• pp.497-500
• /
• 2011

이전 필름 카메라 시대에는 한 필름에 촬영 가능한 사진의 수가 제한되고 인화와 현상에 대한 비용과 시간 소모로 인해, 꼭 필요하거나 중요한 순간에 사진을 촬영하였다. 그러나 최근에는 디지털 카메라의 보급과 대용량화된 메모리로 인해, 이전의 필름 카메라 시대와는 달리 일반 사람들도 한번에 많은 양의 사진을 촬영하는 일이 많아졌다. 이와 같이 관리해야 할 사진의 수가 많아질수록 사진을 분류하고 관리하는 작업에 많은 노력과 비용이 소모된다. 본 논문에서는 윈도우와 우선순위 큐를 이용하여, 촬영시각 문맥 (temporal context)의 흐름이 약한 순서대로 사진들을 클러스터링하는 방법을 제안한다. 제안한 방법의 평가를 위해서, Cooper 가 제안한 이벤트 클러스터링 방법과 정확도와 재현율을 비교하였으며, 사진 촬영 시각 차이의 분포의 편차가 작을수록, 제안한 클러스터링 방법이 높은 정확도를 보였다. 본 논문에서 제안한 촬영 시각 클러스터링은 많은 수의 사진들을 이벤트 기반으로 자동 분류하는데 활용될 수 있으며, 클러스터링된 정보들을 그룹별로 시각화하기 위한 인터페이스를 개발하는 것을 향후 연구과제로 제시한다.

• Journal of the Korean Society of Radiology
• /
• v.17 no.1
• /
• pp.123-129
• /
• 2023

The purpose of this study was to analyze differences in imaging quality and dose difference between intra-venous (IV) and intra-arterial (IA) liver dynamic computed tomography (CT). Herein, retrospective, blinded analysis was conducted to analyze signal-to-noise and contrast-to-noise ratios in cases of patients who underwent IV or IA liver dynamic CT for transarterial chemoembolization (TACE), an interventional procedure for hepatocellular carcinoma. The dose length product (DLP) value stored in Picture Archive and Communication System (PACS) was used to calculate the effective dose and thereby compare differences in the dose between the two methods. The mean liver and spleen signal to noise ratio (SNR) was greater in IV-liver dynamic CT than in IA-liver dynamic CT; however, contrast to noise ratio (CNR) was higher in IA-liver dynamic CT than in IV-liver dynamic CT. However, there were no differences in DLP and effective dose between the two methods. In conclusion, our findings showed that IA-liver dynamic CT showed a similar effective dose and superior CNR compared with IV-liver dynamic CT. Further studies must analyze 3D angiography CT of the hepatic artery to clearly distinguish the feeding artery, which is the essential step in interventional procedures for hepatocellular carcinoma.

• Journal of Veterinary Clinics
• /
• v.27 no.3
• /
• pp.284-288
• /
• 2010

The aim of this report is to compare quantitatively computed radiography (CR) and screen-film radiography (SFR) in the detection of peritoneal effusion in dogs. Normal four beagle dogs and one Maltese dog were used. Each five CR and SFR abdominal images of right lateral and ventro-dorsal position were obtained after lodge of 6 ml, 8 ml, 12 ml, 15 ml, and 18 ml of normal saline by intraperitoneal injection within the abdomen. The reviewers were asked to evaluate each SFR and CR images for the presence of peritoneal effusion using the score by the presence of a peritoneal effusion on a five-point ordinal scale. A receiver operating curve (ROC) analysis compared the two imaging modalities. The present study showed that there was no statistical difference between SFR and CR in the detecting peritoneal effusion, but CR was relatively more sensitive based on the increased area under its ROC analysis. Moreover, Readers were more likely to detect peritoneal effusion on CR images than SFR.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.11a
• /
• pp.1056-1059
• /
• 2020

최근 드론과 1인 미디어 시장의 성장으로, 영상 촬영 분야에서의 드론 산업이 활발하게 발전되고 있다. 본 논문에서는 딥러닝 기반 다중 객체 인식 기술인 Openpose를 활용하여 인물촬영을 위한 음성인식 드론 제어 시스템을 제안한다. 해당 시스템은 자연어 처리된 음성명령어를 통해 드론이 각 촬영 객체에 대한 회전, 초점변화 등 실제 영상촬영기법에 사용되는 다수의 동작을 수행할 수 있도록 한다. 최종적으로 96.2%의 정확도로 음성명령에 따라 동작을 수행하는 것을 확인할 수 있다. 이는 누구나 전문적 지식이나 경험 없이 음성만으로 쉽게 드론을 제어할 수 있을 것으로 기대된다.

• Journal of Korea Society of Industrial Information Systems
• /
• v.15 no.2
• /
• pp.69-81
• /
• 2010

A Cold Cathode Fluorescent Lamp(CCFL) is used as a LCD Monitor's backlight widely. The most common way to check CCFL's defects is an examination with the naked eye. This naked eye examination can cause an examination inconsistency and an industrial disaster. To examine CCFL defects, a shooting equipment and a defect detection algorithm are necessary. This paper shows the shooting environments for checking CCFL and presents some CCFL defect detection algorithms. As a result of experiments, our implementations showed 98.32% of successful defect detection of CCFL.

• Proceedings of the Korean Information Science Society Conference
• /
• 2011.06c
• /
• pp.443-446
• /
• 2011

본 연구는 컴퓨터 단층촬영 영상 (CT, Computed Tomography Image)에서 치열궁 (Dental Arch)을 자동으로 검출하는 기법을 제안한다. 제안된 기법에서는 3차원 컴퓨터 단층촬영 영상을 입력 받고 영역 확장법을 이용하여 하악을 분할 한 후 하악의 단면에서 전체적인 치아의 영역을 분할을 한다. 치아의 영역에서 세선화 작업을 거친 후 곡선 정합법을 이용하여 최종 치열궁을 검출한다. 실험 데이터로 두개골 컴퓨터 단층 촬영 데이터를 사용하였다. 본 연구는 치과 영상 데이터로부터 파노라마 영상을 얻는데 이용 될 수 있고 치과 분야의 질병 진단 및 진찰에 이용될 것으로 기대된다.

• Journal of Veterinary Clinics
• /
• v.31 no.3
• /
• pp.241-245
• /
• 2014

A miniature pinscher and two Maltese dogs were present with dyspnea, and radiography and computed tomography (CT) were performed. All dogs were diagnosed with giant bullous emphysema of the massive, hyperexpanded, and hypoattenuating bulla causing contralateral mediastinal shift. Giant bullous emphysema, the rarest form of bullous lung disease, is defined as a bulla that fills more than 30% of the hemithorax, and multi-detector CT scanning could provide the useful information to confirm a diagnosis of giant bullous emphysema in older dogs. The radiographic and computed tomographic features for giant bullous emphysema were described.

• Journal of Veterinary Clinics
• /
• v.31 no.5
• /
• pp.376-381
• /
• 2014

In this study, we analyzed the computed tomography (CT) measurements of lung volume and density in dogs with relation to body weight, age, sex, and breed. The multi-detector CT examination of the thorax was performed on dogs without respiratory or cardiovascular diseases. Three-dimensional reconstruction of CT images facilitated measurement of lung volume and density. There was a statistical significant correlation between body weight and lung volume (p < 0.0001). Lung density significantly decreased with an increase in body weight (p = 0.0078). However, no correlation was seen between these lung parameters and either sex or age of the dogs. In conclusion, this study shows that body weight is an important factor to consider when interpreting total lung volume and density values measured by quantitative CT. We highlight the need for further study using quantitative CT in identifying the potential effects of sex, age, and disease status on these parameters.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2010.04a
• /
• pp.365-368
• /
• 2010

LCD 모니터의 백라이트로 CCFL 형광체를 많이 사용하고 있으나 그 불량여부는 육안에 의존하고 있다. 그러나 육안 검사는 CCFL 현광체의 불량에 대한 일관성 있는 기준이 결여되고, 노동집약적인 검사로 인해 산업적 재해가 발생할 수 있다. CCFL 불량유무를 자동 판별하기 위해서는 물리적 촬영환경과 자동검출 알고리즘이 필수적이다. 본 논문에서는 CCFL 형광체의 불량을 자동으로 검사하기 위한 촬영환경을 설명하고, 그 촬영환경에 알맞은 대표적인 불량검출 알고리즘을 제안한다. 실험결과에 따르면, 알고리즘은 불량 판별율 98.86%와 과검율 3.34%의 성능을 보였다.

• Journal of the Korean Society of Radiology
• /
• v.11 no.6
• /
• pp.453-458
• /
• 2017

Computed Tomography (CT) provides information on the Diagnostic Reference Level Computed Tomography Dose Index (CTDI) and Dose Length Product (DLP) for accurate diagnosis of patients. However, it does not provide a dose change according to the table height for the diagnostic reference level provided by the CT equipment. The purpose of this study was to evaluate the image and dose according to the table height change using phantom (PMMA: Polymethyl Methacrylate) in order to find the optimal image and the minimum dose during computed tomography examination. When examining using a 32 cm PMMA phantom with the same thickness as the abdomen of an adult, there was little change in dose with table height. However, the noise evaluation of the image caused a high fluctuation of noise depending on the table height. and in the case of the 16 cm PMMA phantom, the change of the noise was small, but the dose change was about 30%. In conclusion, the location of the patient and the center of the detector are important during computed tomography (CT) examinations. In addition, table height setting is considered to be important for examinations with optimized image and minimum dose.