• Smart Media Journal
• /
• v.12 no.3
• /
• pp.112-119
• /
• 2023

Shrimp farming has been becoming a new source of income for fishermen in South Korea. It is often necessary for fishers to measure the size of the shrimp for the purpose to understand the growth rate of the shrimp and to determine the amount of food put into the breeding pond. Traditional methods rely on humans, which has huge time and labor costs. This paper proposes a deep learning-based method for calculating the size of shrimps automatically. Firstly, we use fine-tuning techniques to update the Mask RCNN model with our farm data, enabling it to segment shrimps and generate shrimp masks. We then use skeletonizing method and maximum inscribed circle to calculate the length and width of shrimp, respectively. Our method is simple yet effective, and most importantly, it requires a small hardware resource and is easy to deploy to shrimp farms.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.05a
• /
• pp.417-419
• /
• 2022

농축업에 ICT 기술을 접목한 스마트 팜은 생육환경을 자동으로 조절하여 노동력 등을 줄이고도 생산성과 품질을 향상시키는 것이 큰 장점이다. 하지만, 수익으로 이어지는 출하량과 품질 유지를 위해서 병충해에 주의를 기울여야 함은 여전하다. 따라서 토마토 잎 병충해 발생 시, 적절한 대응을 통해 더 큰 피해를 막을 수 있으므로, 초기 증상을 포착하는 기법을 개발한다. 오픈 데이터 셋인 Ai hub 의 시설작물 질병 데이터셋과 추가로 확보한 샘플을 포함해 2 개의 충해, 4 개의 병해에 1,231 장으로 데이터셋을 직접 구성해서 학습했다. 객체 탐지와 세그먼테이션이 동시에 가능하며 작은 병변도 잘 탐지하는 모델을 사용해서 총 6 가지 병충해에 대한 뚜렷한 증상 탐지를 보여주었다.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2021.11a
• /
• pp.221-222
• /
• 2021

Conventionally, in material management at a construction site, the type, size, and quantity of materials are identified by the eyes of the worker. Labor-intensive material management by manpower is slow, requires a lot of manpower, is prone to errors, and has limitations in that computerization of information on the identified types and quantities is additionally required. Therefore, a method that can quickly and accurately determine the type, size, and quantity of materials with a minimum number of workers is required to reduce labor costs at the construction site and improve work efficiency. In this study, we developed an automated convolution neural network(CNN) and computer vision technology-based rebar size and quantity estimation system that can quickly and accurately determine the type, size, and quantity of materials through images.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.112-114
• /
• 2021

In order to detect urban conditions, the number of means of transportation and traffic flow are essential factors to be identified. This paper improved the detection system capabilities shown in previous studies using the SwinTransformer model, which showed higher performance than existing convolutional neural networks, by learning various vehicle types using existing Mask R-CNN and introducing today's widely used transformer model to detect certain types of vehicles in urban aerial images.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.532-534
• /
• 2022

본 논문에서는 미생 드라마 비디오들을 토대로 구축한 비디오 인물 개체 분할 데이터 집합인 MHIS를 소개하고, 등장인물 클래스 간의 심각한 데이터 불균형 문제를 효과적으로 해결하기 위한 새로운 비디오 데이터 보강 기법인 CDVA를 제안한다. 기존의 비디오 데이터 보강 기법들과는 달리, 새로운 CDVA 보강 기법은 비디오의 시공간적 맥락을 충분히 고려해서 부족한 인물 클래스의 훈련 비디오 데이터들을 추가 생성함으로써, 비디오 개체 분할 신경망 모델의 성능을 효과적으로 개선시킬 수 있다. 본 논문에서는 정량 및 정성 실험들을 통해, 제안 비디오 데이터 보강 기법의 우수성을 입증한다.

• ETRI Journal
• /
• v.46 no.2
• /
• pp.263-276
• /
• 2024

Different diseases occur in the brain. For instance, hereditary and progressive diseases affect and degenerate the white matter. Although addressing, diagnosing, and treating complex abnormalities in the brain is challenging, different strategies have been presented with significant advances in medical research. With state-of-art developments in artificial intelligence, new techniques are being applied to brain magnetic resonance images. Deep learning has been recently used for the segmentation and classification of brain images. In this study, we classified normal and pathological brain images using pretrained deep models through transfer learning. The EfficientNet-B5 model reached the highest accuracy of 98.39% on real data, 91.96% on augmented data, and 100% on pathological data. To verify the reliability of the model, fivefold cross-validation and a two-tier cross-test were applied. The results suggest that the proposed method performs reasonably on the classification of brain magnetic resonance images.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.25 no.4
• /
• pp.65-74
• /
• 2021

In order to efficiently inspect rapidly increasing old tunnels in many well-developed countries, many inspection methodologies have been proposed using imaging equipment and image processing. However, most of the existing methodologies evaluated their performance on a clean concrete surface with a limited area where other objects do not exist. Therefore, this paper proposes a 6-step framework for tunnel crack detection deep learning model development. The proposed method is mainly based on negative sample (non-crack object) training and Cascade Mask R-CNN. The proposed framework consists of six steps: searching for cracks in images captured from real tunnels, labeling cracks in pixel level, training a deep learning model, collecting non-crack objects, retraining the deep learning model with the collected non-crack objects, and constructing final training dataset. To implement the proposed framework, Cascade Mask R-CNN, an instance segmentation model, was trained with 1561 general crack images and 206 non-crack images. In order to examine the applicability of the trained model to the real-world tunnel crack detection, field testing is conducted on tunnel spans with a length of about 200m where electric wires and lights are prevalent. In the experimental result, the trained model showed 99% precision and 92% recall, which shows the excellent field applicability of the proposed framework.

• Journal of Biomedical Engineering Research
• /
• v.31 no.6
• /
• pp.427-433
• /
• 2010

Although tracking methods are efficient and popular for vessel segmentation, they require a seed to initiate an instance of tracking. In this paper, a new method to detect new seeds for tracking of arterial segments from CT angiography (CTA) and to construct a vascular structure is proposed. The proposed algorithm is based on shape analysis of connected components in a volume of interest around a vessel segment which was already extracted by tracking. The eigenvalues of the covariance matrix are used as the shape features for detection. The experimental results on actual clinical data showed that the results totally revealed the arterial tree not hindered by bone or veins. In visual comparison to a method which combines registration and subtraction of both pre-contrast and post-contrast CT volumes, the proposed method produced comparable results to the reference method and were confirmed of its feasibility for clinical use of reducing the cost and burden of patients.

• Korean Journal of Construction Engineering and Management
• /
• v.5 no.5 s.21
• /
• pp.151-162
• /
• 2004

Current methods for construction site modeling employ large, expensive laser range scanners that produce dense range point clouds of a scene from different perspectives. Days of skilled interpretation and of automatic segmentation may be required to convert the clouds to a finished CAD model. The dynamic nature of the construction environment requires that a real-time local area modeling system be capable of handling a rapidly changing and uncertain work environment. However, in practice, large, simple, and reasonably accurate embodying volumes are adequate feedback to an operator who, for instance, is attempting to place materials in the midst of obstacles with an occluded view. For real-time obstacle avoidance and automated equipment control functions, such volumes also facilitate computational tractability. In this research, a human operator's ability to quickly evaluate and associate objects in a scene is exploited. The operator directs a laser range finder mounted on a pan and tilt unit to collect range points on objects throughout the workspace. These groups of points form sparse range point clouds. These sparse clouds are then used to create geometric primitives for visualization and modeling purposes. Experimental results indicate that these models can be created rapidly and with sufficient accuracy for automated obstacle avoidance and equipment control functions.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2021.11a
• /
• pp.223-224
• /
• 2021

The process before the model learning stage in AI R&D can be subdivided into data collection/cleansing-data purification-data labeling. After that, according to the purpose of development, it goes through a stage of verifying the model by performing learning by using the algorithm of the artificial intelligence model. Several studies describe an important part of AI research as the learning stage, and try to increase the accuracy by changing the structure and layer of the AI model. However, if the refinement and labeling process of the learning data is tailored only to the model format and is not made for the purpose of development, the desired AI model cannot be obtained. The latest research reveals that most AI research failures are the failure of the learning data rather than the structure of the AI model. analyzed.