• Annual Conference on Human and Language Technology
• /
• 1996.10a
• /
• pp.98-104
• /
• 1996

일반적으로 비선형 형태 정규화 과정은 필기체 문자에서 발생하는 형태 변형을 보상하기 위하여 사용되며, 현재까지 이진 영상에 대한 비선형 형태 정규화 방법들이 제안되었다. 그러나 현존하는 대부분의 문자 인식 시스템은 스캐너를 통하여 입력된 명도 문자영상을 이진화하여 사용하고 있기 때문에 이진화로 인해 야기되는 물자 영상에 대한 정보 유실 및 잡영 첨가 현상이 비선형 형태 정규화 과정에 누적되어 결과적으로 좋은 특징 추출 결과를 기대하기 어려운 실정이다. 본 연구에서는 이진화에 의한 정보의 손실을 최소화시키고, 필기체 문자에서 발생하는 다양한 형태 변형을 효과적으로 보상할 수 있는 명도 영상에서의 비선형 형태 정규화 방법을 제안한다. 제안된 명도 영상에서의 비선형 형태 정규화 방법들의 성능을 객관적으로 검증하기 위하여 처리 시간 및 복잡도 등을 기준으로 평가하였으며, 다양한 명도 한글 글씨 데이터에 대한 실험을 통하여 이진 영상에서의 비선형 형태 정규화 방법에 비해 제안된 방법이 변형이 심한 한글 글씨 데이타의 품질을 개선하는데 있어서 매우 효율적임을 확인할 수 있었다.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.7 no.5
• /
• pp.47-56
• /
• 2003

In this paper, the cyclic soil behavior model. which can accommodate the cyclic hardening, was developed by modifying the original parallel IWAN model. In order to consider the irrecoverable plastic strain of soil. the cyclic threshold strain, above which the backbone curve deviates from the original curve, was defined and the accumulated strain was determined by summation of the strains above the cyclic threshold in the stress-strain curve with applying Masing rule on unloading and reloading curves. The isotropic hardening elements are attached to the original parallel IWAN model and the slip stresses in the isotropic hardening elements are shown to increase according to the hardening functions. The hardening functions have a single parameter to account for the cyclic hardening and are defined by the symmetric limit cyclic loading test in forms of accumulated shear strain. The model development procedures are included in this paper and the verifications of developed model are discussed in the companion paper.

• Journal of Korea Multimedia Society
• /
• v.13 no.2
• /
• pp.216-224
• /
• 2010

A RAM-based Neural Network(RNN) has the advantages of processing speed and hardware implementation. In spite of these advantages, it has a saturation problem, weakness of repeated learning and extract of a generalized pattern. To resolve these problems of RNN, the 3DNS model using cumulative multi discriminator was proposed. But that model does not solve the saturation problem yet. In this paper, we proposed a adaptive weight cumulative neural net(AWCNN) using the adaptive weight neuron (AWN) for solving the saturation problem. The proposed nets improved a recognition rate and the saturation problem of 3DNS. We experimented with the MNIST database of NIST without preprocessing. As a result of experimentations, the AWCNN was 1.5% higher than 3DNS in a recognition rate when all input patterns were used. The recognition rate using generalized patterns was similar to that using all input patterns.

• 대한생식의학회:학술대회논문집
• /
• 1997.11a
• /
• pp.54-55
• /
• 1997

• Journal of KIISE:Software and Applications
• /
• v.34 no.7
• /
• pp.585-594
• /
• 2007

The back propagation algorithm took a long time to learn the input patterns and was difficult to train the additional or repeated learning patterns. So Aleksander proposed the binary neural network which could overcome the disadvantages of BP Network. But it had the limitation of repeated learning and was impossible to extract a generalized pattern. In this paper, we proposed a dynamic 3 dimensional Neuro System which was consisted of a learning network which was based on weightless neural network and a feedback module which could accumulate the characteristic. The proposed system was enable to train additional and repeated patterns. Also it could be produced a generalized pattern by putting a proper threshold into each learning-net's discriminator which was resulted from learning procedures. And then we reused the generalized pattern to elevate the recognition rate. In the last processing step to decide right category, we used maximum response detector. We experimented using the MNIST database of NIST and got 99.3% of right recognition rate for training data.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2008.10a
• /
• pp.415-418
• /
• 2008

This paper presents a method for moving objects detection, analysis and expression how much move as numerical value from the image which is captured by a network camera. To perform this method, we process few kinds of pre-processing to remove noise that are getting background image, difference image, binarization and so on. And to consider perspective effect, we propose modified ART2 algorithm. Finally, we express the result of ATR2 clustering as numerical value. This method is robust to size of object which is changed by perspective effect.

• Journal of the Institute of Convergence Signal Processing
• /
• v.5 no.2
• /
• pp.119-125
• /
• 2004

The direction component uses mainly the sobel and FFT method. The sobel method is difficult to set representative direction when we wish to extract representative direction when we wish to extract representative direction component and the complicated processing process of sobel mask because same value appears in the low provision image or high provision image and we cannot accumulate tilt size in case of making accumulate after making unit vector to pixel. The method that uses FFT conversion for direction extraction is possible in case that ridge has correct direction specification and must use a special direction filter. After thinning the binary image to supplement above weak point in the paper, we extract direction component by pixel unit, and we extract the most direction components of pixel that exist in block of 8${\times}$8 pixels size as representative direction of ridge.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.24 no.1
• /
• pp.65-71
• /
• 2006

There has been equipped wireless network infrastructure making possible to contact mobile computing at buildings, university, airport etc. Due to increase of mobile user dramatically, it raises interest about application and importance of LBS. The purpose of this study is to develop an indoor positioning system which is position of mobile users using Wireless LAN signal strength. We present Euclidean distance model and Bayesian inference model for analyzing position determination. The experimental results showed that the positioning of Bayesian inference model is more accurate than that of Euclidean distance model. In case of static target, the positioning accuracy of Bayesian inference model is within 2 m and increases when the number of cumulative tracking points increase. We suppose, however, Bayesian inference model using 5- cumulative tracking points is the most optimized thing, to decrease operation rate of mobile instruments and distance error of tracking points by movement of mobile user.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.9 no.5
• /
• pp.205-211
• /
• 2009

The Weightless Neural Network(WNN) has the advantage of the processing speed, less computability than weighted neural network which readjusts the weight. Especially, The behavior information such as sequential gesture has many serial correlation. So, It is required the high computability and processing time to recognize. To solve these problem, Many algorithms used that added preprocessing and hardware interface device to reduce the computability and speed. In this paper, we proposed the Ram based Sequential Cumulative Neural Network(SCNN) model which is sign language recognition system without preprocessing and hardware interface. We experimented with using compound words in continuous korean sign language which was input binary image with edge detection from camera. The recognition system of sign language without preprocessing got 93% recognition rate.

• The KIPS Transactions:PartA
• /
• v.15A no.3
• /
• pp.135-140
• /
• 2008

This paper describes a high-speed CMOS demultiplexer using redundant multi-valued logic (RMVL). The proposed circuit receives serial binary data and is converted to parallel redundant multi-valued data using RMVL. The converted data are reconverted to parallel binary data. By the redundant multi-valued data conversion, the RMVL makes it possible to achieve higher operating speeds than that of a conventional binary logic. The implemented demultiplexer consists of eight integrators. Each integrator is composed of an accumulator, a window comparator, a decoder and a D flip flop. The demultiplexer is designed with TSMC $0.18{\mu}m$ standard CMOS process. The validity and effectiveness are verified through the HSPICE simulation. The demultiplexer is achieved the maximum data rate of 20 Gb/s and the average power consumption of 95.85 mW.