• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.2
• /
• pp.742-762
• /
• 2015

In this study, we propose a distance metric learning approach called discriminant metric learning (DML) for face verification, which addresses a binary-class problem for classifying whether or not two input images are of the same subject. The critical issue for solving this problem is determining the method to be used for measuring the distance between two images. Among various methods, the large margin nearest neighbor (LMNN) method is a state-of-the-art algorithm. However, to compensate the LMNN's entangled data distribution due to high levels of appearance variations in unconstrained environments, DML's goal is to penalize violations of the negative pair distance relationship, i.e., the images with different labels, while being integrated with LMNN to model the distance relation between positive pairs, i.e., the images with the same label. The likelihoods of the input images, estimated using DML and LMNN metrics, are then weighted and combined for further analysis. Additionally, rather than using the k-nearest neighbor (k-NN) classification mechanism, we propose a verification mechanism that measures the correlation of the class label distribution of neighbors to reduce the false negative rate of positive pairs. From the experimental results, we see that DML can modify the relation of negative pairs in the original LMNN space and compensate for LMNN's performance on faces with large variances, such as pose and expression.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.17 no.6
• /
• pp.11-17
• /
• 2017

We proposed malware detection method, which use the feature vector that consist of Opcode(operation code) and Windows API Calls extracted from executable files. And, we implemented our feature vector and measured the performance of it by using Bernoulli Naïve Bayes and K-Nearest Neighbor classifier. In experimental result, when using the K-NN classifier with the proposed method, we obtain 95.21% malware detection accuracy. It was better than existing methods using only either Opcode or Windows API Calls.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.10
• /
• pp.73-81
• /
• 2015

The paper proposes a prototype selection method and evaluates the generalization performance of standard algorithms and prototype based classification learning. The proposed prototype classifier defines multidimensional spheres with variable radii within class areas and generates a small set of training data. The nearest-neighbor classifier uses the new training set for predicting the class of test data. By decomposing bias and variance of the mean expected error value, we compare the generalization errors of k-nearest neighbor, Bayesian classifier, prototype selection using fixed radius and the proposed prototype selection method. In experiments, the bias-variance changing trends of the proposed prototype classifier are similar to those of nearest neighbor classifiers with all training data and the prototype selection rates are under 27.0% on average.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.467-470
• /
• 2003

When user wants to find objects which have the nearest position from him, we use the nearest neighbor (NN) query. The GIS applications, such as navigation system and traffic control system, require processing of NN query for moving objects (MOs). MOs have trajectory with changing their position over time. Therefore, we should be able to find NN object continuously changing over the whole query time when process NN query for MOs, as well as moving nearby on trajectory of query. However, none of previous works consider trajectory information between objects. Therefore, we propose a method of continuous NN query for trajectory of MOs. We call this CTNN (continuous trajectory NN) technique. It ran find constantly valid NN object on the whole query time by considering of trajectory information.

• Journal of IKEEE
• /
• v.26 no.2
• /
• pp.169-175
• /
• 2022

This paper introduces an interference elimination method using k-nearest neighbor (KNN) algorithm for precise distance estimation by reducing interference between ultrasonic sensors. Conventional methods compare current distance measurement result with previous distance measurement results. If the difference exceeds some thresholds, conventional methods recognize them as interference and exclude them, but they often suffer from imprecise distance prediction. KNN algorithm classifies input values measured by multiple ultrasonic sensors and predicts high accuracy outputs. Experiments of distance measurements are conducted where interference frequently occurs by multiple ultrasound sensors of same type, and the results show that KNN algorithm significantly reduce distance prediction errors. Also the results show that the prediction performance of KNN algorithm is superior to conventional voting methods.

• Korean Journal of Remote Sensing
• /
• v.30 no.5
• /
• pp.651-664
• /
• 2014

Recently, the demands of accurate forest carbon stock estimation and mapping are increasing in Korea. This study investigates the feasibility of two methods, k-Nearest Neighbor (kNN) and Regression Tree Analysis (RTA), for carbon stock estimation of pilot areas, Gongju and Sejong cities. The 3rd and 5th ~ 6th NFI data were collected together with Landsat TM acquired in 1992, 2010 and Aster in 2009. Additionally, various vegetation indices and tasseled cap transformation were created for better estimation. Comparison between two methods was conducted by evaluating carbon statistics and visualizing carbon distributions on the map. The comparisons indicated clear strengths and weaknesses of two methods: kNN method has produced more consistent estimates regardless of types of satellite images, but its carbon maps were somewhat smooth to represent the dense carbon areas, particularly for Aster 2009 case. Meanwhile, RTA method has produced better performance on mean bias results and representation of dense carbon areas, but they were more subject to types of satellite images, representing high variability in spatial patterns of carbon maps. Finally, in order to identify the increases in carbon stock of study area, we created the difference maps by subtracting the 1992 carbon map from the 2009 and 2010 carbon maps. Consequently, it was found that the total carbon stock in Gongju and Sejong cities was drastically increased during that period.

• Journal of the Korean Society for Precision Engineering
• /
• v.19 no.3
• /
• pp.52-58
• /
• 2002

The purpose of condition monitoring and fault diagnosis is to detect faults occurring in machinery in order to improve the level of safety in plants and reduce operational and maintenance costs. The recognition performance is important not only to gain a high recognition rate bur a1so to minimize the diagnosis failures error rate by using off effective rejection module. We examined the problem of performance evaluation for the rejection scheme considering the accuracy of individual c1asses in order to increase the recognition performance. We use the Smith's method among the previous studies related to rejection method. Nearest neighbor classifier is used for classifying the machine conditions from the vibration signals. The experiment results for the performance evaluation of rejection show the modified optimum rejection method is superior to others.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2015.07a
• /
• pp.122-124
• /
• 2015

This paper provides the design and implementation of Free Viewpoint TV System with multiple Microsoft Kinects. It generates a virtual view between two views by manipulating texture and depth image captured by Kinects in real-time. In order to avoid this, we propose the hole-filling scheme using Nearest neighbor and inpainting. As a result, holes generated by interference are filled with new depth values calculated by their neighbors. However, the depth values are not accurate, but are similar with their neighbors. And depending on the frequency of running a Nearest Neighbor method, we can see that edge's border would be shifted inner or outer of the object.

• ETRI Journal
• /
• v.35 no.3
• /
• pp.406-413
• /
• 2013

In this paper, an orthogonal frequency division multiple access (OFDMA)-based minimum end-to-end delay (MED) distributed routing scheme for mobile backhaul wireless mesh networks is proposed. The proposed scheme selects routing paths based on OFDMA subcarrier synchronization control, subcarrier availability, and delay. In the proposed scheme, OFDMA is used to transmit frames between mesh routers using type-I hybrid automatic repeat request over multipath Rayleigh fading channels. Compared with other distributed routing algorithms, such as most forward within radius R, farthest neighbor routing, nearest neighbor routing, and nearest with forwarding progress, simulation results show that the proposed MED routing can reduce end-to-end delay and support highly reliable routing using only local information of neighbor nodes.

• Journal of Intelligence and Information Systems
• /
• v.20 no.1
• /
• pp.121-131
• /
• 2014

Traffic speed is an important measure in transportation. It can be employed for various purposes, including traffic congestion detection, travel time estimation, and road design. Consequently, accurate speed prediction is essential in the development of intelligent transportation systems. In this paper, we present an analysis and speed prediction of a certain road section in Busan, South Korea. In previous works, only historical data of the target link are used for prediction. Here, we extract features from real traffic data by considering the neighboring links. After obtaining the candidate features, linear regression, model tree, and k-nearest neighbor (k-NN) are employed for both feature selection and speed prediction. The experiment results show that k-NN outperforms model tree and linear regression for the given dataset. Compared to the other predictors, k-NN significantly reduces the error measures that we use, including mean absolute percentage error (MAPE) and root mean square error (RMSE).