• The Journal of the Acoustical Society of Korea
• /
• v.25 no.4
• /
• pp.151-158
• /
• 2006

In this paper, we proposed a new audio fingerprint extraction method, based on Philips' music retrieval algorithm, which uses the energy difference of neighboring filter-bank and probabilistic characteristics of music. Since Philips method uses too many filter-banks in limited frequency band, it may cause audio fingerprints to be highly sensitive to additive noises and to have too high correlation between neighboring bands. The proposed method improves robustness to noises by reducing the number of filter-banks while it maintains the discriminative power by representing the energy difference of bands with 2 bits where the quantization levels are determined by probabilistic characteristics. The correlation which exists among 4 different levels in 2 bits is not only utilized in similarity measurement. but also in efficient reduction of searching area. Experiments show that the proposed method is not only more robust to various environmental noises (street, department, car, office, and restaurant), but also takes less time for database search than Philips in the case where music is highly degraded.

• The Journal of the Korea Contents Association
• /
• v.14 no.11
• /
• pp.362-369
• /
• 2014

This study aimed to propose an accurate diagnostic method for osteoporosis by realizing a computer-aided diagnosis system with the application of the statistical analysis of texture features using digital images of lateral lumbar spine of patients with osteoporosis and providing reliable supplementary diagnostic information by model experimental research for early diagnosis of diseases. For these purposes, digital images of lateral lumbar spine of normal individuals and patients with osteoporosis were used in the experiments, and the values of statistical texture features on the set ROI were expressed in six parameters. Among the texture feature values of the six parameters of osteoporosis, the highest and lowest recognition rates of 95 and 80% were shown in average gray level and uniformity, respectively. Moreover, all the six parameters showed recognition rates of over 80% for osteoporosis: 82.5% in average contrast, 90% in smoothness, 87.5% in skewness, and 87.5% in entropy. Therefore, if a program developing into a computer-aided diagnosis system for medical images is coded based on the results of this study, it is considered possible to be applied to preliminary diagnostic data for automatic detection of lesions and disease diagnosis using medical images, to provide information for definite diagnosis of diseases, to diagnose by limited device, and to be used to shorten the time to analyze medical images.

• Journal of Korea Multimedia Society
• /
• v.14 no.1
• /
• pp.1-14
• /
• 2011

While the circulation of images become active, various requirements to manage increasing database are raised. The content-based technology is one of methods to satisfy these requirements. The image is represented by feature vectors extracted by various methods in the content-based technology. The global feature method insures fast matching speed because the feature vector extracted by the global feature method is formed into a standard shape. The global feature extraction methods are classified into two categories, the spatial feature extraction and statistical feature extraction. And each group is divided by what kind of information is used, color feature or gray scale feature. In this paper, we introduce various global feature extraction technologies and compare their performance by accuracy, recall-precision graph, ANMRR, feature vector size and matching time. According to the experiments, the spatial features show good performance in non-geometrical modifications, and the extraction technologies that use color and histogram feature show the best performance.

• Journal of Advanced Navigation Technology
• /
• v.16 no.2
• /
• pp.386-393
• /
• 2012

In this paper, we propose an efficient multi-object recognition and tracking scheme based on interest points of objects and their feature descriptors. To do that, we first define a set of object types of interest and collect their sample images. For sample images, we detect interest points and construct their feature descriptors using SURF. Next, we perform a statistical analysis of the local features to select representative points among them. Intuitively, the representative points of an object are the interest points that best characterize the object. in addition, we make the movement vectors of the interest points based on matching between their SURF descriptors and track the object using these vectors. Since our scheme treats all the objects independently, it can recognize and track multiple objects simultaneously. Through the experiments, we show that our proposed scheme can achieve reasonable performance.

• Journal of Digital Convergence
• /
• v.14 no.4
• /
• pp.277-283
• /
• 2016

This study proposed new methods to classify normal and epileptic seizure signals from EEG signals using peaks extracted by wavelet transform(WT) and Hilbert transform(HT) based on a neural network with weighted fuzzy membership functions(NEWFM). This study has the following three steps for extracting inputs for NEWFM. In the first step, the WT was used to remove noise from EEG signals. In the second step, the HT was used to extract peaks from the wavelet coefficients. We also selected the peaks bigger than the average of peaks to extract big peaks. In the third step, statistical methods were used to extract 16 features used as inputs for NEWFM from peaks. The proposed methodology shows that accuracy, specificity, and sensitivity are 99.25%, 99.4%, 99% with 16 features, respectively. Improvement in feature selection method in view to enhancing the accuracy is planned as the future work for selecting good features from 16 features.

• 한국HCI학회:학술대회논문집
• /
• 2006.02a
• /
• pp.1387-1392
• /
• 2006

본 논문에서는 SVM (Support Vector Machine)을 기반으로 하여 인체의 뇌 하부구조인 해마에 대한 지능적 형상분석 방법을 제공한다. 일반적으로 의료 영상으로부터 해마의 형상 분석을 하기 위해서는 충분한 임상 데이터를 필요로 한다. 하지만 현실적으로 많은 양의 표본들을 얻는 것이 쉽지 않기 때문에 전문가의 지식을 기반으로 한 작업이 수반되어야 한다. 결국 이러한 요소들이 분석 작업을 어렵게 한다. 의학 기술이 복잡해 지면서 최근의 형상 분석 연구는 점차 통계적 모델을 기반으로 진행되고 있다. 본 연구에서는 해마로부터 고해상도의 매개변수형 모델을 만들어 형상 표현으로 이용하고, 집단간 분류 작업에 SVM 알고리즘을 적용하는 지능적 분석 방법을 구현한다. 우선 메쉬 데이터로부터 물리변형모델 기반의 매개변수 모델을 구축하고, PDM (point distribution model) 방법을 적용하여 두 집단을 대표하는 평균 모델을 생성한다. 마지막으로 SVM 기반의 이진 분류기를 구축하여 집단간 분류 작업을 수행한다. 구현한 모델링 방법과 분류기의 성능을 평가하기 위하여 본 연구에서는 네 가지 커널 함수 (linear, radial basis function, polynomial, sigmoid)들을 적용한다. 본 논문에서 제시한 매개변수형 모델은 다양한 형태의 의료 데이터로부터 보편적인 3차원 모델을 생성하고, 또한 모델의 전역적, 국부적인 특징들을 복합적으로 표현할 수 있기 때문에 통계적 형상분석에 적합하다. 그리고 SVM 기반의 분류기는 적은 수의 학습 데이터로부터 정상인 해마 집단과 간질 환자 집단간의 정확한 분류를 가능하게 한다.

• Korean Journal of Cognitive Science
• /
• v.6 no.3
• /
• pp.65-81
• /
• 1995

Compound nouns as a composition of multiple nouns exhibit diverse occurence patterns in the texts and have varying degree of meaning coherence.The problem of compound nouns in information retrieval is to find a method to represent and identify the compositive patterns of each words.This paper explains how the cooccurrence patterns are related with the meaning of each compound noun and the information of such relations that can be mechanically acquired from texts is used in ranking the candidated documents for a given query.The main theme of the paper is that compound nouns can be categorized according to their occurrence patterns of simple nouns and these occurrence patterns can be formalized by statistical analysis without large dictionary or complex compositive rules.Our suggested model achieved about 7.75% improvement over the best precision of the other methods at each recall measurements on Korean test collection.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.1
• /
• pp.102-110
• /
• 2015

In this paper, adaptive matching method using the same features for rigid and deformable object images is proposed. Firstly, we determine whether the two images are matched or not using the geometric verification and generate the matching information. Decision boundary which separates deformable matching-pair from non-matching pair is obtained through statistical analysis of matching information. The experimental result shows that the proposed method lowers the computational complexity and increases the matching accuracy compared to the existing method.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.40 no.1
• /
• pp.28-33
• /
• 2003

This paper proposes a database index and retrieval method using the PCA(Principal Component Analysis). We perform a scene change detection and key frame extraction from the DC Image constructed by DCT DC coefficients in the compressed video stream that is video compression standard such as MPEG. In the extracted key frame, we use the PCA, then we can make codebook that has a statistical data as a codeword, which is saved as a database index. We also provide retrieval image that are similar to user's query image in a video database. As a result of experiments, we confirmed that the proposed method clearly showed superior performance in video retrieval and reduced computation time and memory space.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.3B
• /
• pp.212-218
• /
• 2012

An Internet worm is a self-replicating malware program which uses a computer network. As the network connectivity among computers increases, Internet worms have become widespread and are still big threats. There are many approaches to model the propagation of Internet worms such as Code Red, Nimda, and Slammer to get the insight of their behaviors and to devise possible defense methods to suppress worms' propagation activities. The influence of the network characteristics on the worm propagation has usually been modeled by medical epidemic model, named SI model, due to its simplicity and the similarity of propagation patterns. So far, SI model is still dominant and new variations of the SI model, called SI-style models, are being proposed for the modeling of new Internet worms. In this paper, we elaborate the problems of SI-style models and then propose a new accurate stochastic model using an occupancy problem.