• The Transactions of the Korea Information Processing Society
• /
• v.4 no.1
• /
• pp.149-160
• /
• 1997

This paper is concerned with the problem of speaker-adaptive speech synthes is method using a mapped codebook designed by fuzzy mapping on FLVQ (Fuzzy Learning Vector Quantization). The FLVQ is used to design both input and reference speaker's codebook. This algorithm is incorporated fuzzy membership function into the LVQ(learning vector quantization) networks. Unlike the LVQ algorithm, this algorithm minimizes the network output errors which are the differences of clas s membership target and actual membership values, and results to minimize the distances between training patterns and competing neurons. Speaker Adaptation in speech synthesis is performed as follow;input speaker's codebook is mapped a reference speaker's codebook in fuzzy concepts. The Fuzzy VQ mapping replaces a codevector preserving its fuzzy membership function. The codevector correspondence histogram is obtained by accumulating the vector correspondence along the DTW optimal path. We use the Fuzzy VQ mapping to design a mapped codebook. The mapped codebook is defined as a linear combination of reference speaker's vectors using each fuzzy histogram as a weighting function with membership values. In adaptive-speech synthesis stage, input speech is fuzzy vector-quantized by the mapped codcbook, and then FCM arithmetic is used to synthesize speech adapted to input speaker. The speaker adaption experiments are carried out using speech of males in their thirties as input speaker's speech, and a female in her twenties as reference speaker's speech. Speeches used in experiments are sentences /anyoung hasim nika/ and /good morning/. As a results of experiments, we obtained a synthesized speech adapted to input speaker.

• The Transactions of the Korea Information Processing Society
• /
• v.1 no.2
• /
• pp.215-224
• /
• 1994

In this paper, the rotation and size invariant fingerprint recognition using the neural network EART (Extended Adaptive Resonance Theory) is studied ($515{\times}512$) gray level fingerprint images are converted into the binary thinned images based on the adaptive threshold and a thinning algorithm. From these binary thinned images, we extract the ending points and the bifurcation points, which are the most useful critical feature points in the fingerprint images, using the $3{\times}3$ MASK. And we convert the number of these critical points and the interior angles of convex polygon composed of the bifurcation points into the 40*10 critical using the weighted code which is invariant of rotation and size as the input of EART. This system produces very good and efficient results for the rotation and size variations without the restoration of the binary thinned fingerprints.

• Korean Journal of Social Welfare
• /
• v.57 no.2
• /
• pp.231-252
• /
• 2005

The changing Agenda of the family policy is to be influenced by many factors. The structural change of population and aging problem in modern society are generally included in main factors which produce political needs. Therefore, migrant family is recognized as a alternative which is to resolve the socioeconomic problems of aging industrial societies. In Germany, Gerhard Schroeder who leads the SPD governs since 1998. The Schroeder's new conception of welfare policy is known as the third way("neue Mitte") which the social economic structure intends to neo-liberalistic trend. The concern on the welfare policy for the migrant families has grown recently in neo-liberalistic age, but research on german welfare policy for migrant families is very limited in quantity and Depth. The main purpose of this study is to examine how the welfare policy for the migrant families of Schroeder-government has been developed in Germany. And this paper evaluates the results of transforming policy. Based on this evaluation, this paper attempts to be manifest the problem of korean family policy for the migrant population and to find out implications of german model.

• Journal of the Economic Geographical Society of Korea
• /
• v.20 no.4
• /
• pp.486-502
• /
• 2017

This study investigated the formation process and networks of Korean Clean Development Mechanism(CDM) industry. It aimed to reveal the factors and the drivers for the formation processes of this industry in the specific place. Based on the analysis of the Project Design Documents(PDDs) of the CDM projects and the collected project data by international institutions, surveys, and interviews were done. On the basis of these data, the analysis on the industrial change as complex emergent effects by the network evolution caused by adaptive activity of firms is conducted. In the time of the genesis, a kind of serendipity that the industrial activities of Korean firms meet to new system, CDM, In the changing process of the Korean CDM industry, the role of policies fo Korean Government was important to promote the new and renewable energy projects of the power companies. In the time of restructuring, Korean government policies formed new initial conditions for the new domestic GHGs reduction industry. In this processes, the localization of knowledge acted as a key driver for the formation of the Korean CDM industry.

• Journal of Institute of Control, Robotics and Systems
• /
• v.22 no.9
• /
• pp.738-744
• /
• 2016

In this paper, a robust visual odometry system has been proposed and implemented in an environment with dynamic illumination. Visual odometry is based on stereo images to estimate the distance to an object. It is very difficult to realize a highly accurate and stable estimation because image quality is highly dependent on the illumination, which is a major disadvantage of visual odometry. Therefore, in order to solve the problem of low performance during the feature detection phase that is caused by illumination variations, it is suggested to determine an optimal threshold value in the image binarization and to use an adaptive threshold value for feature detection. A feature point direction and a magnitude of the motion vector that is not uniform are utilized as the features. The performance of feature detection has been improved by the RANSAC algorithm. As a result, the position of a mobile robot has been estimated using the feature points. The experimental results demonstrated that the proposed approach has superior performance against illumination variations.

• The KIPS Transactions:PartB
• /
• v.9B no.3
• /
• pp.337-342
• /
• 2002

This paper proposes a new adaptive image labeling algorithm fur object analysis of the binary images. The proposed labeling algorithm need not merge/order of complex equivalent labels like classical labeling algorithm and the processing is done during only 1 Pass. In addition, this algorithm can be extended for gray-level image easily. Experiment result with HIPR image library shows that the proposed algorithm process more than 2 times laster than compared algorithm.

• Proceedings of the Korean Information Science Society Conference
• /
• 2008.06c
• /
• pp.527-532
• /
• 2008

자막은 영상과 관련이 있는 정보를 포함한다. 이러한 영상의 정보를 이용하기 위해서 자막을 추출하는 연구가 진행되고 있다. 기존의 자막 추출 연구는 언어 독립적인 특징으로 자막을 이루는 획의 에지는 일정한 간격을 유지하거나 수평라인으로 존재하는 글자의 분포를 이용한 방법을 제안하였다. 이러한 방법들은 획의 간격이 일정한 자막이나 하나의 글자가 하나의 획으로 이루어진 글자에서만 정상적인 동작을 보장하였다. 본 논문에서는 한글 자막 특징을 고려한 자막 추출 방법을 제안한다. 먼저, 한글 자막의 특징인 가로 획의 다수 분포를 고려한 적응형 에지 이진화를 수행하여 에지 영상을 생성하고 에지 연결 객체를 생성한다. 그 후에 생성한 연결 객체를 특징을 추출하여 사전에 생성한 의사결정 트리로 연결 객체를 자막과 비자막 연결객체로 분류한다. 의사결정 트리를 생성하기 위해서 사용한 연결 객체는 뉴스, 다큐멘터리 프로그램에서 획득하였으며, 성능 평가를 위해서 뉴스, 다큐멘터리, 스포츠 프로그램과 같은 대중 방송에서 획득한 영상에서 자막을 추출하였다. 평가 방법은 찾아진 연결 객체 중에 자막 연결 객체의 비율과 전체 자막 중에서 찾아진 자막 연결 객체의 비율로 분석하였다. 실험 결과에서는 제안한 방법이 한글 자막의 추출에 적용 가능함을 보여준다.

• Journal of the Institute of Convergence Signal Processing
• /
• v.1 no.1
• /
• pp.89-96
• /
• 2000

In this paper, we propose a vision algorithm for the inspection of very small RF-chip inductor which is used in mobile-communication terminal. The proposed method divides coil part from the inductor body by local adaptive thresholding and integral projection method. After dividing work, the coil components are extracted by thinning and labelling techniques. The test items are the number of turns, the intervals in coil, and the measure of uniformity between the extracted lines. If the values of these are more than the specific value a tested product is decided bad one. In the simulation, the proposed method has a good performance.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.48 no.1
• /
• pp.127-132
• /
• 2011

In this paper, we propose a novel method of adult image classification by combining skin color regions and edges in an input image. The proposed method consists of four steps. In the first step, initial skin color regions are detected by logical AND operation of all skin color regions detected by the existing methods of skin color detection. In the second step, a skin color probability map is created by modeling the distribution of skin color in the initial regions. Then, a binary image is generated by using threshold value from the skin color probability map. In the third step, after using the binary image and edge information, we detect final skin color regions using a region growing method. In the final step, adult image classification is performed by support vector machine(SVM). To this end, a feature vector is extracted by combining the final skin color regions and neighboring edges of them. As experimental results, the proposed method improves performance of the adult image classification by 9.6%, compared to the existing method.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.7
• /
• pp.3-8
• /
• 2016

We propose a modified form of the decision-directed least mean square (DD LMS) algorithm that is widely used in the optimization of self-adaptive equalizers, and show the modified version greatly improves the initial convergence properties of the conventional algorithm. Existing DD LMS regards the difference between a equalizer output and a quantization value for it as an error, and achieves an optimization of the equalizer based on minimizing the mean squared error cost function for the equalizer coefficients. This error generating method is useful for binary signal or a single-level signals, however, in the case of multi-level signals, it is not effective in the initialization of the equalizer. The modified DD LMS solves this problem by modifying the error generation. We verified the usefulness and performance of the modified DD LMS through experiments with multi-level signals under distortions due to intersymbol interference and additive noise.