• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.2
• /
• pp.191-196
• /
• 2009

In this paper, we proposed a feature selection method using Binary Particle Swarm Optimization(BPSO) and Mutual information. This proposed method consists of the feature selection part for selecting candidate feature subset by mutual information and the optimal feature selection part for choosing optimal feature subset by BPSO in the candidate feature subsets. In the candidate feature selection part, we computed the mutual information of all features, respectively and selected a candidate feature subset by the ranking of mutual information. In the optimal feature selection part, optimal feature subset can be found by BPSO in the candidate feature subset. In the BPSO process, we used multi-object function to optimize both accuracy of classifier and selected feature subset size. DNA expression dataset are used for estimating the performance of the proposed method. Experimental results show that this method can achieve better performance for pattern recognition problems than conventional ones.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.9C
• /
• pp.922-929
• /
• 2005

In this paper, we propose the MRS p:Ittern classification techniques by the fusion scheme based on the SpPCA and MLP. A conventional PCA teclulique for the dimension reduction has the problem that it can't find a optimal transformation matrix if the property of input data is nonlinear. To overcome this drawback we extract features by the SpPCA technique which use the local patterns rather than whole patterns. In a next classification step, individual classifier based on MLP calculates the similarity of each class for local features. Finally, MRS patterns is classified by the fusion scheme to effectively combine the individual information. As the simulation results to verify the effectiveness, the proposed method showed more improved classification results than conventional methods.

• Journal of Cadastre & Land InformatiX
• /
• v.44 no.2
• /
• pp.103-112
• /
• 2014

The objective of this study is to suggest the possibility on land cover classification using hyperspectal imagery on area which includes lands and waters. After atmospheric correction as a preprocessing work was conducted on hyperspectral imagery acquired by airborne hyperspectral sensor CASI-1500, the effect of atmospheric correction to a few land cover class in before and after atmospheric correction was compared and analyzed. As the result of accuracy of land cover classification by highspectral imagery using reference data as airphoto and digital topographic map, maximum likelihood method represented overall accuracy as 67.0% and minimum distance method showed overall accuracy as 52.4%. Also product accuracy of land cover classification on road, dry field and green house, but that on river, forest, grassland showed low because the area of those was composed of complex object. Therefore, the study needs to select optimal band to classify specific object and to construct spectral library considering spectral characteristics of specific object.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.2
• /
• pp.117-126
• /
• 2019

Legacy study for detecting arrhythmia have mostly used nonlinear method to increase classification accuracy. Most methods are complex to process and manipulate data and have difficulties in classifying various arrhythmias. Therefore it is necessary to classify various arrhythmia based on short-term data. In this study, we propose a feature extraction based on auto regressive modeling and an premature contraction arrhythmia classification method using SVM., For this purpose, the R-wave is detected in the ECG signal from which noise has been removed, QRS and RR interval segment is modelled. Also, we classified Normal, PVC, PAC through SVM in realtime by extracting four optimal segment length and AR order. The detection and classification rate of R wave and PVC is evaluated through MIT-BIH arrhythmia database. The performance results indicate the average of 99.77% in R wave detection and 99.23%, 97.28%, 96.62% in Normal, PVC, PAC classification.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.1
• /
• pp.149-156
• /
• 2014

NGE(Nested Generalized Exemplar Method) can increase the performance of the noisy data at the same time, can reduce the size of the model. It is the optimal distance-based classification method using a matching rule. NGE cross or overlap hyperrectangles generated in the learning has been noted to inhibit the factors. In this paper, We propose the DHGen(Dominant Hyperrectangle Generation) algorithm which avoids the overlapping and the crossing between hyperrectangles, uses interval weights for mixed hyperrectangles to be splited based on the mutual information. The DHGen improves the classification performance and reduces the number of hyperrectangles by processing the training set in an incremental manner. The proposed DHGen has been successfully shown to exhibit comparable classification performance to k-NN and better result than EACH system which implements the NGE theory using benchmark data sets from UCI Machine Learning Repository.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.1
• /
• pp.8-14
• /
• 2017

Deinterlacing is a method to convert interlaced video, including two fields (even and odd), to progressive video. It can be divided into spatial and temporal methods. The deinterlacing method in the spatial domain can easily be hardware-implemented, but yields image degradation if information about the deinterlaced pixel does not exist in the same field. On the other hand, the method in the temporal domain yields a deinterlaced image with higher quality but uses more memory, and hardware implementation is more difficult. Furthermore, the deinterlacing method in the temporal domain degrades image quality when motion is not estimated properly. The proposed method is for deinterlacing in the spatial domain. It uses several deinterlacing methods according to statistical characteristics in neighboring pixel locations. In this procedure, the proposed method uses the C4.5 algorithm, a typical classification algorithm based on entropy for choosing optimal methods from among the candidates. The simulation results show that the proposed algorithm outperforms previous deinterlacing methods in terms of objective and subjective image quality.

• Journal of Korea Multimedia Society
• /
• v.8 no.2
• /
• pp.181-191
• /
• 2005

The classification error of 'ㅁ', 'ㅇ' is one of the main causes of incorrect recognition in Korean characters, but there haven't been enough researches to solve this problem. In this paper, a new feature extraction method from Korean grapheme is proposed to recognize 'ㅁ', 'ㅇ'effectively. First, we defined an optimal neighbor-distance selection measure using modified Hausdorff distance, which we determined the optimal neighbor-distance by. And we extracted neighbor-angle feature which was used as the effective feature to classify the two graphemes 'ㅁ', 'ㅇ'. Experimental results show that the proposed feature extraction method worked efficiently with the small number of features and could recognize the untrained patterns better than the conventional methods. It proves that the proposed method has a generality and stability for pattern recognition.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.10
• /
• pp.37-43
• /
• 2021

Using the acoustic features of speech, important social and linguistic information about the speaker can be obtained, and one of the key features is the dialect. A speaker's use of a dialect is a major barrier to interaction with a computer. Dialects can be distinguished at various levels such as phonemes, syllables, words, phrases, and sentences, but it is difficult to distinguish dialects by identifying them one by one. Therefore, in this paper, we propose a lightweight Korean dialect classification model using only MFCC among the features of speech data. We study the optimal method to utilize MFCC features through Korean conversational voice data, and compare the classification performance of five Korean dialects in Gyeonggi/Seoul, Gangwon, Chungcheong, Jeolla, and Gyeongsang in eight machine learning and deep learning classification models. The performance of most classification models was improved by normalizing the MFCC, and the accuracy was improved by 1.07% and F1-score by 2.04% compared to the best performance of the classification model before normalizing the MFCC.

• Journal of the Korea Society for Simulation
• /
• v.20 no.2
• /
• pp.77-85
• /
• 2011

It is believed that the anomalies or diseases of human organs are identified by the analysis of the patterns. This paper proposes a new classification technique for the identification of cancer disease using the proteome patterns obtained from two-dimensional polyacrylamide gel electrophoresis(2-D PAGE). In the new classification method, three different classification methods such as support vector machine(SVM), multi-layer perceptron(MLP) and k-nearest neighbor(k-NN) are extended by multi-boosting method in an array of subclassifiers and the results of each subclassifier are merged by ensemble method. Genetic algorithm was applied to obtain optimal feature set in each subclassifier. We applied our method to empirical data set from cancer research and the method showed the better accuracy and more stable performance than single classifier.

• The Journal of the Korea Contents Association
• /
• v.21 no.8
• /
• pp.1-9
• /
• 2021

Deep learning in computer vision has made accelerated improvement over a short period but large-scale learning data and computing power are still essential that required time-consuming trial and error tasks are involved to derive an optimal network model. In this study, we propose a similar image classification performance improvement method based on CR (Confusion Rate) that considers only the characteristics of the data itself regardless of network optimization or data reinforcement. The proposed method is a technique that improves the performance of the deep learning model by calculating the CRs for images in a dataset with similar characteristics and reflecting it in the weight of the Loss Function. Also, the CR-based recognition method is advantageous for image identification with high similarity because it enables image recognition in consideration of similarity between classes. As a result of applying the proposed method to the Resnet18 model, it showed a performance improvement of 0.22% in HanDB and 3.38% in Animal-10N. The proposed method is expected to be the basis for artificial intelligence research using noisy labeled data accompanying large-scale learning data.