• Journal of the Korean Society for Precision Engineering
• /
• v.20 no.12
• /
• pp.96-104
• /
• 2003

To assess tunnel safety, cracks in tunnel lining are measured by inspectors who observe cracks with their eyes. A manual inspection is, however, slow and subjective. This paper, therefore, proposes vision-based inspection system for measuring cracks in the tunnel lining that inspects cracks fast and objective. The system is consisted of an on-vehicle system and a lab system. An on-vehicle system acquires image data with line CCD camera. A lab system extracts crack then inform their thickness, length and orientation by using image processing. To improve accuracy of crack recognition the geometric properties of a crack was applied to image processing. The proposed system were verified with experiments in both laboratory and field environment.

• Annual Conference of KIPS
• /
• 2017.11a
• /
• pp.716-719
• /
• 2017

This system is designed for autonomous buses. It controls buses by lane and object recognition using Deep Learning based computer vision technology. Through this system, we can expect to reduce traffic costs and increase stability.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.04b
• /
• pp.715-717
• /
• 2004

This paper proposes a new reconstruction method of high-resolution facial image from a low-resolution facial image based on a recursive error back-projection of top-down machine learning. A face is represented by a linear combination of prototypes of shape and texture. With the shape and texture information about the pixels in a given low-resolution facial image, we can estimate optimal coefficients for a linear combination of prototypes of shape and those of texture by solving least square minimization. Then high-resolution facial image can be obtained by using the optimal coefficients for linear combination of the high-resolution prototypes, In addition to, a recursive error back-projection is applied to improve the accuracy of synthesized high-resolution facial image. The encouraging results of the proposed method show that our method can be used to improve the performance of the face recognition by applying our method to reconstruct high-resolution facial images from low-resolution one captured at a distance.

• International journal of advanced smart convergence
• /
• v.8 no.1
• /
• pp.126-132
• /
• 2019

Creative thinking appears even before it is expressed in language, and its existence is revealed through emotion, intuition, image and body feeling before logic or linguistics rules work. In this study, Lego is intended to present experimental child interactive content that is applied with a computer vision based on image processing techniques. In the case of infants, the main purpose of this content is the development of hand muscles and the ability to implement imagination. The purpose of the analysis algorithm of the OpenCV library and the image processing using the 'VVVV' that is implemented as a 'Node' in the midst of perceptual changes in image processing technology that are representative of object recognition, and the objective is to use a webcam to film, recognize, derive results that match the analysis and produce interactive content that is completed by the user participating. Research shows what Lego children have made, and children can create things themselves and develop creativity. Furthermore, we expect to be able to infer a diverse and individualistic person's thinking based on more data.

• Journal of Institute of Control, Robotics and Systems
• /
• v.13 no.6
• /
• pp.542-546
• /
• 2007

In the vision of ubiquitous computing environment, smart objects would communicate each other and provide many kinds of information about user and their surroundings in the home. This information enables smart objects to recognize context and to provide active and convenient services to the customers. However in most cases, context-aware services are available only with expert systems. In this paper, we present generalized activity recognition application in the smart home based on a naive Bayesian network(BN) and fuzzy clustering. We quantize continuous sensor data with fuzzy c-means clustering to simplify and reduce BN's conditional probability table size. And we apply mutual information to learn the BN structure efficiently. We show that this system can recognize user activities about 80% accuracy in the web based virtual smart home.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.36.4-36
• /
• 2001

A robust and reliable path recognition system is one necessary component for the autonomous navigation of a mobile robot to help determining its current position in its navigation map. This paper describes a computer visual path-recognition system using on-board video camera as vision-based driving assistance for an autonomous navigation mobile robot. The common problem for a visual system is that its reliability was often influenced by different lighting conditions. Here, two different image processing methods for the path detection were developed to reduce the effect of the luminance: one is based on the RGB color model and features of the path, another is based on the HSV color model in the absence of luminance.

• Journal of Information Processing Systems
• /
• v.20 no.3
• /
• pp.328-336
• /
• 2024

Visual communication is widely used and enhanced in modern society, where there is an increasing demand for spirituality. Museum robots are one of many service robots that can replace humans to provide services such as display, interpretation and dialogue. For the improvement of museum guide robots, the paper proposes a human-robot interaction system based on visual communication skills. The system is based on a deep neural mesh structure and utilizes theoretical analysis of computer vision to introduce a Tiny+CBAM mesh structure in the gesture recognition component. This combines basic gestures and gesture states to design and evaluate gesture actions. The test results indicated that the improved Tiny+CBAM mesh structure could enhance the mean average precision value by 13.56% while maintaining a loss of less than 3 frames per second during static basic gesture recognition. After testing the system's dynamic gesture performance, it was found to be over 95% accurate for all items except double click. Additionally, it was 100% accurate for the action displayed on the current page.

• Convergence Security Journal
• /
• v.11 no.4
• /
• pp.59-66
• /
• 2011

Recently, it is very important to establish and predict a traffic policy for expanding social infra structure like road, because the number of cars is significantly increasing. In this paper, we propose and develop an automated system technology based on vision sensor (CCTV) which can provide an efficient information for the traffic policy establishment and expanding the social infra structure. First, the CCTV image is captured as an input of the developed system. With this image, we propose a scheme for extracting vehicles on the road and classifying small-type, large-type vehicles based on color, motion, and geometric features. Also, we develop a DB (database) system for supplying a whole information of traffic for a specified period. Based on the proposed system, we verify 90.1% of recognition ratio in real-time traffic monitering environment.

• The Journal of the Korea Contents Association
• /
• v.20 no.7
• /
• pp.86-96
• /
• 2020

Based on the use cases of image recognition technology, this study looked at how artificial intelligence plays a role in image recognition technology. Through image recognition technology, satellite images can be analyzed with artificial intelligence to reveal the calculation of oil storage tanks in certain countries. And image recognition technology makes it possible for searching images or products similar to images taken or downloaded by users, as well as arranging fruit yields, or detecting plant diseases. Based on deep learning and neural network algorithms, we can recognize people's age, gender, and mood, confirming that image recognition technology is being applied in various industries. In this study, we can look at the use cases of domestic and overseas image recognition technology, as well as see which methods are being applied to the industry. In addition, through this study, the direction of future research was presented, focusing on various successful cases in which image recognition technology was implemented and applied in various industries. At the conclusion, it can be considered that the direction in which domestic image recognition technology should move forward in the future.

• Annual Conference of KIPS
• /
• 2019.05a
• /
• pp.533-534
• /
• 2019

This paper is to summarize the academic status of deep learning in Chinese scientific institutions and universities based on the literatures from CNKI. We analyzed the various development of deep learning in China based on the application of computer vision, voice recognition and natural language processing.