• Journal of the Korea Society of Computer and Information
• /
• v.25 no.2
• /
• pp.49-58
• /
• 2020

In this study, we solve an online multi-object problem which finds object states (i.e. locations and sizes) while conserving their identifications in online-provided images and detections. We handle this problem based on a tracking-by-detection approach by linking (or associating) detections between frames. For more accurate online association, we propose novel online appearance learning with discrete fourier transform and partial least square analysis (PLS). We first transform each object image into a Fourier image in order to extract meaningful features on a frequency domain. We then learn PLS subspaces which can discriminate frequency features of different objects. In addition, we incorporate the proposed appearance learning into the recent confidence-based association method, and extensively compare our methods with the state-of-the-art methods on MOT benchmark challenge datasets.

• Journal of Digital Convergence
• /
• v.15 no.9
• /
• pp.241-248
• /
• 2017

There are various problems in the process of developing the game. Especially, there is a lot of problems in testing the balance of the value of each object. This causes problems such as an increase in development cost and a delay in development time. Therefore, if there is a tool or simulator that can test the mutual value balance of each object in advance, this problem can be solved. However, currently there are few simulators, tools, and platforms that can analyze and evaluate the mutual value balance between these objects. In this paper, I designed a platform to evaluate and test the mutual value balance between these objects based on mock battle. The designed platform tests and evaluates the abilities of each object according to their attack and defense strengths. So, this reduces development costs and shortens development time.

• Journal of the Korea Society for Simulation
• /
• v.25 no.3
• /
• pp.85-95
• /
• 2016

In National Defense field, the importance of M&S and T&E has been increased due to complexity of modern Weapon System. And research reducing time and cost is being conducted continually on using limited resources efficiently. In the existing research, Weapon System's Performance Evaluation System using the Process-based method has been in progress. But, Objective basis or scientific method is insufficient in selecting appropriate function of a target to performance evaluation. Due to this, it's difficult to select functions suitable to the situation in same type. Also, Requirements of user and interrelation of evaluation factors can't be reflected systematically. In this research, it proposes the method to reflecting requirements of user, interrelation of elements in realistic situation for selecting evaluation object in Performance Evaluation Simulation. First, Evaluation Objects is selected using ANP which is multi-criterion decision making method. Second, decision support system is constructed using Programming Language(C#) based on the research result.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10b
• /
• pp.82-84
• /
• 2003

위치 기반 서비스 시스템은 이동객체에 대한 효율적인 검색을 가능하게 해주는 시스템으로 이동객체의 현재와 과거 위치에 대한 정보를 유지한다. 이러한 시스템을 평가하기 위해서는 이동객체들의 사실적인 움직임을 기술한 데이타 집합이 필요하다. 이러한 데이타 집합을 생성하는 대표적인 생성기로는 GSTD가 있다. 하지만 GSTD도 몇 가지 문제점을 지니고 있다. GSTD에 의해서 생성되는 작업공간과 데이타 집합은 실생활 객체의 움직임을 묘사하는데 부적절한 면이 있다. 또한 GSTD가 제공하는 매개변수 안에 사용자가 객체의 이동성과 시간성을 명확히 기술 해줄 수 있는 부분이 제시 되어 있지 않다. 본 논문에서는 GSTD의 여러 매개변수들과 함수들을 변환하거나 추가하여 좀 더 현실적인 객체 움직임을 나타내는 데이타 집합을 만들고, 사용자가 객체의 특성을 세밀히 지정해 줄 수 있는 확장된 GSTD를 구현한다. 또한 확장된 GSTD 기반 LBS 시뮬레이터의 설계와 구현에 관한 원리를 포함한 주요 특징을 기술한다. 확장된 GSTD는 다양한 LBS 응용 시스템의 성능 측정에 활용될 수 있을 것이다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.3
• /
• pp.742-748
• /
• 2014

The research of the object-oriented visual programming education is related to investigate the instructional method for computer programming, the process for program development for students, and the suitability analysis and design methodology for program development. In order to develop the educational methodology for the students who have the difficulties to study the computer programming and to enhance the abilities for the development of object-oriented programs, in this paper, we suggest the evaluation metrics to apply and assess for the object-oriented programming education. Also, we introduce the visual programming education methodology and curriculum using Alice programming. The evaluation metrics for the visual programming education suggested in this paper, is accepted when the visual programming languages/tools are used in the programming education to assess and analyze the suitabilities for the programming education for students, so to utilize the analyses for abilities of understanding, abstraction, and program implementation by level.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.2
• /
• pp.373-384
• /
• 2000

There are many researches about metrics to measure a quality of Object-Oriented(OO) software. However, most of them have only discussed a concept or properties of metrics, and have not shown the detailed procedure for measuring them. They also define a measurement indicator as a threshold, but it has been influenced on a project size or application domains. In this paper, we propose metrics based on characteristics of OO design such as size, complexity, coupling and cohesion, and use a propotion to an average as the measurement indicator. It is easy to classify classes which have a result above the average, and to predict classes which reduced the quality of OO design. They will be modified to hold the average. Proposed metrics are analytically evaluated by Weyuker's nine properties. They are satisfied with seven properties except two properties co not apply to OO metrics. Also, we design a quality assessment system, ASSOD(ASsessment System of Object oriented Design), to measure the quality of the OO design independent of the platform.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.3
• /
• pp.157-165
• /
• 2021

Recently, as the spatial resolution of satellite and aerial images has improved, various studies using remotely sensed data with high spatial resolution have been conducted. In particular, since the building extraction is essential for creating digital thematic maps, high accuracy of building extraction result is required. In this manuscript, building extraction models were generated using SegNet, U-Net, FC-DenseNet, and HRNetV2, which are representative semantic segmentation models in deep learning techniques, and then the evaluation of building extraction results was performed. Training dataset for building extraction were generated by using aerial orthophotos including various buildings, and evaluation was conducted in three areas. First, the model performance was evaluated through the region adjacent to the training dataset. In addition, the applicability of the model was evaluated through the region different from the training dataset. As a result, the f1-score of HRNetV2 represented the best values in terms of model performance and applicability. Through this study, the possibility of creating and modifying the building layer in the digital map was confirmed.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.2 s.34
• /
• pp.59-68
• /
• 2005

Object tracking in a real time image is one of interesting subjects in computer vision and many practical application fields Past couple of years. But sometimes existing systems cannot find object by recognize background noise as object. This paper proposes a method of object detection and tracking using adaptive background image in real time. To detect object which does not influenced by illumination and remove noise in background image, this system generates adaptive background image by real time background image updating. This system detects object using the difference between background image and input image from camera. After setting up MBR(minimum bounding rectangle) using the internal point of detected otject, the system tracks otiect through this MBR. In addition, this paper evaluates the test result about performance of proposed method as compared with existing tracking algorithm.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.410-412
• /
• 2021

This paper proposes an improved pedestrian detection method using object and background histograms. Objects detected through the HOG & SVM algorithm are detected in a square shape. Inside the square area, the background and the object area are mixed. If only the area of the object excluding the background is detected, various object-related information may be easily obtained. The size of the detected rectangle is readjusted using an xy-axis projection algorithm to fit the size of the object. And then, the improved object is detected by dividing the background and the object based on the histogram of the object in the readjusted square. The average values of precision and recall, which are reliability evaluations comparing the detected object with the original object, are 97.9% and 90%, respectively.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2016.11a
• /
• pp.140-141
• /
• 2016

기존의 사람 계수 측정 시스템은 적외선 빔이나 열 감지 영상 장치를 통해 측정하였다. 하지만 이와 같은 방법으로 측정하면 객체가 들어가거나 나가는 정보는 제공하지 않는다. 이에 본 논문은 고정된 카메라를 이용하여 각 사람의 피부색과 옷차림 등의 RGB 정보를 이용한 사람 계수 측정 기법을 제안한다. RGB카메라 영상을 통하여 객체의 RGB 히스토그램을 얻은 후 각 객체에 대해 Bhattacharyya metric을 통한 histogram similarity을 계산하여 객체 추적 및 분류를 통해 사람 계수 측정을 한다. 제안된 시스템은 C/C++을 기반으로 구현하여, 사람 계수 측정 성능을 평가하였다.