• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2004년도 학술대회지
• /
• pp.768-771
• /
• 2004

In this paper, we propose a content-based audio genre classification algorithm that automatically classifies the query audio into five genres such as Classic, Hiphop, Jazz, Rock, Speech using digital signal processing approach. From the 20 second query audio file, 54 dimensional feature vectors, including Spectral Centroid, Rolloff, Flux, LPC, MFCC, is extracted from each query audio. For the classification algorithm, k-NN, Gaussian, GMM classifier is used. In order to choose optimum features from the 54 dimension feature vectors, SFS (Sequential Forward Selection) method is applied to draw 10 dimension optimum features and these are used for the genre classification algorithm. From the experimental result, we verify the superior performance of the SFS method that provides near $90{\%}$ success rate for the genre classification which means $10{\%}$-$20{\%}$ improvements over the previous methods

• Journal of the Korean Data and Information Science Society
• /
• 제14권2호
• /
• pp.167-175
• /
• 2003

In document classification, target of analysis is not document itself but words appeared in the document. Word information, therefore, is a significant factor in document classification. In this study, we are dealing with the classification of Korean document based on words and feature vectors. First, we present the performance of document classification using nouns and keywords. Second, we compare to the results for the size of feature vectors.

• 제어로봇시스템학회논문지
• /
• 제15권4호
• /
• pp.372-378
• /
• 2009

For effective mobility control of unmanned ground vehicles in outdoor off-road environments, terrain cover classification technology using passive sensors is vital. This paper presents a novel method far terrain classification based on color and texture information of off-road images. It uses a neural network classifier and wavelet features. We exploit the wavelet mean and energy features extracted from multi-channel wavelet transformed images and also utilize the terrain class spatial coordinates of images to include additional features. By comparing the classification performance according to applied features, the experimental results show that the proposed algorithm has a promising result and potential possibilities for autonomous navigation.

• 한국컴퓨터정보학회논문지
• /
• 제20권8호
• /
• pp.15-21
• /
• 2015

This study proposed a Sasang constitution classification system that can increase the objectivity and reliability of Sasang constitution diagnosis using the image of frontal face, in order to solve problems in the subjective classification of Sasang constitution based on Sasang constitution specialists' experiences. For classification, characteristics indicating the shapes of the eyes, nose, mouth and chin were defined, and such characteristics were extracted using the morphological statistic analysis of face images. Then, Sasang constitution was classified through a SVM (Support Vector Machine) classifier using the extracted characteristics as its input, and according to the results of experiment, the proposed system showed a correct recognition rate of 93.33%. Different from existing systems that designate characteristic points directly, this system showed a high correct recognition rate and therefore it is expected to be useful as a more objective Sasang constitution classification system.

• International Journal of Computer Science & Network Security
• /
• 제21권6호
• /
• pp.1-6
• /
• 2021

Recently, pre-trained convolutional neural network CNNs have been widely used and applied for medical image classification. These models can utilised in three different ways, for feature extraction, to use the architecture of the pre-trained model and to train some layers while freezing others. In this study, the ResNet18 pre-trained CNNs model is used for feature extraction, followed by the support vector machine for multiple classes to classify medical images from multi-classes, which is used as the main classifier. Our proposed classification method was implemented on Kvasir and PH2 medical image datasets. The overall accuracy was 93.38% and 91.67% for Kvasir and PH2 datasets, respectively. The classification results and performance of our proposed method outperformed some of the related similar methods in this area of study.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권11호
• /
• pp.3658-3679
• /
• 2022

Classification of persons wearing and not wearing face masks in images has emerged as a new computer vision problem during the COVID-19 pandemic. In order to address this problem and scale up the research in this domain, in this paper a hybrid technique by employing ResNet-101 and multi-layer perceptron (MLP) classifier has been proposed. The proposed technique is tested and validated on a self-created face masks classification dataset and a standard dataset. On self-created dataset, the proposed technique achieved a classification accuracy of 97.3%. To embrace the proposed technique, six other state-of-the-art CNN feature extractors with six other classical machine learning classifiers have been tested and compared with the proposed technique. The proposed technique achieved better classification accuracy and 1-6% higher precision, recall, and F1 score as compared to other tested deep feature extractors and machine learning classifiers.

• 대한원격탐사학회지
• /
• 제15권1호
• /
• pp.39-50
• /
• 1999

위성에서 관측된 다 대역 위성영상 데이터를 이용목적에 따라 분류하기 위해서는 복잡한 처리과정과 많은 시간을 필요로 하며, 감독분류시 훈련 데이터의 선택과 고려되는 다양한 특징 값들은 분류 정확도를 좌우할 만큼 민감한 특성을 나타내고 있다. 따라서 본 논문에서는 훈련데이터의 선택과 다양한 특징 값들 중 실제 영상분류에 기여도가 높은 특징을 추출하기 위하여 퍼지 기반의 $\gamma$모델을 이용한 분류네트웍을 구성하였다. 훈련집합 선택시 분류하고자 하는 지역의 밝기 분포도, 텍스쳐 특징 그리고 NDVI(Normalized Difference Vegetation Index)를 분류에 사용될 특징으로 선택하였고, 분류네트웍 출력 값의 오류가 최소화 되도록 Gradient Desoent 방법을 이용하여 각 노드의 $\gamma$파라미터를 훈련시키는 과정을 채택하였다. 이러한 훈련을 통하여 얻어진 파라미터를 이용하면 각 노드의 연결특성을 알 수 있으며, 다양한 입력 노드의 특징들 중 영상분류에 기여도가 적은 특징들을 추출하여 제거할 수 있다.

• International Journal of Computer Science & Network Security
• /
• 제24권2호
• /
• pp.79-88
• /
• 2024

The use of the social media has become part of our daily life activities. The social web channels provide the content generation facility to its users who can share their views, opinions and experiences towards certain topics. The researchers are using the social media content for various research areas. Sentiment analysis, one of the most active research areas in last decade, is the process to extract reviews, opinions and sentiments of people. Sentiment analysis is applied in diverse sub-areas such as subjectivity analysis, polarity detection, and emotion detection. Stance classification has emerged as a new and interesting research area as it aims to determine whether the content writer is in favor, against or neutral towards the target topic or issue. Stance classification is significant as it has many research applications like rumor stance classifications, stance classification towards public forums, claim stance classification, neural attention stance classification, online debate stance classification, dialogic properties stance classification etc. This research study explores different feature sets such as lexical, sentiment-specific, dialog-based which have been extracted using the standard datasets in the relevant area. Supervised learning approaches of generative algorithms such as Naïve Bayes and discriminative machine learning algorithms such as Support Vector Machine, Naïve Bayes, Decision Tree and k-Nearest Neighbor have been applied and then ensemble-based algorithms like Random Forest and AdaBoost have been applied. The empirical based results have been evaluated using the standard performance measures of Accuracy, Precision, Recall, and F-measures.

• 한국음향학회지
• /
• 제24권1호
• /
• pp.29-37
• /
• 2005

본 논문은 한국의 전통 음악, 즉 국악 장르를 자동으로 분류하는 시스템을 제안한다. 제안된 시스템은 입력 음악의 내용기반 분석을 통하여 궁중음악, 풍류방음악, 민속성악, 민속기악, 불교음악, 무속음악 등 6가지 장르중 하나로 자동분류하여 해당 음악의 장르 결과를 보여준다. 국악 장르 분류에 사용된 내용기반 알고리즘은 크게 음악의 특징 벡터 추출 그리고 장르 분류를 위한 패턴인식 과정 2가지로 구성된다. 음악의 특징 벡터 추출은 디지탈 신호 처리기술을 이용하여 해당 음악의 spectral centroid, rolloff, flux 등 STFT (Short Time Fourier Transform) 기반의 특징 계수들과 MFCC (Mel frequency cepstral coefficient), LPC (Linear predictive coding) 등의 계수들을 구한 후 SFS (Sequential Forward Selection) 최적 특징 벡터 열을 선별하여 사용하였으며 패틴 분류 알고리즘으로는 k-NN (k -Nearest Neighbor), Gaussian, GMM (Gaussian Mixture Model), SVM (Support Vector Machine) 분류기를 사용하였다. 특히 본 연구에서는 입력 질의의 패턴 (혹은 구간) 변화에 따른 시스템의 불확실성을 개선하기 위하여 MFC (Multi Feature Clustring) 방법을 이용하여 DB를 구축하였다. 모의실험 결과 k-NN 과 SVM 분류기 모두 $97{\%}$ 이상의 장르 분류 성공률을 보였으나, SVM 이 k-NN에 비해 약 3배 이상의 빠른 분류 성능을 가지고 있음을 확인하였다.

• 한국통신학회논문지
• /
• 제28권9C호
• /
• pp.876-883
• /
• 2003

본 논문에서는 초음파 영상에서 로그 전력 스펙트럼(log power spectrum)을 이용한 장기 인식 알고리듬을 제시한다. 제안한 알고리듬은 크게 특징추출과 특징분류의 두 단계로 구성된다. 특징추출에서는 이동불변의 성질을 가지는 로그 전력 스펙트럼을 이용하여 전처리를 수행한 입력 영상으로부터 장기 조직의 반향(echo of the tissue) 성분을 추출한다. 특징 분류에서는 마하라노비스(Mahalanobis) 거리를 사용하여 입력영상으로부터 추출한 특징벡터와 각 영상 부류의 평균벡터 사이의 유사도를 측정한다. 실제 초음파 영상에 대한 실험결과는 제안된 알고리듬이 전력 스펙트럼(power spectrum)과 유클리드(Euclid) 거리를 이용한 인식 알고리듬보다 최대 30% 향상된 인식률을, 또 가중 큐프런시(weighted quefrency) 복소 켑스트럼(complex cepstrum)을 이용한 알고리듬보다 10∼40% 향상된 인식률을 보여준다.