• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.12
• /
• pp.525-534
• /
• 2023

Gas leak detection system is a key to minimize the loss of life due to the explosiveness and toxicity of gas. Most of the leak detection systems detect by gas sensors or thermal imaging cameras. To improve the performance of gas leak detection system using single-modal methods, the paper propose multimodal approach to gas sensor data and thermal camera data in developing a gas type identification model. MultimodalGasData, a multimodal open-dataset, is used to compare the performance of the four models developed through multimodal approach to gas sensors and thermal cameras with existing models. As a result, 1D CNN and GasNet models show the highest performance of 96.3% and 96.4%. The performance of the combined early fusion model of 1D CNN and GasNet reached 99.3%, 3.3% higher than the existing model. We hoped that further damage caused by gas leaks can be minimized through the gas leak detection system proposed in the study.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.9 no.5
• /
• pp.175-181
• /
• 2009

In this paper, we introduce the implementation of the security system using iris recognition. This system acquires images with infrared camera and extracts the 2D code from a infrared image which uses scale-space filtering and concavity. We examine the system by (i) extract 2D code and (ii) compare the code that stored on the server (iii) mearsure FAR and FRR using pattern matching. Experiment results show that the proposed method is very suitable.

• Korean Journal of Computational Design and Engineering
• /
• v.18 no.4
• /
• pp.283-289
• /
• 2013

Presented in this paper is an adaptive measuring procedure of dental impression using the Structured Light System. While measuring a complex object, such as dental impression, in the reverse engineering, it is not possible to acquire all parts of the scanned surface. Missing scanned data is resulted in holes in a created triangular mesh. The focus of this paper is to introduce an algorithm for automatic identification of additional scanning orientations to fill holes that are created by a default scan. The proposed algorithm was developed by the three major technological requirements: camera visibility, projector visibility, data reliability. In order to satisfy the requirements, the proposed algorithm determines additional scanning orientation from the orientation of a projection plane derived from the average normal vector of boundary triangles.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.22 no.4
• /
• pp.723-729
• /
• 2013

The contraction and expansion noise of a refrigerator are investigated, and some effective methods are proposed to reduce the level and occurrence frequency of noise. First, the noise of a refrigerator is measured to estimate the frequency spectrum and occurrence frequency of noise. Second, a sound visualization was conducted using an acoustic camera to determine the location of the noise source. From the results, it was observed that the internal part mainly producing noise was the third shelf in the freezer room. A method to estimate the acceleration on the location of the noise source is introduced to analyze the contraction and expansion noise precisely and accommodate experimental convenience. Noise reduction methods such as the replacement of the existing shelf with glass shelves, adoption of rail slides, and increase of roughness on the contact surface of the shelf are proposed and tested.

• Proceedings of the Korea Society of Information Technology Applications Conference
• /
• 2005.11a
• /
• pp.153-156
• /
• 2005

Skin region detection in images is an important process in many computer vision applications targeting humans such as hand gesture recognition and face identification. It usually starts at a pixel-level, and involves a pre-process of color spae transformation followed by a classification process. A color space transformation is assumed to increase separability between skin classes and other classes, to increase similarity among different skin tones, and to bring a robust performance under varying imaging conditions, without any complicated analysis. In this paper, we examine if the color space transformation actually brings those benefits to the problem of skin region detection on a set of human hand images with different postures, backgrounds, people, and illuminations. Our experimental results indicate that color space transfomation affects the skin detection performance. Although the performance depends on camera and surround conditions, normalized [R, G, B] color space may be a good choice in general.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.6 no.4
• /
• pp.517-522
• /
• 2011

Conventional projection welding systems suffer from lots of defective products caused by manual handling. In this paper, we introduce a projection welding system that performs automatic identification, welding and counting of components and products. The proposed system checks the existence and identifies placement of components to be welded by a vision camera. After welding of the components, it automatically updates product counts and dressing items. We show that the proposed welding system can reduce the defect rate and improve the productivity through experimental test with a existing system.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2008.04c
• /
• pp.6-9
• /
• 2008

A 3D depth measurement system is proposed for mobile vehicles. Depth measurement system which is composed of a rotating mirror, a line laser and mono-camera is employed to detect depth, where the laser light is reflected by the mirror and projected to- the scene objects whose locations are to be determined. The obtained depth information is converted into an image. Such depth images of the road region represent even and plane while that of off-road region is irregular or textured. Road region is detected employing a simple spatial differentiation technique to detect the plain textured area. Identification results of the diverse situation of Non-linear trail are included in this paper.

• International Journal of Aeronautical and Space Sciences
• /
• v.7 no.2
• /
• pp.86-103
• /
• 2006

Using a combination of force transducer measurement to quantify net lift force, high frame rate camera to quantify and subtract inertial contributions, and Digital Particle Image Velocimetry (DPIV) to calculate aerodynamic contributions in the spanwise plane, the contribution of spanwise flow to the generation of lift force in wings undergoing a pure flapping motion in hover is shown as a function of flapping angle throughout the flapping cycle. These experiments were repeated at various flapping frequencies and for various wing planform sizes for flat plate and span wise cambered wings. Despite the previous identification of the importance of span wise fluid structures in the generation of lift force in flapping wings throughout the existing body of literature, the direct contribution of spanwise flow to lift force generated has not previously been quantified. Therefore, in the same manner as commonly applied to investigate the chordwise lift distribution across an airfoil in flapping wings, spanwise flow due to bulk flow and rotational fluid dynamic mechanisms will be investigated to validate the existence of a direct component of the lift force originating from the flapping motion in the spanwise plane instead.

• International Journal of Control, Automation, and Systems
• /
• v.5 no.1
• /
• pp.51-60
• /
• 2007

This paper describes a visual tracking control law of an Unmanned Aerial Vehicle(UAV) for monitoring of structures and maintenance of bridges. It presents a control law based on computer vision for quasi-stationary flights above a planar target. The first part of the UAV's mission is the navigation from an initial position to a final position to define a desired trajectory in an unknown 3D environment. The proposed method uses the homography matrix computed from the visual information and derives, using backstepping techniques, an adaptive nonlinear tracking control law allowing the effective tracking and depth estimation. The depth represents the desired distance separating the camera from the target.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.5 no.3
• /
• pp.111-117
• /
• 2010

The easily accessible handheld devices equipped with camera are widely available as common commodities. According to this trend, utilization of images is popular among common users for various purposes resulting in huge amount of images in local or network based storage systems. In this environment, identification of an image with a solid and effective manner is demanded in behalf of safe distribution and efficient management of images. The generated identifiers can be used as a file name in file systems or an index in image databases utilizing the uniqueness of the identifiers. In this paper, we propose a method that generates image identifiers using linear components in images. Some experiments of generation of identifiers are performed, and the results evaluate that the proposed method has feasible effectiveness.