• IEIE Transactions on Smart Processing and Computing
• /
• v.5 no.3
• /
• pp.153-163
• /
• 2016

Local feature extraction methods for images and videos are widely applied in the fields of image understanding and computer vision. However, robust features are detected differently when using the latest feature detectors and descriptors because of diverse image environments. This paper analyzes various feature extraction methods by summarizing algorithms, specifying properties, and comparing performance. We analyze eight feature extraction methods. The performance of feature extraction in various image environments is compared and evaluated. As a result, the feature detectors and descriptors can be used adaptively for image sequences captured under various image environments. Also, the evaluation of feature detectors and descriptors can be applied to driving assistance systems, closed circuit televisions (CCTVs), robot vision, etc.

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.11
• /
• pp.23-30
• /
• 2016

In this paper, we propose an adaptive histogram projection technique for dynamic range compression and an efficient detail enhancement method which is enhancing strong edge while reducing noise. First, The high dynamic range image is divided into low-pass component and high-pass component by applying 'guided image filtering'. After applying 'guided filter' to high dynamic range image, second, the low-pass component of the image is compressed into 8-bit with the adaptive histogram projection technique which is using global standard deviation value of whole image. Third, the high-pass component of the image adaptively reduces noise and intensifies the strong edges using standard deviation value in local path of the guided filter. Lastly, the monitor display image is summed up with the compressed low-pass component and the edge-intensified high-pass component. At the end of this paper, the experimental result show that the suggested technique can be applied properly to the IR images of various scenes.

• Proceedings of the KOSOMBE Conference
• /
• v.1997 no.05
• /
• pp.465-468
• /
• 1997

A Study on Computer Assisted Diagnosis (CAD) system extract ing lung cancer part from Digital X-ray Computerized Tomography(CT) image is discussed in this paper. It is very crucial to segment the image of lung into the three organ area such as inside, outside and the hilum so that the variant image processing algorithm can be applied an each area respectively. In this paper, the efficient algorithm extracting lung cancer part is proposed with characterizing lung hilum part and its associated vessel patterns.

• Proceedings of the KSRS Conference
• /
• 2004.10a
• /
• pp.242-244
• /
• 2004

In this paper, we design and implement an integrated management system for geographic data and vehicular images using a Geographic Information System (GIS). Integrated management of geographic data and vehicular images is very important to manage and to provide them to users effectively because of a large volume of vehicular images. To manipulate these data together, we consider a vehicular image as a polygon which is a type of popular geographic data types. The polygon represents a region in which spatial objects appear the vehicular image.

• 한국정보디스플레이학회:학술대회논문집
• /
• 2002.08a
• /
• pp.153-156
• /
• 2002

We have developed a pixel structure for reflective color TFT-LCD which can display 27-color in still-image. The proposed pixel can display 3 gray scale in still image; white, black and median gray. This paper shows the concept and the driving method of the proposed pixel. Finally this paper compares power consumption and area with the Toshiba's DMOG technology.

• 한국정보디스플레이학회:학술대회논문집
• /
• 2008.10a
• /
• pp.873-878
• /
• 2008

Doosan Mecatec has developed alignment system for Organic Light-Emitting Diode (OLED) display production using large size substrate. In the present article, The alignment system between the substrate and the mask, which is a core technology for producing the OLED product using the fourth-generation substrate with $730{\times}920mm^2$ or more, will be described by dividing into a substrate loader, a magnet unit, a CCD camera, etc. The substrate loader is optimized through the simulation where the central portion of the substrate droops by about 1.5mm by clamping each of a long side (920mm direction) and a short side (730mm direction) thereof by 6 point and 4 point. A magnet unit using a sheet type of rubber magnet is constituted and a CCD camera model with the specifications capable of minimizing the errors between a clear image and the same image is selected. The system to which an upward evaporation technique of small molecular organic materials will be applied has been developed so that repeatability and position accuracy becomes ${\pm}1{\mu}m$ or less using an UVW type of stage. Also, the vision accuracy of the CCD camera becomes ${\pm}1{\mu}m$ or less and the align process TACT becomes 30sec. or less so that the final alignment accuracy between the substrate and the mask becomes ${\pm}3{\mu}m$ or less. In order to meet an extra-large glass substrate, an evaporation system using an extra-large AMOLED substrate has been developing through a vertical type of an alignment system.

• Journal of Sensor Science and Technology
• /
• v.28 no.3
• /
• pp.157-163
• /
• 2019

The LiDAR sensor, which is widely regarded as one of the most important sensors, has recently undergone active commercialization owing to the significant growth in the production of ADAS and autonomous vehicle components. The LiDAR sensor technology involves radiating a laser beam at a particular angle and acquiring a three-dimensional image by measuring the lapsed time of the laser beam that has returned after being reflected. The LiDAR sensor has been incorporated and utilized in various devices such as drones and robots. This study focuses on object detection and recognition by employing sensor fusion. Object detection and recognition can be executed as a single function by incorporating sensors capable of recognition, such as image sensors, optical sensors, and propagation sensors. However, a single sensor has limitations with respect to object detection and recognition, and such limitations can be overcome by employing multiple sensors. In this paper, the performance of an eight-channel scanning LiDAR was evaluated and an object detection algorithm based on it was implemented. Furthermore, object detection characteristics during daytime and nighttime in a real road environment were verified. Obtained experimental results corroborate that an excellent detection performance of 92.87% can be achieved.

• Journal of Internet Computing and Services
• /
• v.19 no.4
• /
• pp.45-52
• /
• 2018

A single-image super-resolution is a process of restoring a high-resolution image from a low-resolution image. Recently, the super-resolution using the deep neural network has shown good results. In this paper, we propose a neural network structure that improves speed and performance over conventional neural network based super-resolution methods. To do this, we analyze the conventional neural network based super-resolution methods and propose solutions. The proposed method reduce the 5 stages of the conventional method to 3 stages. Then we have studied the optimal width and depth by experimenting on the width and depth of the network. Experimental results have shown that the proposed method improves the disadvantages of the conventional methods. The proposed neural network structure showed superior performance and speed than the conventional method.

• Journal of Fashion Business
• /
• v.16 no.4
• /
• pp.64-74
• /
• 2012

The purpose of this study would be to find out the relationship of social as a sign of fashion image and the followership. This study is classified into theoretical and experimental research. Following are the summary of the results revealed through the experimental study. First, The relationship of oneself pursuit of fashion image types and leader's favourite fashion image types for regression analysis result indicated significant difference. Second, The factor analysis of followership are used, developed by Colangeol is asking. The results of factor analysis are four types classification as to Active Participation, Convergence objective, Team Spirit, Critical Thinking. Third, The relationship of types of fashion images and factor variance of followership indicated a difference in Active Participation factors. But The relationship of types of leader's fashion images and factor variance of followership indicated a difference in Convergence objective factors. Analysis of the fashion image based on the conceptual properties of followership is to understand the characteristics of followers, and the leader's image based on research for building materials will be provided.

• Journal of the Korea Society of Computer and Information
• /
• v.6 no.3
• /
• pp.143-151
• /
• 2001

Fashion Industry is one of the Knowledge Based Industry. The firms which produce the fashion merchandises should shirt to new paradigm of creating new image to delight customers. New image and design is derived from each designer's creative mind. However, self management team work of designers is more creative power than each independent designer's work. This study recommends that the Korean fashion firms implement self management team, concurrent product design and development and Just-in-Time Ⅱ based on sharing information and knowledge with the supplier to set competitive advantage. It can not only reduce cost and improve speed of management and the qualify of products but also satisfy and delight customers.