• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.579-580
• /
• 2023

본 논문에서는 동물의 십자인대 질환의 진단지표인 경골고원각도(TPA)를 자동으로 측정하는 딥러닝 소프트웨어 기법을 제안한다. 동물 X-ray 영상에서 나타나는 피사체의 위치와 형태에 대한 다양한 변이는 TPA(Tibial Plateau Angle) 지표 산출에 필요한 특징점 검출과정에서 학습 효율을 현저하게 저하시킨다. 이에 본 연구에서는 YOLO(You Only Look Once) 기반 모델을 사용하여 일차적으로 경골영역의 분할 단계를 수행하고, 이어서 경골 상단부의 과간융기와 복사뼈의 중심점을 찾는 과정을 Resnet 기반의 특징점 추출 모듈로서 구현함으로써 학습의 효율과 지표 검출의 정확도를 향상시켰다. 총 201 개의 실제 X-ray 영상을 사용하여 학습 속도와 영역 분할 및 특징점 추출의 정확도 측면을 고려함으로 제안된 이론의 타당성을 실험적으로 평가하였다.

• Journal of Information Processing Systems
• /
• v.19 no.6
• /
• pp.791-802
• /
• 2023

The manual inspection of photovoltaic (PV) panels to meet the requirements of inspection work for large-scale PV power plants is challenging. We present a hot spot detection and positioning method to detect hot spots in batches and locate their latitudes and longitudes. First, a network based on the YOLOv3 architecture was utilized to identify hot spots. The innovation is to modify the RU_1 unit in the YOLOv3 model for hot spot detection in the far field of view and add a neural network residual unit for fusion. In addition, because of the misidentification problem in the infrared images of the solar PV panels, the DeepLab v3+ model was adopted to segment the PV panels to filter out the misidentification caused by bright spots on the ground. Finally, the latitude and longitude of the hot spot are calculated according to the geometric positioning method utilizing known information such as the drone's yaw angle, shooting height, and lens field-of-view. The experimental results indicate that the hot spot recognition rate accuracy is above 98%. When keeping the drone 25 m off the ground, the hot spot positioning error is at the decimeter level.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.19 no.4
• /
• pp.117-126
• /
• 2023

Traditional markets, formed naturally, often feature aged buildings and facilities that are susceptible to fire. However, the lack of adequate fire detection systems in these markets can easily lead to large-scale fires upon ignition. Therefore, this study was conducted with the aim of detecting fires in traditional markets, utilizing thermal imaging cameras for data collection and the YOLOv8 model for object detection experiments. Data were collected in the night markets within traditional markets of xx city and by simulating fire scenarios. A comparative analysis of the Nano and XL models of YOLOv8 revealed that the XL model is more effective in detecting fires. The XL model not only demonstrated higher accuracy in correctly identifying flames but also tended to miss fewer fires compared to the Nano model. In the case of objects other than flames, the XL model showed superior performance over the Nano model. Taking all these factors into account, it is anticipated that with further data collection and improvement in model performance, a suitable fire detection system for traditional markets can be developed.

• Journal of Software Engineering Society
• /
• v.28 no.2
• /
• pp.7-13
• /
• 2019

Consumers see many external factors when they buy a product, but the most important thing is the function and composition of the product. In particular, it is important to accurately identify the components of chemicals in order to prevent misuse, and a systematic management method is required. However, there is a lack of a systematic way for consumers to manage chemical components, and much effort is required to manage them manually. Therefore, in this paper, we propose a method of extracting product's information of real-time recognition based products that can be used to systematically manage chemical components, and design and develop a mobile application prototype of the method. The prototype can be developed to systematically manage multiple chemical products in the future.

• Journal of the Korea Convergence Society
• /
• v.12 no.12
• /
• pp.31-37
• /
• 2021

In this paper proposed a feature fusion and spatial attention-based modified YOLOv4S for small and occluded detection. Conventional YOLOv4S is a lightweight network and lacks feature extraction capability compared to the method of the deep network. The proposed method first combines feature maps of different scales with feature fusion to enhance semantic and low-level information. In addition expanding the receptive field with dilated convolution, the detection accuracy for small and occluded objects was improved. Second by improving the conventional spatial information with spatial attention, the detection accuracy of objects classified and occluded between objects was improved. PASCAL VOC and COCO datasets were used for quantitative evaluation of the proposed method. The proposed method improved mAP by 2.7% in the PASCAL VOC dataset and 1.8% in the COCO dataset compared to the Conventional YOLOv4S.

• Journal of Broadcast Engineering
• /
• v.26 no.2
• /
• pp.184-196
• /
• 2021

This paper proposes statistics adaptive linear regression-based object size prediction method for object detection. YOLOv2 and YOLOv3, which are typical deep learning-based object detection algorithms, designed the last layer of a network using statistics adaptive exponential regression model to predict the size of objects. However, an exponential regression model can propagate a high derivative of a loss function into all parameters in a network because of the property of an exponential function. We propose statistics adaptive linear regression layer to ease the gradient exploding problem of the exponential regression model. The proposed statistics adaptive linear regression model is used in the last layer of the network to predict the size of objects with statistics estimated from training dataset. We newly designed the network based on the YOLOv3tiny and it shows the higher performance compared to YOLOv3 tiny on the UFPR-ALPR dataset.

• Journal of Auto-vehicle Safety Association
• /
• v.11 no.2
• /
• pp.35-43
• /
• 2019

This paper describes the improved environment recognition algorithms using some type of sensors like LiDAR and cameras. Additionally, integrated control algorithm for an autonomous vehicle is included. The integrated algorithm was based on C++ environment and supported the stability of the whole driving control algorithms. As to the improved vision algorithms, lane tracing and traffic sign recognition were mainly operated with three cameras. There are two algorithms developed for lane tracing, Improved Lane Tracing (ILT) and Histogram Extension (HIX). Two independent algorithms were combined into one algorithm - Enhanced Lane Tracing with Histogram Extension (ELIX). As for the enhanced traffic sign recognition algorithm, integrated Mutual Validation Procedure (MVP) by using three algorithms - Cascade, Reinforced DSIFT SVM and YOLO was developed. Comparing to the results for those, it is convincing that the precision of traffic sign recognition is substantially increased. With the LiDAR sensor, static and dynamic obstacle detection and obstacle avoidance algorithms were focused. Therefore, improved environment recognition algorithms, which are higher accuracy and faster processing speed than ones of the previous algorithms, were proposed. Moreover, by optimizing with integrated control algorithm, the memory issue of irregular system shutdown was prevented. Therefore, the maneuvering stability of the autonomous vehicle in severe environment were enhanced.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.6
• /
• pp.2086-2097
• /
• 2021

The current cultural assets are being restored depending on the opinions of experts (craftsmen). We intend to introduce digitalized artificial intelligence techniques, excluding the personal opinions of experts on reconstruction of such cultural properties. The first step toward restoring digitized cultural properties is separation. The restoration of cultural properties should be reorganized based on recorded documents, period historical backgrounds and regional characteristics. The cultural properties in the form of photographs or images should be collected by separating the background. In addition, when restoring cultural properties most of them depend a lot on the tendency of the restoring person workers. As a result, it often occurs when there is a problem in the accuracy and reliability of restoration of cultural properties. In this study, we propose a search method for learning stored digital cultural assets using AI technology. Pagoda was selected for restoration of Cultural Properties. Pagoda data collection was collected through the Internet and various historical records. The pagoda data was classified by period and region, and grouped into similar buildings. The collected data was learned by applying the well-known CNN algorithm for artificial intelligence learning. The pagoda search used Yolo Marker to mark the tower shape. The tower was used a total of about 100-10,000 pagoda data. In conclusion, it was confirmed that the probability of searching for a tower differs according to the number of pagoda pictures and the number of learning iterations. Finally, it was confirmed that the number of 500 towers and the epochs in training of 8000 times were good. If the test result exceeds 8,000 times, it becomes overfitting. All so, I found a phenomenon that the recognition rate drops when the enemy repeatedly learns more than 8,000 times. As a result of this study, it is believed that it will be helpful in data gathering to increase the accuracy of tower restoration.

• Smart Media Journal
• /
• v.11 no.4
• /
• pp.9-18
• /
• 2022

Recently, research and system using AI is rapidly increasing in various fields. Smart farm using artificial intelligence and information communication technology is also being studied in agriculture. In addition, data-based precision agriculture is being commercialized by convergence various advanced technology such as autonomous driving, satellites, and big data. In Korea, the number of commercialization cases of facility agriculture among smart agriculture is increasing. However, research and investment are being biased in the field of facility agriculture. The gap between research and investment in facility agriculture and open-air agriculture continues to increase. The fields of fruit trees and plant factories have low research and investment. There is a problem that the big data collection and utilization system is insufficient. In this paper, we are proposed the system for determining the fruit tree growth stage using a deep learning-based object detection model. The system was proposed as a hybrid app for use in agricultural sites. In addition, we are implemented an object detection function for the fruit tree growth stage determine.

• Imaging Science in Dentistry
• /
• v.52 no.2
• /
• pp.219-224
• /
• 2022

Purpose: This study aimed to evaluate the performance of transfer learning in a deep convolutional neural network for classifying implant fixtures. Materials and Methods: Periapical radiographs of implant fixtures obtained using the Superline (Dentium Co. Ltd., Seoul, Korea), TS III(Osstem Implant Co. Ltd., Seoul, Korea), and Bone Level Implant(Institut Straumann AG, Basel, Switzerland) systems were selected from patients who underwent dental implant treatment. All 355 implant fixtures comprised the total dataset and were annotated with the name of the system. The total dataset was split into a training dataset and a test dataset at a ratio of 8 to 2, respectively. YOLOv3 (You Only Look Once version 3, available at https://pjreddie.com/darknet/yolo/), a deep convolutional neural network that has been pretrained with a large image dataset of objects, was used to train the model to classify fixtures in periapical images, in a process called transfer learning. This network was trained with the training dataset for 100, 200, and 300 epochs. Using the test dataset, the performance of the network was evaluated in terms of sensitivity, specificity, and accuracy. Results: When YOLOv3 was trained for 200 epochs, the sensitivity, specificity, accuracy, and confidence score were the highest for all systems, with overall results of 94.4%, 97.9%, 96.7%, and 0.75, respectively. The network showed the best performance in classifying Bone Level Implant fixtures, with 100.0% sensitivity, specificity, and accuracy. Conclusion: Through transfer learning, high performance could be achieved with YOLOv3, even using a small amount of data.