• 한국환경복원기술학회지
• /
• 제24권6호
• /
• pp.89-96
• /
• 2021

In this study, a deep-learning image analysis model was established and validated for AI-based monitoring of the tidal flat ecosystem for marine protected creatures Ocypode stimpsoni and their habitat. The data in the study was constructed using an unmanned aerial vehicle, and the U-net model was applied for the deep learning model. The accuracy of deep learning model learning results was about 0.76 and about 0.8 each for the Ocypode stimpsoni and their burrow whose accuracy was higher. Analyzing the distribution of crabs and burrows by putting orthomosaic images of the entire study area to the learned deep learning model, it was confirmed that 1,943 Ocypode stimpsoni and 2,807 burrow were distributed in the study area. Through this study, the possibility of using the deep learning image analysis technology for monitoring the tidal ecosystem was confirmed. And it is expected that it can be used in the tidal ecosystem monitoring field by expanding the monitoring sites and target species in the future.

• Journal of Information Processing Systems
• /
• 제19권3호
• /
• pp.289-301
• /
• 2023

Due to the development and dissemination of modern technology, anyone can easily communicate using services such as social network service (SNS) through a personal computer (PC) or smartphone. The development of these technologies has caused many beneficial effects. At the same time, bad effects also occurred, one of which was the spam problem. Spam refers to unwanted or rejected information received by unspecified users. The continuous exposure of such information to service users creates inconvenience in the user's use of the service, and if filtering is not performed correctly, the quality of service deteriorates. Recently, spammers are creating more malicious spam by distorting the image of spam text so that optical character recognition (OCR)-based spam filters cannot easily detect it. Fortunately, the level of transformation of image spam circulated on social media is not serious yet. However, in the mail system, spammers (the person who sends spam) showed various modifications to the spam image for neutralizing OCR, and therefore, the same situation can happen with spam images on social media. Spammers have been shown to interfere with OCR reading through geometric transformations such as image distortion, noise addition, and blurring. Various techniques have been studied to filter image spam, but at the same time, methods of interfering with image spam identification using obfuscated images are also continuously developing. In this paper, we propose a deep learning-based spam image detection model to improve the existing OCR-based spam image detection performance and compensate for vulnerabilities. The proposed model extracts text features and image features from the image using four sub-models. First, the OCR-based text model extracts the text-related features, whether the image contains spam words, and the word embedding vector from the input image. Then, the convolution neural network-based image model extracts image obfuscation and image feature vectors from the input image. The extracted feature is determined whether it is a spam image by the final spam image classifier. As a result of evaluating the F1-score of the proposed model, the performance was about 14 points higher than the OCR-based spam image detection performance.

• 한국지리정보학회지
• /
• 제22권3호
• /
• pp.82-98
• /
• 2019

정보화 사회로 진입하면서 공간정보의 중요성은 급격하게 부각되고 있다. 특히 스마트시티, 디지털트윈과 같은 Real World Object의 3차원 공간정보 구축 및 모델링은 중요한 핵심기술로 자리매김하고 있다. 구축된 3차원 공간정보는 국토관리, 경관분석, 환경 및 복지 서비스 등 다양한 분야에서 활용된다. 영상기반의 3차원 모델링은 객체 벽면에 대한 텍스처링을 생성하여 객체의 가시성과 현실성을 높이고 있다. 하지만 이러한 텍스처링은 영상 취득 당시의 가로수, 인접 객체, 차량, 현수막 등의 물리적 적치물에 의해 필연적으로 폐색영역이 발생한다. 이러한 폐색영역은 구축된 3차원 모델링의 현실성과 정확성 저하의 주요원인이다. 폐색영역 해결을 위한 다양한 연구가 수행되고 있으며, 딥러닝을 이용한 폐색영역 검출 및 해결방안에 대한 연구가 수행되고 있다. 딥러닝 알고리즘 적용한 폐색영역 검출 및 해결을 위해서는 충분한 학습 데이터가 필요하며, 수집된 학습 데이터 품질은 딥러닝의 성능 및 결과에 직접적인 영향을 미친다. 따라서 본 연구에서는 이러한 학습 데이터의 품질에 따라 딥러닝의 성능 및 결과를 확인하기 위하여 다양한 영상품질을 이용하여 영상의 폐색영역 검출 능력을 분석하였다. 폐색을 유발하는 객체가 포함된 영상을 인위적이고 정량화된 영상품질별로 생성하여 구현된 딥러닝 알고리즘에 적용하였다. 연구결과, 밝기값 조절 영상품질은 밝은 영상일수록 0.56 검출비율로 낮게 나타났고 픽셀크기와 인위적 노이즈 조절 영상품질은 원본영상에서 중간단계의 비율로 조절된 영상부터 결과 검출비율이 급격히 낮아지는 것을 확인할 수 있었다. F-measure 성능평가 방법에서 노이즈 조절한 영상품질 변화가 0.53으로 가장 높게 나타났다. 연구결과로 획득된 영상품질별에 따른 폐색영역 검출 능력은 향후 딥러닝을 실제 적용을 위한 귀중한 기준으로 활용될 것이다. 영상 취득 단계에서 일정 수준의 영상 취득과 노이즈, 밝기값, 픽셀크기 등에 대한 기준을 마련함으로써 딥러닝을 실질적인 적용에 많은 기여가 예상된다.

• 디지털콘텐츠학회 논문지
• /
• 제19권6호
• /
• pp.1177-1183
• /
• 2018

최근 기계학습의 발달로 인공지능을 구현하는 머신러닝과 딥러닝 같은 기술이 많은 관심을 받고 있다. 본 논문에서는 딥러닝 기반의 R-CNN을 이용한 바이너리 악성코드를 이미지화 하고 이미지에서 특징을 추출해 패밀리를 분류한다. 본 논문에서는 딥러닝에서 두 단계를 이용해 악성코드를 CNN을 이용해 이미지화하고, 악성코드의 패밀리가 갖는 특징을 R-CNN을 이용해 분류함으로 악성코드를 이미지화하여 특징을 분류하고 패밀리를 분류한 후 악성코드의 진화를 자동 분류한다. 제안 기법은 검출율이 93.4%로 우수한 탐지 성능을 보였고 정확도는 98.6%로 매우 높은 성능을 보였다. 또한 악성코드를 이미지화 하는 CNN 처리속도가 23.3ms, 하나의 샘플을 분류하기 위해서 R-CNN처리 속도는 4ms로 비교적 빠르게 악성코드를 판별하고 분류가 가능함을 실험을 통해 증명하였다.

• 한국컴퓨터그래픽스학회논문지
• /
• 제28권3호
• /
• pp.45-54
• /
• 2022

영상의 깊이 추정은 다양한 영상 분석의 기반이 되는 기술이다. 딥러닝 모델을 활용한 분석 방법이 대두되면서, 영상의 깊이 추정 분야 또한 딥러닝을 활용하는 연구가 활발하게 이루어지고 있다. 현재 대부분의 딥러닝 영상 깊이 추정 모델들은 깨끗하고 이상적인 환경에서 학습되고 있다. 하지만 연무, 안개가 낀 열악한 환경에서도 깊이 추정 기술이 잘 동작할 수 있으려면 이러한 환경의 데이터를 포함하여야 한다. 하지만 열악한 환경의 영상을 충분히 확보하는 것이 어려운 실정이며, 불균일한 안개 데이터를 얻는 것은 특히 어려운 문제이다. 이를 해결하기 위해, 본 연구에서는 불균일 안개 영상 합성 방법과 이를 활용한 단안 기반의 깊이 추정 딥러닝 모델의 학습을 제안한다. 안개가 주로 실외에서 발생하는 것을 고려하여, 실외 위주의 데이터 세트를 구축한다. 그리고 실험을 통해 제안된 방법으로 학습된 모델이 합성 데이터와 실제 데이터에서 깊이를 잘 추정하는 것을 보인다.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회 2005년도 추계학술대회 및 정기총회
• /
• pp.175-181
• /
• 2005

This paper describes the development of on-line computer vision laboratories to teach the detailed image processing and pattern recognition techniques. The computer vision laboratories include distant image acquisition method, basic image processing and pattern recognition methods, lens and light, and communication. This study introduces a case study that teaches computer vision in distance learning environment. It shows a schematic of a distant loaming workstation and contents of laboratories with image processing examples. The study focus more on the contents of the vision Labs rather than internet application method. The study proposes the ways to improve the on-line computer vision laboratories and includes the further research perspectives

• 한국멀티미디어학회논문지
• /
• 제24권10호
• /
• pp.1311-1318
• /
• 2021

Agriculture is being damaged by harmful animals such as wild boars and water deer. It need to get permission to catch a wild boar and farmers are using a lot of methods to chase harmful animals. The methods through deep learning and image processing capture harmful animals with cameras. It is difficult to analyze harmful animals that are active at night. In this case, In this case, using deep learning by image correction can achieve a higher recognition rate.

• 전기전자학회논문지
• /
• 제27권1호
• /
• pp.59-64
• /
• 2023

딥러닝은 다양한 컴퓨터 비전 문제를 해결할 수 있지만, 대량의 데이터셋이 필요하다. 본 논문에서는 대형 이미지 데이터셋을 구축하기 위해 이미지 이진화 기반 데이터 증강 기법을 제안한다. 이미지 이진화를 사용하여 특성을 추출하고 추출된 나머지 픽셀을 랜덤하게 배치하여 새로운 이미지를 생성한다. 생성된 이미지는 원본 이미지와 유사한 품질을 보여주며, 딥러닝 모델에서도 뛰어난 성능을 보였다.

• Journal of Radiation Protection and Research
• /
• 제44권4호
• /
• pp.149-155
• /
• 2019

Background: Magnetic resonance (MR) image guided radiation therapy system, enables real time MR guided radiotherapy (RT) without additional radiation exposure to patients during treatment. However, MR image lacks electron density information required for dose calculation. Image fusion algorithm with deformable registration between MR and computed tomography (CT) was developed to solve this issue. However, delivered dose may be different due to volumetric changes during image registration process. In this respect, synthetic CT generated from the MR image would provide more accurate information required for the real time RT. Materials and Methods: We analyzed 1,209 MR images from 16 patients who underwent MR guided RT. Structures were divided into five tissue types, air, lung, fat, soft tissue and bone, according to the Hounsfield unit of deformed CT. Using the deep learning model (U-NET model), synthetic CT images were generated from the MR images acquired during RT. This synthetic CT images were compared to deformed CT generated using the deformable registration. Pixel-to-pixel match was conducted to compare the synthetic and deformed CT images. Results and Discussion: In two test image sets, average pixel match rate per section was more than 70% (67.9 to 80.3% and 60.1 to 79%; synthetic CT pixel/deformed planning CT pixel) and the average pixel match rate in the entire patient image set was 69.8%. Conclusion: The synthetic CT generated from the MR images were comparable to deformed CT, suggesting possible use for real time RT. Deep learning model may further improve match rate of synthetic CT with larger MR imaging data.

• Journal of Korean Medical Science
• /
• 제33권43호
• /
• pp.239.1-239.12
• /
• 2018

Background: We described a novel multi-step retinal fundus image reading system for providing high-quality large data for machine learning algorithms, and assessed the grader variability in the large-scale dataset generated with this system. Methods: A 5-step retinal fundus image reading tool was developed that rates image quality, presence of abnormality, findings with location information, diagnoses, and clinical significance. Each image was evaluated by 3 different graders. Agreements among graders for each decision were evaluated. Results: The 234,242 readings of 79,458 images were collected from 55 licensed ophthalmologists during 6 months. The 34,364 images were graded as abnormal by at-least one rater. Of these, all three raters agreed in 46.6% in abnormality, while 69.9% of the images were rated as abnormal by two or more raters. Agreement rate of at-least two raters on a certain finding was 26.7%-65.2%, and complete agreement rate of all-three raters was 5.7%-43.3%. As for diagnoses, agreement of at-least two raters was 35.6%-65.6%, and complete agreement rate was 11.0%-40.0%. Agreement of findings and diagnoses were higher when restricted to images with prior complete agreement on abnormality. Retinal/glaucoma specialists showed higher agreements on findings and diagnoses of their corresponding subspecialties. Conclusion: This novel reading tool for retinal fundus images generated a large-scale dataset with high level of information, which can be utilized in future development of machine learning-based algorithms for automated identification of abnormal conditions and clinical decision supporting system. These results emphasize the importance of addressing grader variability in algorithm developments.