• Journal of the Korea Convergence Society
• /
• v.11 no.10
• /
• pp.97-106
• /
• 2020

The factory information provided by the Industrial Location Information System (ILIS) is provided as raw data by the Korea Industrial Complex Corporation and registered after a filtering process, so the new factory information update is slow. In this study, to solve the problem of updating factory information of industrial location information system, using building data of road name address with relatively fast renewal cycle and building data of real estate, we compared the factory information of existing ILIS and extracted new factory information. In the process of comparison, a method was proposed to compare spatial objects of different types with point data and polygon data. Attribute information matching and object matching were performed, and attribute values of new factory information were extracted. The accuracy evaluation of the proposed spatial analysis method showed 79% accuracy, and the above matching technique was used to confirm the possibility of convergence of road name address data, real estate data and factory information of ILIS.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.8
• /
• pp.132-139
• /
• 2015

Visual SLAM(Simultaneous Localization And Mapping) has been used widely to estimate a mobile robot's location. Visual SLAM estimates relative motions with static visual features over image sequence. Because visual SLAM methods assume generally static features in the environment, we cannot obtain precise results in dynamic situation including many moving objects: cars and human beings. This paper presents a stereo vision based SLAM method in dynamic environment. First, we extract disparity map with stereo vision and compute optical flow. We then compute disparity change that is the estimated flow field between stereo views. After examining the disparity change value, we detect ROIs(Region Of Interest) in disparity space to determine dynamic scene objects. In indoor environment, many structural planes like walls may be determined as false dynamic elements. To solve this problem, we segment the scene into planar structure. More specifically, disparity values by the stereo vision are projected to X-Z plane and we employ Hough transform to determine planes. In final step, we remove ROIs nearby the walls and discriminate static scene elements in indoor environment. The experimental results show that the proposed method can obtain stable performance in dynamic environment.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.41 no.2
• /
• pp.65-76
• /
• 2004

Today, most documents are electronically produced and their paleography is digitalized by imaging, resulting in a tremendous number of electronic documents in the shape of images. Therefore, to process these document images, many methods of document structure analysis and recognition have already been proposed, including font classification. Accordingly, the current paper proposes a font classification method for document images that uses non-negative matrix factorization (NMF), which is able to learn part-based representations of objects. In the proposed method, spatially total features of font images are automatically extracted using NMF, then the appropriateness of the features specifying each font is investigated. The proposed method is expected to improve the performance of optical character recognition (OCR), document indexing, and retrieval systems, when such systems adopt a font classifier as a preprocessor.

• Journal of KIISE:Software and Applications
• /
• v.30 no.9
• /
• pp.885-893
• /
• 2003

In this paper, a new practical implementation of a person verification system using features of longitudinal section and transection and other facial, rotation compensated 3D face image, is proposed. The approach works by finding the nose tip that has a protrusion shape on the face. In feature recognition of 3D face image, one has to take into consideration the orientated frontal posture to normalize. Next, the special points in regions, such as nose, eyes and mouth are detected. The depth of nose, the area of nose and the volume of nose based both on the 3 longitudinal section and a transection are calculated. The eye interval and mouth width are also computed. Finally, the 12 features on the face were extracted. The Ll measure for comparing two feature vectors were used, because it is simple and robust. In the experimental results, proposed method achieves recognition rate of 95.5% for the longitudinal section and transection.

• Journal of Korea Multimedia Society
• /
• v.10 no.7
• /
• pp.923-933
• /
• 2007

In this paper, we present a method for performance enhancement of the marker detection system by using SVM(Support Vector Machine) and LDA(Linear Discriminant Analysis). It converts the input image to a binary image and extracts contours of objects in the binary image. After that, it approximates the contours to a list of line segments. It finds quadrangle by using geometrical features which are extracted from the approximated line segments. It normalizes the shape of extracted quadrangle into exact squares by using the warping technique and scale transformation. It extracts feature vectors from the square image by using principal component analysis. It then checks if the square image is a marker image or a non-marker image by using a SVM classifier. After that, it computes feature vectors by using LDA for the extracted marker images. And it calculates the distance between feature vector of input marker image and those of standard markers. Finally, it recognizes the marker by using minimum distance method. Experimental results show that the proposed method achieves enhancement of recognition rate with smaller feature vectors by using LDA and it can decrease false detection errors by using SVM.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.12
• /
• pp.546-554
• /
• 2019

Typically, closed-circuit television (CCTV) monitoring is mainly used for post-processes (i.e. to provide evidence after an incident has occurred), but by using a streaming video feed, machine-based learning, and advanced image recognition techniques, current technology can be extended to respond to crimes or reports of missing persons in real time. The multi-CCTV cooperation technique developed in this study is a program model that delivers similarity information about a suspect (or moving object) extracted via CCTV at one location and sent to a monitoring agent to track the selected suspect or object when he, she, or it moves out of range to another CCTV camera. To improve the operating efficiency of local government CCTV control centers, we describe here the partial automation of a CCTV control system that currently relies upon monitoring by human agents. We envisage an integrated crime prevention service, which incorporates the cooperative CCTV network suggested in this study and that can easily be experienced by citizens in ways such as determining a precise individual location in real time and providing a crime prevention service linked to smartphones and/or crime prevention/safety information.

• The Journal of the Korea Contents Association
• /
• v.5 no.4
• /
• pp.103-110
• /
• 2005

In this paper, we present a shadow detection algorithm using the shadow features which appear in color images. Shadow regions have lower luminance and saturation than those of nearby regions, and is generally shown as dark colors. The regions are detected by means of analysing and applying their properties to images represented as the HSI color model. The proposed algorithm is consisted of two steps: at the first step, the candidate regions of shadow are found with using shadow features, and then, real shadow regions are detected only in candidate regions by using their information to reduce real objects and dark marks. The experimental results show that the algorithm is effective.

• The KIPS Transactions:PartB
• /
• v.15B no.5
• /
• pp.407-412
• /
• 2008

Color segmentation is one of the most challenging problems in image processing especially in case of handling the images with cluttered background. Great amount of color segmentation methods have been developed and applied to real problems. In this paper, we suggest a new methodology. Our approach is focused on background extraction, as a complimentary operation to standard foreground object segmentation, using self-organizing feature selective property of unsupervised self-learning paradigm based on the competitive algorithm. The results of our studies show that background segmentation can be achievable in efficient manner.

• Journal of Broadcast Engineering
• /
• v.21 no.6
• /
• pp.861-877
• /
• 2016

This paper first discusses the disadvantages of the existing CamShift Algorithm for real time face tracking, and then proposes a new Camshift Algorithm that performs better than the existing algorithm. The existing CamShift Algorithm shows unstable tracking when tracing similar colors in the background of objects. This drawback of the existing CamShift is resolved by using Kinect’s pixel-by-pixel depth information and the Skin Detection algorithm to extract candidate skin regions based on HSV color space. Additionally, even when the tracking object is not found, or when occlusion occurs, the feature point-based matching algorithm makes it robust to occlusion. By applying the improved CamShift algorithm to face tracking, the proposed real-time face tracking algorithm can be applied to various fields. The results from the experiment prove that the proposed algorithm is superior in tracking performance to that of existing TLD tracking algorithm, and offers faster processing speed. Also, while the proposed algorithm has a slower processing speed than CamShift, it overcomes all the existing shortfalls of the existing CamShift.

• Journal of Convergence for Information Technology
• /
• v.11 no.9
• /
• pp.28-34
• /
• 2021

In the case of object detection using deep learning, both accuracy and real-time are required. However, it is difficult to use a deep learning model that processes a large amount of data in a limited resource environment. To solve this problem, this paper proposes an object detection model for small embedded devices. Unlike the general detection model, the model size was minimized by using a structure in which the pre-trained feature extractor was removed. The structure of the model was designed by repeatedly stacking lightweight convolution blocks. In addition, the number of region proposals is greatly reduced to reduce detection overhead. The proposed model was trained and evaluated using the public dataset PASCAL VOC. For quantitative evaluation of the model, detection performance was measured with average precision used in the detection field. And the detection speed was measured in a Raspberry Pi similar to an actual embedded device. Through the experiment, we achieved improved accuracy and faster reasoning speed compared to the existing detection method.