• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.6
• /
• pp.1209-1216
• /
• 2020

In this study, we propose a method for detecting coastal surface wastes using an UAV(Unmanned Aerial Vehicle) remote sensing method and an object detection algorithm based on deep learning. An object detection algorithm based on deep neural networks was proposed to detect coastal debris in aerial images. A deep neural network model was trained with image datasets of three classes: PET, Styrofoam, and plastics. And the detection accuracy of each class was compared with Darknet-53. Through this, it was possible to monitor the wastes landing on the shore by type through unmanned aerial vehicles. In the future, if the method proposed in this study is applied, a complete enumeration of the whole beach will be possible. It is believed that it can contribute to increase the efficiency of the marine environment monitoring field.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.120-122
• /
• 2021

Currently, as demand for cars at home and abroad increases, the number of vehicles is decreasing and the number of vehicles is increasing. This is the main cause of the traffic jam. To solve this problem, it operates a high-ocompancy vehicle (HOV) lane, a multi-passenger vehicle, but many people ignore the conditions of use and use it illegally. Since the police visually judge and crack down on such illegal activities, the accuracy of the crackdown is low and inefficient. In this paper, we propose a system design that enables more efficient detection using imaging techniques using computer vision to solve such problems. By improving the existing vehicle detection method that was studied, the trigger was set in the image so that the detection object can be selected and the image analysis can be conducted intensively on the target. Using the YOLO model, a deep learning object recognition model, we propose a method to utilize the shift amount of the center point rather than judging by the bounding box in the image to obtain real-time object detection and accurate signals.

• Journal of the Semiconductor & Display Technology
• /
• v.20 no.3
• /
• pp.140-145
• /
• 2021

This paper deals with the development of a deep-learning-based runway cleaning robot using an optical camera. A suitable model to realize real-time object detection was investigated, and the differences between the selected YOLOv3 and other deep learning models were analyzed. In order to check whether the proposed system is applicable to the actual runway, an experiment was conducted by making a prototype of the robot and a runway model. As a result, it was confirmed that the robot was well developed because the detection rate of FOD (Foreign Object Debris) and cracks was high, and the collection of foreign substances was carried out smoothly.

• Journal of the Semiconductor & Display Technology
• /
• v.20 no.1
• /
• pp.88-93
• /
• 2021

The automotive industry is developing day by day. Among them, it is very important to prevent accidents while driving. However, despite the importance of developing automobile industry technology, accidents due to road defects increase every year, especially in the rainy season. To this end, we proposed a road defect detection system for road management by converging deep learning and raspberry pi, which show various possibilities. In this paper, we developed a system that visually displays through a map after analyzing the images captured by the Raspberry Pi and the route GPS. The deep learning model trained for this system achieved 96% accuracy. Through this system, it is expected to manage road defects efficiently at a low cost.

• Journal of Multimedia Information System
• /
• v.7 no.1
• /
• pp.41-44
• /
• 2020

In autonomous driving, object recognition based on machine learning is one of the core software technologies. In particular, the object recognition using deep learning becomes an essential element for autonomous driving software to operate. In this paper, we introduce a drivable area detection method based on Region-based CNN model to support autonomous driving. To effectively detect the drivable area, we used the BDD dataset for model training and demonstrated its effectiveness. As a result, our R-CNN model using BDD datasets showed interesting results in training and testing for detection of drivable areas.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2023.05a
• /
• pp.173-174
• /
• 2023

In this paper, we introduce a model for detecting Personal Protective Equipment (PPE) using YOLO (You Only Look Once), an object detection neural network. PPE is used to maintain a safe working environment, and proper use of PPE protects workers' safety and health. However, failure to wear PPE or wearing it improperly can cause serious safety issues. Therefore, a PPE detection system is crucial in industrial settings.

• Journal of the Korea Convergence Society
• /
• v.11 no.12
• /
• pp.39-47
• /
• 2020

In this paper, we proposed a deep learning-based detection method using Saliency Attention to detect salient objects in images. The salient object detection separates the object where the human eye is focused from the background, and determines the highly relevant part of the image. It is usefully used in various fields such as object tracking, detection, and recognition. Existing deep learning-based methods are mostly Autoencoder structures, and many feature losses occur in encoders that compress and extract features and decoders that decompress and extend the extracted features. These losses cause the salient object area to be lost or detect the background as an object. In the proposed method, Saliency Attention is proposed to reduce the feature loss and suppress the background region in the Autoencoder structure. The influence of the feature values was determined using the ELU activation function, and Attention was performed on the feature values in the normalized negative and positive regions, respectively. Through this Attention method, the background area was suppressed and the projected object area was emphasized. Experimental results showed improved detection results compared to existing deep learning methods.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.9 no.5
• /
• pp.472-477
• /
• 2016

The most important and critical to the performance of video surveillance systems is to be detected exactly how much. In order to accurately track the object must be able to accurately separate the background and object. However, the system itself rather than the human vision exactly distinguish the object and the background, to assess the situation, it is not easy. If we can accurately detect the background and the object, to be able to accurately track an object, it is possible to increase the reliability of the system, have a significant impact on the success of the entire production system. In this paper, we propose a way to distinguish more precisely the background and the object being to determine the background environment changes more accurately.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.5
• /
• pp.605-611
• /
• 2023

Accidents at construction sites have a very high rate of fatalities due to the nature of being prone to severe injury patients. In order to reduce the mortality rate of severely injury patients, quick response is required, and some systems that detect accidents using AI technology and cameras have been devised to respond quickly to accidents. However, since existing accident detection systems use only a single camera, there are blind spots, Thus, they cannot detect all accidents at a construction site. Therefore, in this paper, we present the system that minimizes the detection blind spot by using multiple cameras. Our implemented system extracts feature points from the images of multiple cameras with the YOLO-pose library, and inputs the extracted feature points to a Long Short Term Memory-based recurrent neural network in order to detect accidents. In our experimental result, we confirme that the proposed system shows high accuracy while minimizing detection blind spots by using multiple cameras.

• Journal of Korea Multimedia Society
• /
• v.15 no.3
• /
• pp.323-335
• /
• 2012

In this paper, we propose an AR(augmented reality) system using user's bare hand based on computer vision. It is important for registering a virtual object on the real input image to detect and track correct feature points. The AR systems with markers are stable but they can not register the virtual object on an acquired image when the marker goes out of a range of the camera. There is a tendency to give users inconvenient environment which is limited to control a virtual object. On the other hand, our system detects fingertips as fiducial features using adaptive ellipse fitting method considering the geometric characteristics of hand. It registers the virtual object stably by getting movement of fingertips with determining the shortest distance from a palm center. We verified that the accuracy of fingertip detection over 82.0% and fingertip ordering and tracking have just 1.8% and 2.0% errors for each step. We proved that this system can replace the marker system by tacking a camera projection matrix effectively in the view of stable augmentation of virtual object.