• Korean Journal of Computational Design and Engineering
• /
• v.15 no.1
• /
• pp.33-42
• /
• 2010

In this study, new medial surface calculation methods using Voronoi diagrams are investigated for the point samples extracted on closed surface models. The medial surface is defined by the closure of all points having more than one closest point on the shape boundary. It is a one of essential geometric information in 3D and can be used in many areas such as 3D shape analysis, dimension reduction, freeform shape deformation, image processing, computer vision, FEM analysis, etc. In industrial parts, the idealized solid parts and shell shapes including sharp edges and vertices are frequently used. Other medial surface extraction methods using Voronoi diagram have inherent separation and branch problems, so that they are not appropriate to the sharp edged objects and have difficulties to be applied to industrial parts. In addition, the branched surfaces on sharp edges in shell shapes should be eliminated to obtain representative medial shapes. In order to avoid separation and branch problems, the new approach by analyzing the shapes and specially sampling on surfaces has been developed.

• Journal of Korea Society of Industrial Information Systems
• /
• v.10 no.2
• /
• pp.102-108
• /
• 2005

CIELAB coordinate is represented by one lightness component and two chromaticity components and similar to human visual system. Visual devices such as computer monitor display images using RGB coordinate. We propose a technique for inserting the watermark of visually recognizable mark into the middle frequency domain of image. RGB coordinate image is transformed into CIELAB coordinate, which include the characteristics of Human vision and then a* component is transformed into DFT(Discrete Fourier transform) transform.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.29 no.4
• /
• pp.107-112
• /
• 2006

This study presents the new dispatching rules for improving performance measures of job shop scheduling related to completion time and due dates. The proposed dispatching rule considers information, which includes the comparison value of job workload, work remaining, operation time, and operation due dates. Through computer experiments, the performance of the new dispatching rules is compared and analyzed with the existing rules. The results provide a guidance for the researchers to develop new dispatching rules and for practitioners to choose rules of job shop scheduling.

• International Journal of Advanced Culture Technology
• /
• v.10 no.1
• /
• pp.302-309
• /
• 2022

Occupational safety accidents are caused by various factors, and it is difficult to predict when and why they occur, and it is directly related to the lives of workers, so the interest in safety accidents is increasing every year. Therefore, in order to reduce safety accidents at industrial fields, workers are required to wear personal protective equipment. In this paper, we proposes a method to automatically check whether workers are wearing safety helmets among the protective equipment in the industrial field. It detects whether or not the helmet is worn using YOLOv5, a computer vision-based deep learning object detection algorithm. We transfer learning the s model among Yolov5 models with different learning rates and epochs, evaluate the performance, and select the optimal model. The selected model showed a performance of 0.959 mAP.

• Proceedings of the IEEK Conference
• /
• 2002.07a
• /
• pp.273-276
• /
• 2002

Pulse-Coupled Neural Network (PCNN), which can explain the synchronous burst of neurons in a cat visual cortex, is a fundamental model for the biomimetic vision. The PCNN is a kind of pulse coded neural network models. In order to get deep understanding of the visual information Processing, it is important to simulate the visual system through such biologically plausible neural network model. In this paper, we construct the motion detection model based on the PCNN with the receptive field models of neurons in the lateral geniculate nucleus and the primary visual cortex. Then it is shown that this motion detection model can detect the movements and the direction of motion effectively.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2003.05a
• /
• pp.409-412
• /
• 2003

The 3-D shape use to effect of movie, animation, industrial design, medical treatment service, education, engineering etc... But it is not easy to make 3-D shape from the information of 2-D image. There are two methods in restoring 3-D video image through 2-D image; First the method of using a laser; Second, the method of acquiring 3-D image through stereo vision. Instead of doing two methods with many difficulties, I study the method of simple 3-D image in this research paper. We present here a simple and efficient method, called direct calibration, which does not require any equations at all. The direct calibration procedure builds a lookup table(LUT) linking image and 3-D coordinates by a real 3-D triangulation system. The LUT is built by measuring the image coordinates of a grid of known 3-D points, and recording both image and world coordinates for each point; the depth values of all other visible points are obtained by interpolation.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.3
• /
• pp.1-7
• /
• 2019

In this paper, we propose an approach to design a practical filter and a mathematical modeling for images. In the areas of signal processing, including high-dimensional image processing, the filtering process has been fundamental and crucial in diverse practical applications such as image processing, computer vision, and pattern recognition. In general, the ideal filter is modeled as circular-shaped in the 2D frequency domain as the rectangular shape is ideal for the 1D frequency domain. This paper proposes an approach to modeling practical and efficient image filter in the 2D frequency domain. Instead of employing a circular-shaped filter, this study proposes a polygon-shaped filter inspired by the concept of a hexagon cellular system for frequency reuse in wireless communication systems. By employing the concept of frequency reuse, bandwidth efficiency is also achieved in the frequency domain. To substantiate the proposed approach, quantitative evaluation is performed using PSNR.

• Journal of Multimedia Information System
• /
• v.6 no.4
• /
• pp.225-234
• /
• 2019

Deep learning has become increasingly popular in both academic and industrial areas nowadays. Various domains including pattern recognition, Computer vision have witnessed the great power of deep neural networks. However, current studies on deep learning mainly focus on quality data sets with balanced class labels, while training on bad and imbalanced data set have been providing great challenges for classification tasks. We propose in this paper a method of data analysis-based data reduction techniques for selecting good and diversity data samples from a large dataset for a deep learning model. Furthermore, data sampling techniques could be applied to decrease the large size of raw data by retrieving its useful knowledge as representatives. Therefore, instead of dealing with large size of raw data, we can use some data reduction techniques to sample data without losing important information. We group PCB characters in classes and train deep learning on the ResNet56 v2 and SENet model in order to improve the classification performance of optical character recognition (OCR) character classifier.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.1458-1463
• /
• 2004

Tracking is one of the most important pre-required task for many application such as human-computer interaction through gesture and face recognition, motion analysis, visual servoing, augment reality, industrial assembly and robot obstacle avoidance. Recently, 3D information of object is required in realtime for many aforementioned applications. 3D tracking is difficult problem to solve because during the image formation process of the camera, explicit 3D information about objects in the scene is lost. Recently, many vision system use stereo camera especially for 3D tracking. The 3D feature based tracking(3DFBT) which is on of the 3D tracking system using stereo vision have many advantage compare to other tracking methods. If we assumed the correspondence problem which is one of the subproblem of 3DFBT is solved, the accuracy of tracking depends on the accuracy of camera calibration. However, The existing calibration method based on accurate camera model so that modelling error and weakness to lens distortion are embedded. Therefore, this thesis proposes 3D feature based tracking method using SVM which is used to solve reconstruction problem.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.14 no.6
• /
• pp.3023-3031
• /
• 2013

In this paper, we propose a robot control data generation system applying context aware mechanism in order to control the robot manipulator which automatically sorts parcels. The context aware mechanism generates intelligent information to control a robot using context data such as the parcel shape, weight, location and barcodes. The proposed system collects context data of the parcel and generates robot control data to pick up and drop parcels. Then a robot manipulator, which receives control data of picking-up and dropping, processes the automated sorting of parcels according to delivery persons and delivery routes. It will contribute not only to save much time and cost but also to reduce the industrial accidents.