• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.8
• /
• pp.2298-2315
• /
• 2024

This paper introduces a method aimed at diagnosing the presence or absence of lesions by detecting anomalies in Chest X-ray images. The proposed approach is based on the PatchCore anomaly detection method, which extracts a feature vector containing location information of an image patch from normal image data and calculates the anomaly distance from the normal vector. However, applying PatchCore directly to medical image processing presents challenges due to the possibility of diseases occurring only in specific organs and the presence of image noise unrelated to lesions. In this study, we present an image alignment method that utilizes affine transformation parameter prediction to standardize already captured X-ray images into a specific composition. Additionally, we introduce a region-specific abnormality detection method that requires affine-transformed chest X-ray images. Furthermore, we propose a method to enhance application efficiency and performance through feature map hard masking. The experimental results demonstrate that our proposed approach achieved a maximum AUROC (Area Under the Receiver Operating Characteristic) of 0.774. Compared to a previous study conducted on the same dataset, our method shows a 6.9% higher performance and improved accuracy.

• IEIE Transactions on Smart Processing and Computing
• /
• v.1 no.3
• /
• pp.133-142
• /
• 2012

This paper presents a salient region detection algorithm for auto-focusing based on the characteristics of a human's visual attention. To describe the saliency at the local, regional, and global levels, this paper proposes a set of novel features including multi-scale local contrast, variance, center-surround entropy, and closeness to the center. Those features are then prioritized to produce a saliency map. The major advantage of the proposed approach is twofold; i) robustness to changes in focus and ii) low computational complexity. The experimental results showed that the proposed method outperforms the existing low-level feature-based methods in the sense of both robustness and accuracy for auto-focusing.

• Journal of Institute of Control, Robotics and Systems
• /
• v.14 no.11
• /
• pp.1110-1116
• /
• 2008

In this paper, we suggest GP operators based on tree structure considering tree distributions in structure space and structural difficulties. The main idea of the proposed genetic operators is to place generated offspring into the specific region which nodes and depths are balanced and most of solutions exist. To enable that, the proposed operators are designed to utilize region information where parents belong and node/depth rates of selected subtree. To demonstrate the effectiveness of our proposed approach, experiments of binomial-3 regression, multiplexer and even parity problem are executed. The experiments results show that the proposed operators based on tree structure is superior to the results of standard GP for all three test problems in both success rate and number of evaluations.

• International Journal of Aeronautical and Space Sciences
• /
• v.10 no.1
• /
• pp.37-45
• /
• 2009

A method of conflict detection and resolution is described by using simple geometric approach. Two VAVs are dealt with and considered as point masses with constant velocity. This paper discusses en route aircraft which are assumed to be linked by real time data bases like ADS-B. With this data base, all DAVs share the information each other. Calculating PCA (Point of Closest Approach), we can evaluate the worst conflict condition between two VAVs. This paper proposes one resolution maneuvering logic, which can be called 'Vector Sharing Resolution'. In case of conflict, using miss distance vector in PCA, we can decide the directions for two VAVs to share the conflict region. With these directions, VAVs are going to maneuver cooperatively. First of all, this paper describes some '2-D' conflict scenarios and then extends to '3-D' conflict scenarios.

• Geomechanics and Engineering
• /
• v.20 no.2
• /
• pp.147-154
• /
• 2020

A novel approach for predicting lateral displacement caused by pile installation in anisotropic clay is presented, on the basis of the cylindrical and spherical cavities expansion theory. The K₀-based modified Cam-clay (K₀-MCC) model is adopted for the K₀-consolidated clay and the process of pile installation is taken as the cavity expansion problem in undrained condition. The radial displacement of plastic region is obtained by combining the cavity wall boundary and the elastic-plastic (EP) boundary conditions. The predicted equations of lateral displacement during single pile and multi-pile installation are proposed, and the hydraulic fracture problem in the vicinity of the pile tip is investigated. The comparison between the lateral displacement obtained from the presented approach and the measured data from Chai et al. (2005) is carried out and shows a good agreement. It is suggested that the presented approach is a useful tool for the design of soft subsoil improvement resulting from the pile installation.

• Journal of the Korean Society of Safety
• /
• v.15 no.3
• /
• pp.30-39
• /
• 2000

The present paper deals with the numerical simulation for the spray characteristics with swirling turbulent flows and dilution flows from swirl injectors in a simplified can type of gas turbine combustor. The main objective is to investigate the characteristics of swirling turbulent flows with dilution flows and to provide the qualitative results for the spray characteristics such as the droplet distribution and Sauter Mean Diameter(SMD). The gas-phase equations based on Eulerian approach were discretized by Finite Volume Method, together with SIMPLE algorithm and the Reynolds -Stress-Model. The liquid-phase equations based on Lagrangian method were used to predict the droplet behavior. The results of preliminary test are generally in good agreement with experimental data, and show that the anisotropy exists in the primary zone due to swirl velocity and injected air from primary injector, and then gradually decays due to turbulent mixing and consequently near-isotropy occurs in the region between primary and dilution zones. For the spray characteristics, it is indicated that the swirling flows of primary jet region increase the droplet atomization. In addition, it is showed that the swirling flows at the inlet region lead the air-fuel mixture to be distributed near the igniter and can significantly affect the spray behavior in the primary jet region.

• Geomechanics and Engineering
• /
• v.13 no.3
• /
• pp.505-515
• /
• 2017

Based on the collapse characteristics of a shallow rectangular cavity, a three-dimensional failure mechanism which can be used to study the collapsing region of the rock mass above a shallow cavity roof is constructed. Considering the effects of surcharge pressure and surface bolt on the collapsing block, the external rate of works produced by surcharge pressure and surface bolt are included in the energy dissipation calculation. Using variational approach, an analytic expression of surface equation for the collapsing block, which can be used to study the collapsing region of the rock mass above a shallow cavity roof, is derived in the framework of upper bound theorem. Based on the analytic expression of surface equation, the shape of the collapsing block for shallow cavity is drawn. Moreover, the changing law of the collapsing region for different parameters indicates that the collapsing region of rock mass decreases with the increase of the density of surface bolt. This conclusion can provide reference for practicing geotechnical engineers to achieve an optimal design of supporting structure for a shallow cavity.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.223-228
• /
• 2009

We propose an on-line machine learning approach for object recognition, where new images are continuously added and the recognition decision is made without delay. Random forest (RF) classifier has been extensively used as a generative model for classification and regression applications. We extend this technique for the task of building incremental component-based detector. First we employ object descriptor model based on bag of covariance matrices, to represent an object region then run our on-line RF learner to select object descriptors and to learn an object classifier. Experiments of the object recognition are provided to verify the effectiveness of the proposed approach. Results demonstrate that the propose model yields in object recognition performance comparable to the benchmark standard RF, AdaBoost, and SVM classifiers.

• Journal of Cadastre & Land InformatiX
• /
• v.48 no.2
• /
• pp.213-225
• /
• 2018

For constructing three-dimensional (3D) spatial information occlusion region problem arises in the process of taking the texture of the building. In order to solve this problem, it is necessary to investigate the automation method to automatically recognize the occlusion region, issue it, and automatically complement the texture. In fact there are occasions when it is possible to generate a very large number of structures and occlusion, so alternatives to overcome are being considered. In this study, we attempt to apply an approach to automatically create an occlusion region based on learning by patterning the blocked region using the recently emerging deep learning algorithm. Experiment to see the performance automatic detection of people, banners, vehicles, and traffic lights that cause occlusion in building walls using two advanced algorithms of Convolutional Neural Network (CNN) technique, Faster Region-based Convolutional Neural Network (R-CNN) and Mask R-CNN. And the results of the automatic detection by learning the banners in the pre-learned model of the Mask R-CNN method were found to be excellent.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.4
• /
• pp.124-131
• /
• 2009

A reliable detection of regions in radiography is one of the most important task before the evaluation of defects on welded joints. The extracted features is to be classified into distinctive clusters for each segmented region. But conventional segmentation techniques give unsatisfactory results for this task due to the spatial superposition of intensity and low signal-to-ratio(SNR) in radiographic images. The usage of global or local processes not only provide the necessary noise resistance but also fail in classification of regions. In this paper, we presents an appropriate approach for segmentation of region-based indications in industrial Computed Radiography(CR) images. The geometric differences between welded and non-welded area which is generated on radiography as the representative regions(background, thickness, middle and welded region in steel tube image) have constructed the hierarchical structure. Although this structure is contaminated by noise, the scheme between regions can be selected by the help of local clustering based on distinctive geometric property of each region. Because of the geometric nature of the considered region and so that the region is selected layer by layer, and that the real class represents the boundary between regions, the vertical and horizontal clustering process in each layer must be judicious. In order to show the effectiveness of this approach, a comparative experiment of various segmentation method is performed on industrial steel tube CR images.