• Annual Conference on Human and Language Technology
• /
• 1993.10a
• /
• pp.493-501
• /
• 1993

본 논문에서는 한글의 자소간 흘림의 연속 필기를 허용하는 원고 작성기의 구현을 연구하였다. 이러한 온라인 한글 필기의 응용에서는 신속한 인식속도를 갖는 인식방법이 요구되며, 인식중에도 계속적인 필기가 가능하도록 하여 사용자에게 편의를 제공할 수 있어야 한다. 본 논문에서는 이와같은 요구사항을 만족시키기 위하여 스트링 정합방법에 기반한 신속한 인식 방법을 사용한다. 또한, 글자인식과 필기데이타 수집이 병행적으로 처리되도록 구성됨으로써 원고작성시에 자유로운 필기동작이 가능하도록 하였다. 실험결과 50명이 쓴 21,076자에 대하여 88.96%의 인식률을 제공하였으며, 제안하는 구현 방법이 원고작성 응용에 적합하게 동작함을 알 수 있었다.

• Proceedings of the KIEE Conference
• /
• 2003.07d
• /
• pp.2381-2383
• /
• 2003

This paper presents an effective approach to dynamic obstacle avoidance for mobile robot. The main concept of this approach includes modified polar mapping for recognition of the moving obstacle in vision-based robot systems. To simplify the segmentation of the moving obstacle from the background and to obtain its relative position data the modified polar mapping is proposed. Dynamic moving obstacles are avoided with a vision sensor and stationary obstacles are avoided with a sonar sensor.

• International Journal of Computer Science & Network Security
• /
• v.21 no.9
• /
• pp.63-66
• /
• 2021

The technological advancement in computer vision has made system like grab-and-go grocery a reality. Now all the shoppers have to do now is to walk in grab the items and go out without having to wait in the long queues. This paper presents an intelligent retail environment system that is capable of monitoring and tracking customer's activity during shopping based on their interaction with the shelf. It aims to develop a system that is low cost, easy to mount and exhibit adequate performance in real environment.

• International Journal of Advanced Culture Technology
• /
• v.8 no.4
• /
• pp.119-128
• /
• 2020

Object detection is an important field of computer vision and is applied to applications such as security, autonomous driving, and face recognition. Recently, as the application of artificial intelligence technology including deep learning has been applied in various fields, it has become a more powerful tool that can learn meaningful high-level, deeper features, solving difficult problems that have not been solved. Therefore, deep learning techniques are also being studied in the field of object detection, and algorithms with excellent performance are being introduced. In this paper, a deep learning-based object detection algorithm used to detect multiple objects in an image is investigated, and future development directions are presented.

• 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 Auto-vehicle Safety Association
• /
• v.14 no.2
• /
• pp.45-50
• /
• 2022

The vehicle motion in urban environment is determined by surrounding traffic flow, which cause understanding the flow to be a factor that dominantly affects the motion planning of the vehicle. The traffic flow in this urban environment is accessed using various urban infrastructure information. This paper represents a color recognition algorithm for traffic lights to perceive traffic condition which is a main information among various urban infrastructure information. Deep learning based vision open source realizes positions of traffic lights around the host vehicle. The data are processed to input data based on whether it exists on the route of ego vehicle. The colors of traffic lights are estimated through pixel values from the camera image. The proposed algorithm is validated in intersection situations with traffic lights on the test track. The results show that the proposed algorithm guarantees precise recognition on traffic lights associated with the ego vehicle path in urban intersection scenarios.

• International conference on construction engineering and project management
• /
• 2022.06a
• /
• pp.426-432
• /
• 2022

HVAC systems play a critical role in reducing energy consumption in buildings. Integrating occupants' thermal comfort evaluation into HVAC control strategies is believed to reduce building energy consumption while minimizing their thermal discomfort. Advanced technologies, such as visual sensors and deep learning, enable the recognition of occupants' discomfort-related actions, thus making it possible to estimate their thermal discomfort. Unfortunately, it remains unclear how accurate a deep learning-based classifier is to recognize occupants' discomfort-related actions in a working environment. Therefore, this research evaluates the classification performance of occupants' discomfort-related actions while sitting at a computer desk. To achieve this objective, this study collected RGB video data on nine college students' cold discomfort-related actions and then trained a deep learning-based classifier using the collected data. The classification results are threefold. First, the trained classifier has an average accuracy of 93.9% for classifying six cold discomfort-related actions. Second, each discomfort-related action is recognized with more than 85% accuracy. Third, classification errors are mostly observed among similar discomfort-related actions. These results indicate that using human action data will enable facility managers to estimate occupants' thermal discomfort and, in turn, adjust the operational settings of HVAC systems to improve the energy efficiency of buildings in conjunction with their thermal comfort levels.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.1
• /
• pp.111-118
• /
• 2022

In computer vision research, the two-dimensional human pose is a very extensive research direction, especially in pose tracking and behavior recognition, which has very important research significance. The acquisition of human pose targets, which is essentially the study of how to accurately identify human targets from pictures, is of great research significance and has been a hot research topic of great interest in recent years. Human pose recognition is used in artificial intelligence on the one hand and in daily life on the other. The excellent effect of pose recognition is mainly determined by the success rate and the accuracy of the recognition process, so it reflects the importance of human pose recognition in terms of recognition rate. In this human body gesture recognition, the human body is divided into 17 key points for labeling. Not only that but also the key points are segmented to ensure the accuracy of the labeling information. In the recognition design, use the comprehensive data set MS COCO for deep learning to design a neural network model to train a large number of samples, from simple step-by-step to efficient training, so that a good accuracy rate can be obtained.

• Journal of Software Assessment and Valuation
• /
• v.17 no.2
• /
• pp.1-10
• /
• 2021

The ability to recognize human emotions by computer vision is a very important task, with many potential applications. Therefore the demand for emotion recognition using not only RGB images but also thermal images is increasing. Compared to RGB images, thermal images has the advantage of being less affected by lighting conditions but require a more sophisticated recognition method with low-resolution sources. In this paper, we propose a Divide and Conquer-based CNN training strategy to improve the performance of facial thermal image-based emotion recognition. The proposed method first trains to classify difficult-to-classify similar emotion classes into the same class group by confusion matrix analysis and then divides and solves the problem so that the emotion group classified into the same class group is recognized again as actual emotions. In experiments, the proposed method has improved accuracy in all the tests than when recognizing all the presented emotions with a single CNN model.

• Journal of Platform Technology
• /
• v.12 no.2
• /
• pp.23-33
• /
• 2024

The development of mixed reality (MR) technology has a great influence on the research and development of medical support equipment. In particular, it supports to respond effectively to emergencies occurring in the field. MR technology enables access to first aid and field support by combining virtual information with the real world so that users can see virtual objects in the real world. However, due to the nature of the equipment, there is a limitation in accurately matching virtual objects based on user vision. To improve these limitations, this paper proposes a 3D biometric object recognition and matching algorithm in the MR environment. As a result of the experiment, when a virtual object is rendered and visualized while equipped with an optical-based HMD from the user's side, it was possible to reduce the user's field of view error and eliminate the joint-loss phenomenon during skeleton recognition. The proposed method can reduce errors between the real user's field of view and the virtual image and provide a basis for reducing errors that occur in the process of virtual object recognition and matching. It is expected that this study will contribute to improving the accuracy of the telemedicine support system for first aid.