• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.68-73
• /
• 2005

Various techniques have been proposed for detection and tracking of targets in order to develop a real-world computer vision system, e.g., visual surveillance systems, intelligent transport systems (ITSs), and so forth. Especially, the idea of distributed vision system is required to realize these techniques in a wide-spread area. In this paper, we develop a ubiquitous vision system for location-awareness of multiple targets. Here, each vision sensor that the system is composed of can perform exact segmentation for a target by color and motion information, and visual tracking for multiple targets in real-time. We construct the ubiquitous vision system as the multiagent system by regarding each vision sensor as the agent (the vision agent). Therefore, we solve matching problem for the identity of a target as handover by protocol-based approach. We propose the identified contract net (ICN) protocol for the approach. The ICN protocol not only is independent of the number of vision agents but also doesn't need calibration between vision agents. Therefore, the ICN protocol raises speed, scalability, and modularity of the system. We adapt the ICN protocol in our ubiquitous vision system that we construct in order to make an experiment. Our ubiquitous vision system shows us reliable results and the ICN protocol is successfully operated through several experiments.

• Proceedings of the KSR Conference
• /
• 2010.06a
• /
• pp.2122-2136
• /
• 2010

Existing model formula in the 4 phase model is limited in the estimation of the demand for urban railway because the administrative region-based formula reflects no spatial characteristics of station surrounding area(SSA) that urban railway forms. The purpose of this study is both to analyse the behavior in selecting the method regarding spatial range of SSA and to do the basic research for the development of new model through the survey conducted in the stations of the metropolitan area. This study will review the domestic and foreign cases about designation of SSA, study the spatial range of SSA through case studies, analyze the selection of methods by the spatial range and estimate the demand of the station on the basis of social and economic indices regarding SSA. This study focuses on the verification of real results and model estimates, due to the time constraint and lack of resources for collecting and analysing the data. According to this study, 500m,1000m division of SSA shows the closest results of the model estimates to the real demand of the targeted stations.

• Journal of Digital Contents Society
• /
• v.13 no.2
• /
• pp.235-245
• /
• 2012

This paper introduce a quiz game contents based on gesture interface. We analyzed the off-line quiz games, extracted its presiding components, and digitalized them so that the proposed game contents is able to substitute for the off-line quiz games. We used the Kinect camera to obtain the depth images and performed the preprocessing including vertical human segmentation, head detection and tracking and hand detection, and gesture recognition for hand-up, hand vertical movement, fist shape, pass and fist-and-attraction. Especially, we defined the interface gestures designed as a metaphor for natural gestures in real world so that users are able to feel abstract concept of movement, selection and confirmation tangibly. Compared to our previous work, we added the card compensation process for completeness, improved the vertical hand movement and the fist shape recognition methods for the example selection and presented an organized test to measure the recognition performance. The implemented quiz application program was tested in real time and showed very satisfactory gesture recognition results.

• The KIPS Transactions:PartB
• /
• v.15B no.6
• /
• pp.561-574
• /
• 2008

The spot detection phase of the 2-DE image analysis program segments a gel image into spot regions by an image segmentation algorithm and fits the spot regions to a spot shape model and quantifies the spot informations for the next phases. Currently the watershed algorithm is generally used as the segmentation algorithm and there are the Gaussian model and the diffusion model for the shape model. The diffusion model is closer to real spot shapes than the Gaussian model however spots have very various shapes and especially an asymmetric formation in x-coordinate and y-coordinate. The reason for asymmetric formation of spots is known that a protein could not be diffused completely because the 2-DE could not be processed under the ideal environment usually. Accordingly we propose an asymmetric diffusion model in this paper. The asymmetric diffusion model assumes that a protein spot is diffused from a disc at initial time of diffusing process, but is diffused asymmetrically for x-axis and y-axis respectively as time goes on. In experiments we processed spot matching for 19 gel images by using three models respectively and evaluated averages of SNR for comparing three models. As averages of SNR we got 14.22dB for the Gaussian model, 20.72dB for the diffusion model and 22.85dB for the asymmetric diffusion model. By experimental results we could confirm the asymmetric diffusion model is more efficient and more adequate for spot matching than the Gaussian model and the diffusion model.

• Journal of Broadcast Engineering
• /
• v.20 no.3
• /
• pp.449-463
• /
• 2015

Video object tracking is important for variety of applications, such as security, video indexing and retrieval, video surveillance, communication, and compression. This paper proposes an object tracking method in HEVC bitstreams. Without pixel reconstruction, motion vector (MV) and size of prediction unit in the bitstream are employed in an Spatio-Temporal Markov Random Fields (ST-MRF) model which represents the spatial and temporal aspects of the object's motion. Coefficient-based object shape adjustment is proposed to solve the over-segmentation and the error propagation problems caused in other methods. In the experimental results, the proposed method provides on average precision of 86.4%, recall of 79.8% and F-measure of 81.1%. The proposed method achieves an F-measure improvement of up to 9% for over-segmented results in the other method even though it provides only average F-measure improvement of 0.2% with respect to the other method. The total processing time is 5.4ms per frame, allowing the algorithm to be applied in real-time applications.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.1A
• /
• pp.80-88
• /
• 2002

Visual media(image, graphic, and video) processing poses challenge from several perpectives, specifically from the point of view of real-time implementation and scalability. There have been several approaches to obtain speedups to meet the computing demands in multimedia processing ranging from media processors to special purpose implementations. A variety of parallel processing strategies are adopted in these implementations in order to achieve the required speedups. We have investigated a parallel processing system for improving the processing speed o f visual media related applications. The parallel processing system we proposed is similar to a pipelined memory stystem(MAMS). The multi-access memory system is made up of m memory modules and a memory controller to perform parallel memory access with a variety of combinations of 1${\times}$pq, pq${\times}$1, and p${\times}$q subarray, which improves both cost and complexity of control. Facial recognition, Phong shading, and automatic segmentation of moving object in image sequences are some that have been applied to the parallel processing system and resulted in faithful processing speed. This paper describes the parallel processing systems for the speedup and its utilization to three time-consuming applications.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.30 no.2
• /
• pp.147-156
• /
• 2024

The sustainable management and enhancement of marine resources are becoming increasingly important issues worldwide. This study was conducted in response to these challenges, focusing on the development and performance comparison of fish detection and classification models as part of a deep learning-based technique for assessing the effectiveness of marine resource enhancement projects initiated by the Korea Fisheries Resources Agency. The aim was to select the optimal model by training various sizes of YOLOv8-Seg models on a fish image dataset and comparing each performance metric. The dataset used for model construction consisted of 36,749 images and label files of 12 different species of fish, with data diversity enhanced through the application of augmentation techniques during training. When training and validating five different YOLOv8-Seg models under identical conditions, the medium-sized YOLOv8m-Seg model showed high learning efficiency and excellent detection and classification performance, with the shortest training time of 13 h and 12 min, an of 0.933, and an inference speed of 9.6 ms. Considering the balance between each performance metric, this was deemed the most efficient model for meeting real-time processing requirements. The use of such real-time fish detection and classification models could enable effective surveys of marine resource enhancement projects, suggesting the need for ongoing performance improvements and further research.

• Journal of Internet Computing and Services
• /
• v.25 no.5
• /
• pp.131-140
• /
• 2024

Aids to navigation, such as buoys used in maritime environments, play a crucial role in providing accurate information to navigating vessels, enabling them to precisely determine their position and maintain safe routes by marking surrounding hazardous areas. However, collisions between ships and these aids result in substantial costs for buoy damage and repair. While high-end equipment is currently used to prevent such accidents, its widespread adoption is hindered by cost concerns. This paper presents research on a maritime object detection algorithm utilizing embedded systems to address this issue. Previous studies employed the Hough transform for horizon detection, but its high computational demands posed challenges for real-time processing. To overcome this limitation, our approach first performs image segmentation, followed by an optimized Otsu algorithm for horizon detection. Subsequently, we establish a Region of Interest (ROI) based on the detected horizon, focusing on areas with a high risk of ship collision. Within this ROI, particularly below the horizon line, maritime objects are detected. A Convolutional Neural Network (CNN) model is then applied to determine whether the detected objects are ships. Objects classified as ships within the ROI are considered potential collision risks.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.37 no.2
• /
• pp.169-176
• /
• 2013

To generate the walking path of a humanoid robot in an unknown environment, the shapes of obstacles around the robot should be detected accurately. However, doing so incurs a very large computational cost. Therefore this study proposes a method to classify the obstacle shape into three types: a shape small enough for the robot to go over, a shape planar enough for the robot foot to make contact with, and an uncertain shape that must be avoided by the robot. To classify the obstacle shape, first, the range and the number of the obstacles is detected. If an obstacle can make contact with the robot foot, the shape of an obstacle is accurately derived. If an obstacle has uncertain shape or small size, the shape of an obstacle is not detected to minimize the computational load. Experimental results show that the proposed algorithm efficiently classifies the shapes of obstacles around the robot in real time with low computational load.

• Journal of Satellite, Information and Communications
• /
• v.9 no.1
• /
• pp.57-64
• /
• 2014

Recently, many human-friendly communication methods have been studied for human-machine interface(HMI) without using any physical devices. One of them is the vision-based gesture recognition that this paper deals with. In this paper, we define some gestures for interaction with objects in a predefined virtual world, and propose an efficient method to recognize them. For preprocessing, we detect and track the both hands, and extract their silhouettes from the low-resolution hand images captured by a webcam. We modeled skin color by two Gaussian distributions in RGB color space and use blob-matching method to detect and track the hands. Applying the foodfill algorithm we extracted hand silhouettes and recognize the hand shapes of Thumb-Up, Palm and Cross by detecting and analyzing their modes. Then, with analyzing the context of hand movement, we recognized five predefined one-hand or both-hand gestures. Assuming that one main user shows up for accurate hand detection, the proposed gesture recognition method has been proved its efficiency and accuracy in many real-time demos.