• Proceedings of the IEEK Conference
• /
• 2000.11d
• /
• pp.65-68
• /
• 2000

This paper describes an approach for extracting invariant features using a view-based representation and recognizing an object with a high speed search method in FLIR. In this paper, we use a reformulated eigenspace technique based on robust estimation for extracting features which are robust for outlier such as noise and clutter. After extracting feature, we recognize an object using a partial distance search method for calculating Euclidean distance. The experimental results show that the proposed method achieves the improvement of recognition rate compared with standard PCA.

• Journal of information and communication convergence engineering
• /
• v.7 no.2
• /
• pp.119-124
• /
• 2009

In this paper, we propose a novel approach to estimating the real-time moving trajectory of an object is proposed in this paper. The object's position is obtained from the image data of a CCD camera, while a state estimator predicts the linear and angular velocities of the moving object. To overcome the uncertainties and noises residing in the input data, a Extended Kalman Filter(EKF) and neural networks are utilized cooperatively. Since the EKF needs to approximate a nonlinear system into a linear model in order to estimate the states, there still exist errors as well as uncertainties. To resolve this problem, in this approach the Kohonen networks, which have a high adaptability to the memory of the input-output relationship, are utilized for the nonlinear region. In addition to this, the Kohonen network, as a sort of neural network, can effectively adapt to the dynamic variations and become robust against noises. This approach is derived from the observation that the Kohonen network is a type of self-organized map and is spatially oriented, which makes it suitable for determining the trajectories of moving objects. The superiority of the proposed algorithm compared with the EKF is demonstrated through real experiments.

• Smart Media Journal
• /
• v.12 no.9
• /
• pp.45-59
• /
• 2023

Weakly supervised object localization (WSOL) is a task of localizing an object in an image using only image-level labels. Previous studies have followed the conventional class activation mapping (CAM) pipeline. However, we reveal the current CAM approach suffers from problems which cause original CAM could not capture the complete defects features. This work utilizes a convolutional neural network (CNN) pretrained on image-level labels to generate class activation maps in a multi-scale manner to highlight discriminative regions. Additionally, a vision transformer (ViT) pretrained was treated to produce multi-head attention maps as an auxiliary detector. By integrating the CNN-based CAMs and attention maps, our approach localizes defective regions without requiring bounding box or pixel-level supervision during training. We evaluate our approach on a dataset of apple images with only image-level labels of defect categories. Experiments demonstrate our proposed method aligns with several Object Detection models performance, hold a promise for improving localization.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.57 no.3
• /
• pp.354-362
• /
• 2008

This paper presents the new method of power system transient stability simulation, which combines the desirable features of both the time domain technique based on OOP(Object-oriented Programming) and the direct method of transient stability analysis using detailed generator model. OOP is an alternative to overcome the problems associated with the development, maintenance and update of large software by electrical utilities. Several papers have already evaluated this approach for power system applications in areas such as load flow, security assessment and graphical interface. This paper applied the object-oriented approach to the problem of power system dynamics simulation. The modeling method is that each block of dynamic system block diagram is implemented as an object and connected each other. In the transient energy method, the detailed synchronous generator model is so-called two-axis model. For the excitation model, IEEE type1 model is used. The developed mothed was successfully applied to New England Test System.

• Journal of Information Processing Systems
• /
• v.9 no.4
• /
• pp.567-574
• /
• 2013

In this paper, we have proposed an architectural solution for a system for the visualization and modification of large amounts of data. The pattern is based on an asynchronous execution of programmable commands and a reflective approach of an object structure composition. The described pattern provides great flexibility, which helps adopting it easily to custom application needs. We have implemented a system based on the described pattern. The implemented system presents an innovative approach for a dynamic data object initialization and a flexible system for asynchronous interaction with data sources. We believe that this system can help software developers increase the quality and the production speed of their software products.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2010.04a
• /
• pp.405-408
• /
• 2010

This paper presents a new feature extraction technique, correlation coefficient and Manhattan distance (MD) based method for recognition of rotated object in an image. This paper also represented a new concept of intensity invariant. We extracted global features of an image and converts a large size image into a one-dimensional vector called circular feature vector's (CFVs). An especial advantage of the proposed technique is that the extracted features are same even if original image is rotated with rotation angles 1 to 360 or rotated. The proposed technique is based on fuzzy sets and finally we have recognized the object by using histogram matching, correlation coefficient and manhattan distance of the objects. The proposed approach is very easy in implementation and it has implemented in Matlab7 on Windows XP. The experimental results have demonstrated that the proposed approach performs successfully on a variety of small as well as large scale rotated images.

• Proceedings of the IEEK Conference
• /
• 2003.07e
• /
• pp.1771-1774
• /
• 2003

Image segmentation is very important technique as preprocessing. It is used for various applications such as object recognition, computer vision, object based image compression. In this paper, a method which segments the multidimensional image using a hierarchical histogram approach, is proposed. The hierarchical histogram approach is a method that decomposes the multi-dimensional situation into multi levels of 1 dimensional situations. It has the advantage of the rapid and easy calculation of the histogram, and at the same time because the histogram is applied at each level and not as a whole, it is possible to have more detailed partitioning of the situation.

• Proceedings of the IEEK Conference
• /
• 2006.06a
• /
• pp.299-300
• /
• 2006

In this paper, we propose a segmentation algorithm which combines the ideas from local watershed transforms and the region merging algorithm based hierarchical queue. Only the process of watershed and region merging algorithm can be restricted area. A fast region merging approach is proposed to extract the video object from the regions of watershed segmentation. Results show the effectiveness and convenience of the approach.

• Proceedings of the Korea Society for Simulation Conference
• /
• 2003.11a
• /
• pp.51-57
• /
• 2003

In operating distributed systems, proactive management is one of the major concerns for better quality of service and future capacity planning. In order to handle this management problem effectively, it is necessary to analyze performances of the distributed system and events generated by components in the system. This paper provides a rule-based event parsing engine for proactive management. Our event parsing engine uses object hooking-based and event-token approaches. The object hooking-based approach prepares new conditions and actions in Java classes and allows dynamically exchange them as hook objects in run time. The event-token approach allows the event parsing engine consider a proper sequence and relationship among events as an event token to trigger an action. We analyze the performance of our event parsing engine with two different implementations of rule structure; one is table-based and the other is tree-based.

• Electronics and Telecommunications Trends
• /
• v.33 no.4
• /
• pp.33-42
• /
• 2018

Object classification and detection are fundamental technologies in computer vision and its applications. Recently, a deep-learning based approach has shown significant improvement in terms of object classification and detection. This report reviews the progress of deep-learning based object classification and detection in views of the ImageNet Large Scale Visual Recognition Challenge (ILSVRC), and analyzes recent trends of object classification and detection technology and its applications.