• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.213-215
• /
• 2022

Recently, deep learning technology that processes images has been widely used in fire detection, autonomous driving, and defective product detection. In particular, in order to determine whether a product is contaminated or not, it can be identified through the contaminants passed from the existing sensor data, but technologies for recognizing cracks in products or contaminants themselves as images are being actively studied in various fields. In this paper, a system for classifying uncontaminated normal cups and contaminated cups through images was designed and implemented. The image was analyzed using an open image and a photographed image, and the image was analyzed by extracting the upper part of the cup image using Google Objectron for 3D object recognition. Through this study, it is thought that it will be used in various ways for research that can extract the contamination level of products required in the hygiene field based on images.

• Journal of Korean Society for Geospatial Information Science
• /
• v.17 no.3
• /
• pp.71-79
• /
• 2009

We expanded the 3D building information extraction method using shadow and vertical line from single high resolution image with meta information into the method for single high resolution image without meta information. Our method guesses an azimuth angle and an elevation angle of the sensor and the sun using reference building, selected by user, on an image. For test, we used an IKONOS image and an image extracted from the Google Earth. We calculated the Root Mean Square (RMS) error of heights extracted by our method using the building height extracted from stereo IKONOS image as reference, and the RMS error from the IKONOS image and the Google Earth image was under than 3 m. We also calculated the RMS error of horizontality position by comparison between building position extracted from only the IKONOS image and it from 1:1,000 digital map, and the result was under than 3 m. This test results showed that the height pattern of building models by our method was similar with it by the method using meta information.

• 한국HCI학회:학술대회논문집
• /
• 2008.02a
• /
• pp.210-214
• /
• 2008

It is actually impossible to model and store all objects which exist in real home environment into robot's database in advance. To resolve this problem, this paper proposes new object modeling method that can be available for robot self-modeling, which is capable of estimating whole model's shape from partial surface data using Generic Model. And this whole produce is conducted to cylindrical objects like cup, bottles and cans which can be easily found at indoor environment. The detailed process is firstly we obtain cylinder's initial principle axis using points coordinates and normal vectors from object's surface after we separate cylindrical object from 3D image. This 3D image is obtained from 3D sensor. And second, we compensate errors in the principle axis repeatedly. Then finally, we do modeling whole cylindrical object using cross sectional principal axis and its radius To show the feasibility of the algorithm, We implemented it and evaluated its accuracy.

• Journal of IKEEE
• /
• v.16 no.4
• /
• pp.369-374
• /
• 2012

This paper handles the UX implementation for system control using the visual input information of hand motion. Kinect sensor from Microsoft is used to acquire the user's skeleton image from the 3-D depth map at a rate of 30 frames per sec. and eventually knows the x-y coordinates of hand joints. The x-y coordinate value changes of hands between the present frame and next frame shows the direction of changes and rotation of changes and the various hand motion is used as a UX input command for remote image control on smart TV, etc. Through the experiments, we showed the implementation of the proposed idea.

• Proceedings of the KSRS Conference
• /
• 2005.10a
• /
• pp.79-82
• /
• 2005

Recently the necessity of road data is still being increased in industrial society, so there are many repairing and new constructions of roads at many areas. According to the development of government, city and region, the update and acquisition of road data for GIS (Geographical Information System) is very necessary. In this study, the fusion method with range data(3D Ground Coordinate System Data) and Intensity data in stand alone LiDAR data is used for road extraction and then digital image processing method is applicable. Up to date Intensity data of LiDAR is being studied. This study shows the possibility method for road extraction using Intensity data. Intensity and Range data are acquired at the same time. Therefore LiDAR does not have problems of multi-sensor data fusion method. Also the advantage of intensity data is already geocoded, same scale of real world and can make ortho-photo. Lastly, analysis of quantitative and quality is showed with extracted road image which compare with I: 1,000 digital map.

• Journal of Institute of Control, Robotics and Systems
• /
• v.16 no.10
• /
• pp.982-990
• /
• 2010

In a target tracking system using image information from a CCD (Charged Couple Device) or an IIR (Imaging Infra-red) sensor, occluded targets can result in track losses. If the target is occlued by background objects such as buildings or trees, probability of track existence will be reduced sharply and track will be terminated due to track maintenance algorithms. This paper proposes data association algorithm based on target existence for the robust tracking performance. we suggest the HPDA (Highest Probability Data Association) algorithm based on target existence and the tracking performance is compared with the established method based on target perceivability. Image tracking simulation that utilizes virtual 3D images and real IR images is employed to evaluate the robustness of the proposed tracking algorithm.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.387-389
• /
• 2003

Rational function model (RFM) is a universal sensor model for remote sensing image restitution. It is able to substitute for models of all known sensors. In this paper, RFM generation by CCD linear image models is described in detail. A principle of RFM-based 3D reconstruction and its implementation in JX4 DPW is also described. Experiments using IKONOS and SPOT5 images are carried out on JX4 DPW. Results show that RFM generated is feasible for photogrammetric restitution of CCD linear images.

• Journal of the Korean Society for Precision Engineering
• /
• v.12 no.4
• /
• pp.119-126
• /
• 1995

This paper presents a non-contact measuring system using one point measuring method to measure surface profiles of dies and clay models for practical use in the field of production engineering. The system has a laser beam probe similar to a measuring probe in a contact measuring system and CCD linear sensor used to detect 300mm measurement range, displacement of measured surfaces, from an origin. There is no mechanical interference between this measuring system and a measured surface in this system. In this measuring system, it was needed 500-600ms including data processing time to measure one point. The experiments showed that the standard deviation was 800 .mu.m and the reproducibility was also 100-210 .mu. m.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2016.10a
• /
• pp.733-734
• /
• 2016

This study proposes a novel approach for ground segmentation of 3D point cloud. We combine two techniques: gradient threshold segmentation, and mean height evaluation. Acquired 3D point cloud is represented as a graph data structures by exploiting the structure of 2D reference image. The ground parts nearing the position of the sensor are segmented based on gradient threshold technique. For sparse regions, we separate the ground and nonground by using a technique called mean height evaluation. The main contribution of this study is a new ground segmentation algorithm which works well with 3D point clouds from various environments. The processing time is acceptable and it allows the algorithm running in real time.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.29 no.8 s.145
• /
• pp.1168-1175
• /
• 2005

Clothing pressure has close relation with clothing comfort and depends on the pattern and properties of textile fabrics. Choosing a suitable clothing pressure is an essential factor for designing functional clothing such as the foundation for reshaping of a body contour or medical items for bum patient, and etc. However, it is hard to measure pressure values at the curved surface of a human body correctly. Recently, an air pack type pressure sensor, which has relatively excellent performance has been used to measure clothing pressure, however, it is still inconvenient to apply because it is a contact- type sensor. Therefore, in this paper, we suggest an indirect method that can measure clothing pressure without touching the subject by improving the equation of Kirk and Ibrahim (1966). However, confusions have been occurred when someone use the equation since the definition of parameters are somewhat vague. Furthermore, the estimated clothing pressure obtained by the previous method are quite different from the real values because this method does not consider the 3D effect of a human body and property changes of a transformed fabric. In this paper, the direction of principal stress and the radius of curvature in the principal direction were searched in the 3D image of the deformed girdle to get more accurate clothing pressure. The estimated clothing pressure was verified by comparing the result of the air pack type pressure sensor. It was found that the accuracy of the pressure estimation was improved by considering the 3D curvature of human body and the directional characteristics of textile fabrics.