• Journal of information and communication convergence engineering
• /
• v.12 no.2
• /
• pp.128-134
• /
• 2014

In this paper, we present a graphics processing unit (GPU)-based matching technique for the purpose of fast feature matching between different images. The scale invariant feature transform algorithm developed by Lowe for various feature matching applications, such as stereo vision and object recognition, is computationally intensive. To address this problem, we propose a matching technique optimized for GPUs to perform computations in less time. We optimize GPUs for fast computation of keypoints to make our system quick and efficient. The proposed method uses a self-organizing map feature matching technique to perform efficient matching between the different images. The experiments are performed on various image sets to examine the performance of the system under varying conditions, such as image rotation, scaling, and blurring. The experimental results show that the proposed algorithm outperforms the existing feature matching methods, resulting in fast feature matching due to the optimization of the GPU.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1997.04a
• /
• pp.25-29
• /
• 1997

In this thesis, we construct the microrobot succor system and navigate the real-time path planning and visual tracking of each robot. The system consists robots, vision system and a host computer. Because the robots are free-ranging mobile robot, it is needed to make and gallow the path. The path is planned and controlled by a host computer, ie. Supervisory control system. In path planning, we suggest a cost function which consists of three terms. One is the smoothness of the path, another is the total distance or time, and the last one is to avoid obstacles. To minimize the cost function, we choose the parametric cubic spline and update the coefficients in real time. We perform the simulation for the path planing and obstacle avoidance and real experiment for visual tracking

• Journal of IKEEE
• /
• v.22 no.2
• /
• pp.250-257
• /
• 2018

In this paper, a smart bus system that automates public bus traffic payment by applying beacon and computer vision and provides bus route information, real-time location information, getting off alarm is proposed. By using the beacon to recognize busses near the stop and to board the bus to be boarded, this system automatically processes the payment when boarding by using the distance from the beacon and the information provided by the beacon and the face comparison. After the payment processing, the system provides the route information of the boarded bus and the real-time bus location information to the user, and when the user sets an alarm using these informations, the alarm is activated when the bus leaves the bus stop.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.16 no.2
• /
• pp.211-218
• /
• 2021

Construction sites have various risk factors. There are various approaches to reduce safety accidents, but they have limitations to some extent. By utilizing the wireless communication technology of IT and the rapidly developing image processing technology, it will be possible to reduce accidents at the construction site if risk factors are identified and actively responded to. Therefore, in this study, a system that can detect risk factors of construction sites in advance is constructed, and a system is proposed to discover and respond to risk factors of construction sites using OpenCV for the purpose of real-time computer vision.

• International Journal of Advanced Culture Technology
• /
• v.5 no.1
• /
• pp.51-57
• /
• 2017

Digital signage is becoming part of daily life across a wide range of visual advertisements segments market used in stations, hotels, retail stores, hotels, etc. The current digital signage system used in market is generally works on limited user interactivity with static contents. In this paper, a new approach is proposed using computer vision based dynamic audience adaptive cost-effective digital signage system. The proposed design uses the Camera attached Raspberry Pi Open source platform to employ the real-time audience interaction using computer vision algorithms to extract facial features of the audience. The real-time facial features are extracted using Haar Cascade algorithm which are used for audience gender specific rendering of dynamic digital signage content. The audience facial characterization using Haar Cascade is evaluated on the FERET database with 95% accuracy for gender classification. The proposed system, developed and evaluated with male and female audiences in real-life environments camera embedded raspberry pi with good level of accuracy.

• Journal of Practical Engineering Education
• /
• v.13 no.3
• /
• pp.491-496
• /
• 2021

In this thesis, we intend to use CCTVs installed in existing parking lots to understand the current status of parking lots and provide real-time information to users through Android applications. It describes how to set the ROI in the parking area using YOLO V3 and how to provide the number of vacancies that change in real time through the set ROI, and describes how to link CCTV-server-user using IMAGE ZMQ and FIREBASE. The user can know the real-time situation of the parking lot near the destination before arriving through the application and can come up with various measures accordingly.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.36C no.8
• /
• pp.62-69
• /
• 1999

This paper presents a vision-based real-time traffic surveillance system. Current research in machine vision applied to traffic is due to its potential for more powerful process, its flexibility and its lower cost. Our traffic surveillance system consists of FPA and TSM on the Internet. The FPA provides image data, vehicle speed and vehicle density for the TSM. Traffic parameters, the vehicle speed and density, are extracted over many frames using differences between the sampling points of a background image and those of incoming frames. The FPA works by just processing groups of pixels without any understanding of the image. Although this system has an accuracy limitation, it can be used in applications requiring an approximate vehicle density and vehicle speed in real-time.

• Journal of Korea Multimedia Society
• /
• v.18 no.5
• /
• pp.583-598
• /
• 2015

We need many assembly lines to produce industrial product such as automobiles that require a lot of composited parts. Big portion of such assembly line are still operated by manual works of human. Such manual works sometimes cause critical error that may produce artifacts. Also, once the assembly is completed, it is really hard to verify whether of not the product has some error. In this paper, for monitoring behaviors of manual human work in an assembly line automatically, we proposes a realtime hybrid monitoring system that combines 2D vision sensor tracking technique with 3D motion recognition sensors.

• Proceedings of the IEEK Conference
• /
• 2001.06d
• /
• pp.239-242
• /
• 2001

As systems for real time computer vision are confronted with prodigious amounts of visual information, it has become a priority to locate and analyze just that information essential to the task at hand, while ignoring the vast flow of irrelevant detail. A method of achieving this is to using human visual attention mechanism. In this paper, short review of human visual attention mechanisms and some computation models of visual attention were shown. This paper can be used as the basic data for researches on development of visual attention system that can perform various complex tasks more efficiently.

• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10b
• /
• pp.1340-1343
• /
• 1996

In this study, we developed the drowziness prevention system to detect the driver's drowziness and relieve the driver. A computer vision method was proposed to detect the driver's drowziness. We extracted the eyes and mouth in the face image and tracked the positions from the image sequencies in real time. The eye blink duration and yawning was used as the parameters of drowziness. Wehn the drowziness state of a driver is detected, the driver is refreshed by the scent generator and the alarm. Also, the driver's bio-signal is acquired and analysed the vigilance state.