• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.2
• /
• pp.692-712
• /
• 2022

Buoys used for Aid to Navigation systems are widely used to guide the sea paths and are powered by batteries, requiring continuous battery replacement. However, since human labor is required to replace the batteries, humans can be exposed to dangerous situation, including even collision with shipping vessels. In addition, Maritime sensors are installed on the route signs, so that these are often damaged by collisions with small and medium-sized ships, resulting in significant financial loss. In order to prevent these accidents, maritime object detection technology is essential to alert ships approaching buoys. Existing studies apply a number of filters to eliminate noise and to detect objects within the sea image. For this process, most studies directly access the pixels and process the images. However, this approach typically takes a long time to process because of its complexity and the requirements of significant amounts of computational power. In an emergent situation, it is important to alarm the vessel's rapid approach to buoys in real time to avoid collisions between vessels and route signs, therefore minimizing computation and speeding up processes are critical operations. Therefore, we propose Fast Connected Component Labeling (FCCL) which can reduce computation to minimize the processing time of filter applications, while maintaining the detection performance of existing methods. The results show that the detection performance of the FCCL is close to 30 FPS - approximately 2-5 times faster, when compared to the existing methods - while the average throughput is the same as existing methods.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.11a
• /
• pp.548-551
• /
• 2020

최근 의료시장에 상용화된 모바일 수술 중 CT(intra-operative computed tomography, iCT)는 수술실 내 이동이 자유로울 뿐만 아니라 수술 후 즉각적인 환자 모니터링이 실시간으로 이루어져 수술 후 환자의 예후 향상과 재수술 확률을 낮출 수 있다. 이동성을 갖춘 iCT 는 편의성과 유용성이 검증되었지만, 이동시 발생될 수 있는 충돌사고의 단점이 존재한다. 따라서, iCT 의 이동시 발생할 수 있는 위험을 최소화 할 수 있는 안전장치가 요구된다. 본 연구에서는 모바일 iCT 의 구동 제어의 편의성과 안전성을 확보 할 수 있는 CT 촬영을 제어하기 위한 리모트 컨트롤러, 이동시 전방 시야를 확보하기 위한 전방 모니터링 카메라 출력, 충돌 위험을 알릴 수 있는 초음파 센서를 통합하는 임베디드 컨트롤러를 개발하고자 한다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2016.05a
• /
• pp.499-501
• /
• 2016

In this paper, we suggest an advance method for maintaining a perceived behavior as triangle formation and preventing collision between each other in case of a flying drone. In the existing studies, the collision of the drone is only controlled by using light entered in the camera or the image processing. However, when there is no light, it is difficult to confirm the position of each other and they can collide because this system can not confirm the each other's position. Therefore, in this paper, we propose the system to solve the problems by using the distance and the relative coordinates of the three drones that were determined using the ALPS(Ad hoc network Localized Positioning System) algorithm. This system can be a new algorithm that will prevent collisions between each other during flying the drone object. The proposed algorithm is that we make drones maintaining a determined constant value of the distance between coordinates of each drone and the measured center of the drone of triangle formation. Therefore, if the form of fixed formation is disturbed, they reset the position of the drone so as to keep the distance between each drone and the center coordinates constant. As a result of the simulation, if we use the system where the supposed algorithm is applied, we can expect that it is possible to prevent malfunction or an accident due to collisions by preventing collisions of drones in advanced behavior system.

• International conference on construction engineering and project management
• /
• 2022.06a
• /
• pp.1243-1244
• /
• 2022

In recent years, the growing interest in off-site construction has led to factories scaling up their manufacturing and production processes in the construction sector. Consequently, continuous large-scale site monitoring in low-variability environments, such as prefabricated components production plants (precast concrete production), has gained increasing importance. Although many studies on computer vision-based site monitoring have been conducted, challenges for deploying this technology for large-scale field applications still remain. One of the issues is collecting and transmitting vast amounts of video data. Continuous site monitoring systems are based on real-time video data collection and analysis, which requires excessive computational resources and network traffic. In addition, it is difficult to integrate various object information with different sizes and scales into a single scene. Various sizes and types of objects (e.g., workers, heavy equipment, and materials) exist in a plant production environment, and these objects should be detected simultaneously for effective site monitoring. However, with the existing object detection algorithms, it is difficult to simultaneously detect objects with significant differences in size because collecting and training massive amounts of object image data with various scales is necessary. This study thus developed a large-scale site monitoring system using edge computing and a small-object detection system to solve these problems. Edge computing is a distributed information technology architecture wherein the image or video data is processed near the originating source, not on a centralized server or cloud. By inferring information from the AI computing module equipped with CCTVs and communicating only the processed information with the server, it is possible to reduce excessive network traffic. Small-object detection is an innovative method to detect different-sized objects by cropping the raw image and setting the appropriate number of rows and columns for image splitting based on the target object size. This enables the detection of small objects from cropped and magnified images. The detected small objects can then be expressed in the original image. In the inference process, this study used the YOLO-v5 algorithm, known for its fast processing speed and widely used for real-time object detection. This method could effectively detect large and even small objects that were difficult to detect with the existing object detection algorithms. When the large-scale site monitoring system was tested, it performed well in detecting small objects, such as workers in a large-scale view of construction sites, which were inaccurately detected by the existing algorithms. Our next goal is to incorporate various safety monitoring and risk analysis algorithms into this system, such as collision risk estimation, based on the time-to-collision concept, enabling the optimization of safety routes by accumulating workers' paths and inferring the risky areas based on workers' trajectory patterns. Through such developments, this continuous large-scale site monitoring system can guide a construction plant's safety management system more effectively.

• Journal of Biomedical Engineering Research
• /
• v.33 no.1
• /
• pp.25-31
• /
• 2012

The audio vision system was developed for visually impaired people and usability was verified. In this study ten normal volunteers were included in the subject group and their mean age was 28.8 years old. Male and female ratio was 7:3. The usability of audio vision system was verified by as follows. First, volunteers learned distance of obstacles and up-down discrimination. After learning of audio vision system, indoor and outdoor walking examination was performed. The test was scored by ability of up-down and lateral discrimination, distance recognition and walking without collision. Each parameter was scored by 1 to 5. The results were 93.5 +- SD(ranges, 86 to 100) of 100. In this study, we could convert visual information to auditory information by audio-vision system and verified possibility of applying to daily life for visually impaired people.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.37 no.2
• /
• pp.161-168
• /
• 2013

Nowadays, one of the most interested area of automobile industry is the production of vehicle which has collision safety and ability to produce less amount of $CO_2$. The achievement of such a dual performance is done by choosing the materials like dual phase steel, ferrite bainite steel, etc. These steels have been used in automotive chassis and body parts, and also used to be formed by hole flanging to meet the goal of strength and design requirement. The formability of sheet material was experimented by hole expansion test and the judgement relies on human eye and his experience. This manual judgement involves many errors and large deviation. This paper develops the automatic crack recognition system which finds a crack based on CCD image to complement the problem of the current method depending on human's sense.

• Journal of Computational Design and Engineering
• /
• v.1 no.2
• /
• pp.140-151
• /
• 2014

Storing, and the loading and unloading of materials at production sites in the manufacturing sector for mass production is a critical problem that affects various aspects: the layout of the factory, line-side space, logistics, workers' work paths and ease of work, automatic procurement of components, and transfer and supply. Traditionally, the nesting problem has been an issue to improve the efficiency of raw materials; further, research into mainly 2D optimization has progressed. Also, recently, research into the expanded usage of 3D models to implement packing optimization has been actively carried out. Nevertheless, packing algorithms using 3D models are not widely used in practice, due to the large decrease in efficiency, owing to the complexity and excessive computational time. In this paper, the problem of efficiently loading and unloading freeform 3D objects into a given container has been solved, by considering the 3D form, ease of loading and unloading, and packing density. For this reason, a Group Packing Approach for workers has been developed, by using analyzed truck packing work patterns and Group Technology, which is to enhance the efficiency of storage in the manufacturing sector. Also, an algorithm for 3D packing has been developed, and implemented in a commercial 3D CAD modeling system. The 3D packing method consists of a grouping algorithm, a sequencing algorithm, an orientating algorithm, and a loading algorithm. These algorithms concern the respective aspects: the packing order, orientation decisions of parts, collision checking among parts and processing, position decisions of parts, efficiency verification, and loading and unloading simulation. Storage optimization and examination of the ease of loading and unloading are possible, and various kinds of engineering analysis, such as work performance analysis, are facilitated through the intelligent 3D packing method developed in this paper, by using the results of the 3D model.

• Journal of Institute of Control, Robotics and Systems
• /
• v.20 no.12
• /
• pp.1272-1277
• /
• 2014

This article describes a path planning algorithm for fixed-wing UAVs when a real terrain should be considered. Nowadays, many UAVs are required to perform mission flights near given terrain for surveillance, reconnaissance, and infiltration, as well as flight altitude of many UAVs are relatively lower than typical manned aerial vehicles. Therefore, real terrain should be considered in path planning algorithms of fixed-wing UAVs. In this research, we have extended a spline-$RRT^*$ algorithm to three-dimensional planner. The spline-$RRT^*$ algorithm is a $RRT^*$ based algorithm, and it takes spline method to extend the tree structure over the workspace to generate smooth paths without any post-processing. Direction continuity of the resulting path is guaranteed via this spline technique, and it is essential factor for the paths of fixed-wing UAVs. The proposed algorithm confirm collision check during the tree structure extension, so that generated path is both geometrically and dynamically feasible in addition to direction continuity. To decrease degrees of freedom of a random configuration, we designed a function assigning directions to nodes of the graph. As a result, it increases the execution speed of the algorithm efficiently. In order to investigate the performance of the proposed planning algorithm, several simulations are performed under real terrain environment. Simulation results show that this proposed algorithm can be utilized effectively to path planning applications considering real terrain.

• The KIPS Transactions:PartC
• /
• v.14C no.7
• /
• pp.589-596
• /
• 2007

Although flooding has the disadvantages like a transmission of duplicated packets and a packet collision, it has been used frequently to find a path between a source and a sink node in a wireless ad hoc network. Clustering is one of the techniques that have been proposed to overcome those disadvantages. In this paper, we propose a new flooding mechanism in ad hoc networks using cluster formation based on the link density which means the number of neighbors within a node's radio reach. To reduce traffic overhead in the cluster is to make the number of non-flooding nodes as large as possible. Therefore, a node with the most links in a cluster will be elected as cluster header. This method will reduce the network traffic overhead with a reliable network performance. Simulation results using NS2 show that cluster formation based on the link density can reduce redundant flooding without loss of network performance.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.7 no.4
• /
• pp.163-172
• /
• 2012

A driver assistance system is critical to improve a convenience and stability of vehicle driving. Several systems have been already commercialized such as adaptive cruise control system and forward collision warning system. Efficient vehicle detection is very important to improve such driver assistance systems. Most existing vehicle detection systems are based on a radar system, which measures distance between a host and leading (or oncoming) vehicles under various weather conditions. However, it requires high deployment cost and complexity overload when there are many vehicles. A camera based vehicle detection technique is also good alternative method because of low cost and simple implementation. In general, night time vehicle detection is more complicated than day time vehicle detection, because it is much more difficult to distinguish the vehicle's features such as outline and color under the dim environment. This paper proposes a method to detect vehicles at night time using analysis of a captured color space with reduction of reflection and other light sources in images. Four colors spaces, namely RGB, YCbCr, normalized RGB and Ruta-RGB, are compared each other and evaluated. A suboptimal threshold value is determined by Otsu algorithm and applied to extract candidates of taillights of leading vehicles. Statistical features such as mean, variance, skewness, kurtosis, and entropy are extracted from the candidate regions and used as feature vector for SVM(Support Vector Machine) classifier. According to our simulation results, the proposed statistical feature based SVM provides relatively high performances of leading vehicle detection with various distances in variable nighttime environments.