• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2018.11a
• /
• pp.117-118
• /
• 2018

본 논문에서는 작은 객체를 검출하기 위한 수정 된 YOLOv3-tiny 를 제안한다. 컴퓨터 비전에서 작은 객체 검출은 제한된 해상도와 정보로 검출하기 어렵다. 이 문제를 해결하기 위해 기존 방법의 대부분은 높은 정확도 향상을 위해 속도를 희생한다. 본 논문은 정확도와 속도가 균형적인 성능을 통해 빠른 속도로 작은 객체를 검출하는 것을 목표로 한다. 실험은 WIDER FACE 와 자체 수집한 데이터베이스에서 기존 YOLOv3-tiny 보다 높은 87.48% mAP 를 얻었으며, 속도는 각각 100.5FPS 로 YOLOv3-tiny 보다는 느리지만 높은 정확도와 YOLOv3 보다는 빠르지만 낮은 정확도를 통해 균형적인 성능을 얻을 수 있다.

• Journal of the Semiconductor & Display Technology
• /
• v.21 no.3
• /
• pp.7-11
• /
• 2022

Object detection technology is one of the main research topics in the field of computer vision and has established itself as an essential base technology for implementing various vision systems. Recent DNN (Deep Neural Networks)-based algorithms achieve much higher recognition accuracy than traditional algorithms. However, it is well-known that the DNN model inference operation requires a relatively high computational power. In this paper, we analyze the inference time complexity of the state-of-the-art object detection architecture Yolov7 in various environments. Specifically, we compare and analyze the time complexity of four types of the Yolov7 model, YOLOv7-tiny, YOLOv7, YOLOv7-X, and YOLOv7-E6 when performing inference operations using CPU and GPU. Furthermore, we analyze the time complexity variation when inferring the same models using the Pytorch framework and the Onnxruntime engine.

• Journal of the Korean Society of Radiology
• /
• v.16 no.1
• /
• pp.25-34
• /
• 2022

Brain computed tomography (CT) is useful for brain lesion diagnosis, such as brain hemorrhage, due to non-invasive methodology, 3-dimensional image provision, low radiation dose. However, there has been numerous misdiagnosis owing to a lack of radiologist and heavy workload. Recently, object detection technologies based on artificial intelligence have been developed in order to overcome the limitations of traditional diagnosis. In this study, the applicability of a deep learning-based YOLOv5s model was evaluated for brain hemorrhage detection using brain CT images. Also, the effect of hyperparameters in the trained YOLOv5s model was analyzed. The YOLOv5s model consisted of backbone, neck and output modules. The trained model was able to detect a region of brain hemorrhage and provide the information of the region. The YOLOv5s model was trained with various activation functions, optimizer functions, loss functions and epochs, and the performance of the trained model was evaluated in terms of brain hemorrhage detection accuracy and training time. The results showed that the trained YOLOv5s model is able to provide a bounding box for a region of brain hemorrhage and the accuracy of the corresponding box. The performance of the YOLOv5s model was improved by using the mish activation function, the stochastic gradient descent (SGD) optimizer function and the completed intersection over union (CIoU) loss function. Also, the accuracy and training time of the YOLOv5s model increased with the number of epochs. Therefore, the YOLOv5s model is suitable for brain hemorrhage detection using brain CT images, and the performance of the model can be maximized by using appropriate hyperparameters.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.1161-1162
• /
• 2023

모바일 로봇의 자율주행을 위하여 인터넷이 제약된 환경에서도 가능한 Edge computing 에서의 Object Detection 이 필수적이다. 본 논문에서는 이를 위해 Orin 보드에서 YOLOv7 과 Complex_YOLOv4 를 구현하였다. 직접 취득한 데이터를 통해 YOLOv7 을 구현한 결과 0.56 의 mAP 로 프레임당 133ms 가 소요되었다. Kitti Dataset 을 통해 Complex_YOLOv4 를 구현한 결과 0.88 의 mAP 로 프레임당 236ms 가 소요되었다. Comple_YOLOv4 가 YOLOv7 보다 더 많은 데이터를 예측하기에 시간은 더 소요되지만 높은 정확성을 가지는 것을 확인할 수 있었다.

• Korean Journal of Remote Sensing
• /
• v.39 no.3
• /
• pp.297-309
• /
• 2023

The degradation of coastal ecosystems and fishery environments is accelerating due to the recent phenomenon of invertebrate grazers. To effectively monitor and implement preventive measures for this phenomenon, the adoption of remote sensing-based monitoring technology for extensive maritime areas is imperative. In this study, we compared and analyzed the robustness of deep learning-based object detection modelsfor detecting and monitoring invertebrate grazersfrom underwater videos. We constructed an image dataset targeting seven representative species of invertebrate grazers in the coastal waters of South Korea and trained deep learning-based object detection models, You Only Look Once (YOLO)v7 and YOLOv8, using this dataset. We evaluated the detection performance and speed of a total of six YOLO models (YOLOv7, YOLOv7x, YOLOv8s, YOLOv8m, YOLOv8l, YOLOv8x) and conducted robustness evaluations considering various image distortions that may occur during underwater filming. The evaluation results showed that the YOLOv8 models demonstrated higher detection speed (approximately 71 to 141 FPS [frame per second]) compared to the number of parameters. In terms of detection performance, the YOLOv8 models (mean average precision [mAP] 0.848 to 0.882) exhibited better performance than the YOLOv7 models (mAP 0.847 to 0.850). Regarding model robustness, it was observed that the YOLOv7 models were more robust to shape distortions, while the YOLOv8 models were relatively more robust to color distortions. Therefore, considering that shape distortions occur less frequently in underwater video recordings while color distortions are more frequent in coastal areas, it can be concluded that utilizing YOLOv8 models is a valid choice for invertebrate grazer detection and monitoring in coastal waters.

• 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 the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.40 no.5
• /
• pp.381-391
• /
• 2022

This paper is intended to find one of the prevailing deep learning models that are a type of AI (Artificial Intelligence) that helps rapidly detect damaged buildings where disasters occur. The models selected are SSD-512, RetinaNet, and YOLOv3 which are widely used in object detection in recent years. These models are based on one-stage detector networks that are suitable for rapid object detection. These are often used for object detection due to their advantages in structure and high speed but not for damaged building detection in disaster management. In this study, we first trained each of the algorithms on xBD dataset that provides the post-disaster imagery with damage classification labels. Next, the three models are quantitatively evaluated with the mAP(mean Average Precision) and the FPS (Frames Per Second). The mAP of YOLOv3 is recorded at 34.39%, and the FPS reached 46. The mAP of RetinaNet recorded 36.06%, which is 1.67% higher than YOLOv3, but the FPS is one-third of YOLOv3. SSD-512 received significantly lower values than the results of YOLOv3 on two quantitative indicators. In a disaster situation, a rapid and precise investigation of damaged buildings is essential for effective disaster response. Accordingly, it is expected that the results obtained through this study can be effectively used for the rapid response in disaster management.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.3
• /
• pp.115-124
• /
• 2022

In this paper, YOLO and EAST models are tested to analyze their performance in text area detecting for real-world and normal text images. The earl ier YOLO models which include YOLOv3 have been known to underperform in detecting text areas for given images, but the recently released YOLOv4 and YOLOv5 achieved promising performances to detect text area included in various images. Experimental results show that both of YOLO v4 and v5 models are expected to be widely used for text detection in the filed of scene text recognition in the future.

• Journal of Korean Society on Water Environment
• /
• v.37 no.4
• /
• pp.275-285
• /
• 2021

Algal bloom is an important issue in maintaining the safety of the drinking water supply system. Fast detection and classification of algae images are essential for the management of algal blooms. Conventional visual identification using a microscope is a labor-intensive and time-consuming method that often requires several hours to several days in order to obtain analysis results from field water samples. In recent decades, various deep learning algorithms have been developed and widely used in object detection studies. YOLO is a state-of-the-art deep learning algorithm. In this study the third version of the YOLO algorithm, namely, YOLOv3, was used to develop an algae image detection model. YOLOv3 is one of the most representative one-stage object detection algorithms with faster inference time, which is an important benefit of YOLO. A total of 1,114 algae images for 30 genera collected by microscope were used to develop the YOLOv3 algae image detection model. The algae images were divided into four groups with five, 10, 20, and 30 genera for training and testing the model. The mean average precision (mAP) was 81, 70, 52, and 41 for data sets with five, 10, 20, and 30 genera, respectively. The precision was higher than 0.8 for all four image groups. These results show the practical applicability of the deep learning algorithm, YOLOv3, for algae image detection.

• International Journal of Advanced Culture Technology
• /
• v.11 no.1
• /
• pp.320-324
• /
• 2023

As urbanization accelerates and facilities that use energy increase, human life and property damage due to fire is increasing. Therefore, a fire monitoring system capable of quickly detecting a fire is required to reduce economic loss and human damage caused by a fire. In this study, we aim to develop an improved artificial intelligence model that can increase the accuracy of low fire alarms by mixing DeepSORT, which has strengths in object tracking, with the YOLOv5s model. In order to develop a fire detection model that is faster and more accurate than the existing artificial intelligence model, DeepSORT, a technology that complements and extends SORT as one of the most widely used frameworks for object tracking and YOLOv5s model, was selected and a mixed model was used and compared with the YOLOv5s model. As the final research result of this paper, the accuracy of YOLOv5s model was 96.3% and the number of frames per second was 30, and the YOLOv5s_DeepSORT mixed model was 0.9% higher in accuracy than YOLOv5s with an accuracy of 97.2% and number of frames per second: 30.