• Journal of the Korean Institute of Telematics and Electronics
• /
• v.26 no.10
• /
• pp.1554-1560
• /
• 1989

This paper suggests the algorihtm which extracts the classification parameter relative to the only feature of document patterns even though they are rotated or scaled, and also classifies them. With the complex logarithmic conformal mapping, the sample of the document pattern image makes the pattern image of the complex logarithmic plane. Because the power spectrum of this plane is invariant to the rotation, and scale of the pattern image, it is used as the characteristics parameter of the patten image. By using the coherence function, this method analyzes the standard and input power spectrum. additionally, it classifies the input pattern image. Even though input image is rotated, our algorithm can classify it without reference to the rotation, and this is possible when the scale is in the range of 0.5-1.5.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.5
• /
• pp.141-148
• /
• 1994

Both the neural network classifier utilizing multi-layer perceptron and the linear tree classifier composed of hierarchically structured linear discriminating functions can form arbitrarily complex decision boundaries in the feature space and have very similar decision making processes. In this paper, a new method for automatically choosing the number of neurons in the hidden layers and for initalzing the connection weights between the layres and its supporting theory are presented by mapping the sequential structure of the linear tree classifier to the parallel structure of the neural networks having one or two hidden layers. Experimental results on the real data obtained from the military ship images show that this method is effective, and that three exists no siginificant difference in the classification acuracy of both classifiers.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.3
• /
• pp.1650-1669
• /
• 2017

In this study, each pixel in an ear is used as a centroid to generate a cake. Subsequently the major axis length of this cake is computed and obtained. This obtained major axis length serves as a feature to recognize an ear. Later, the ear hole is used as a centroid and a 16-circle template is generated to extract the major axis lengths of the ear. The 16-circle template extracted signals are used to recognize an ear. In the next step, a ring-to-line mapping technique is used to map these major axis lengths to several straight-line signals. Next, the complex plane vector computing technique is used to determine the similarity of these major axis lengths, whereby a solution to the image-rotating problem is achieved. The aforementioned extracted signals are also compared to the ones that are extracted from its neighboring pixels, whereby solving the image-shifting problem. The algorithm developed in this study can precisely identify an ear image by solving the image rotation and image shifting problems.

• The Bulletin of The Korean Astronomical Society
• /
• v.39 no.2
• /
• pp.96.2-96.2
• /
• 2014

Using solar occultation data obtained by Cassini/Visual Infrared Mapping Spectrometer (VIMS), we were able to retrieve the 1 - 5 mm optical-depth spectra of the Titanian haze, for which only selected wavelength and altitudinal ranges were previously analyzed. We found that the gross 1 - 5 mm shapes of the retrieved haze spectra are significantly different from the spectra of tholin samples in the literature. We also derived the vertical variation of the spectral structure of the $3.3-3.4{\mu}m$ absorption feature of the Titan haze from the VIMS data recorded between 250 and 700 km altitude. We found a marked change between 480 and 580 km in the relative amplitudes of the 3.33 and $3.38{\mu}m$ features which are characteristic of aromatic (double C=Cchains or rings) or aliphatic (single C-C chains) structural groups, respectively. Dicussions on this spectral and altitudinal variation will be presented.

• Proceedings of the KSPS conference
• /
• 1996.10a
• /
• pp.448-455
• /
• 1996

In this paper, we have proposed a VQ algorithm which uses a generating order to make quantize feature vector of speech signal. The proposed algorithm inspects what codeword follows a(ter present codeword and adds new index to established codebook, when mapping speech signal. We present a variable bit rate for new codebook, and propose an efficient compressed way of information. In this way, the number of computation and the number of codewords to be searched are reduced considerably. The performance of the proposed VQ algorithm is evaluated by spectrum distortion measure and bit rate. The obtained spectrum distortion is reduced about 0.22 [db], and the bit rate is saved over 0.21 bit/frame.

• The Bulletin of The Korean Astronomical Society
• /
• v.35 no.1
• /
• pp.58.2-58.2
• /
• 2010

L1251B, which was revealed as a small group of protostars by the Spitzer Space Telescope (SST), presents a great site of studying chemical evolution in gas and ice as well as various dynamical processes associated with star formation (infall, rotation, and outflow). We have mapped L1251B with the Infrared Spectrograph (IRS) aboard the SST to study the chemical distribution in the phases of gas and ice and the dynamical feature related to shock in the region. Various atomic lines and the $H_2$ pure rotational lines, which trace different shock velocities, were detected. In addition, the distribution of the water and $CO_2$ ices hints variety of the ice desorption mechanism in L1251B.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.04b
• /
• pp.532-534
• /
• 2000

복합적인 감성기반 영상 검색 시스템을 구축하기 위해서는 감성속성으로 영상을 찾는 검색은 물론이고, 주어진 영상의 감성특성을 알아내는 과정이 필요하다. 본 논문에서는 영상의 특성으로부터 감성을 매핑하는 신경망을 구축하고 다양한 실험으로 그 가능성을 보인다. 여기에서 영상특징으로 웨이블렛계수와 위치정보를 사용했고, 감성공간으로는 SD법으로부터 14개의 형용사쌍을 추출했다. 이 두 공간의 매핑에 사용된 신경망의 입력으로 영상에서 얻은 RGB 색상당 36개의 총 108개의 웨이블렛 개수를 사용했고, 출력은 14개의 감속속성당 7등급으로 총 98개로 구성했다. 총 6명이 영상을 보고 평가한 감성평가데이터중에서 2명이 각각 평가한 데이터로 신경망을 학습시키고 나머지 10개로 테스트한 경우는 90%이상의 인식률을 보였다. 4명이 각각 90개씩 평가한 데이터로 신경망을 학습시키고 나머지 10개로 테스트한 경우는 90%의 인식률을 보였다. 또한 공통된 감성을 신경망을 통해 인식할 수 있는지 판단하기 위해 600개씩 2명으로부터 얻은 1200개의 데이터에 대해서 1000개를 학습시키고 200개를 테스트하고, 100개씩 4명으로부터 데이터에 대해서 360개를 학습시키고 40개를 테스트해 본 결과, 전자의 경우 오류율 8, 후자의 경우 0.7~0.8 범위였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2016.10a
• /
• pp.234-237
• /
• 2016

Address Resolution Protocol (ARP) is used for mapping a network address to physical address in many network technologies. However, since ARP protocol has no security feature, it always abused by attackers for performing ARP-based attacks. Researchers presented many technologies to improve ARP protocol, but most of them require a high implementation cost or scarify the network performance for using ARP protocol securely. In this paper, we present an ARP-disabled network system to neutralize the ARP-based attacks. "ARP-disabled" means suppress the ARP messages like request, response and broadcast messages, but not the ARP table. In our system, ARP tables are used for managing static ARP entries without prior knowledge (e.g. IP, MAC list of client devices). This is possible because the MAC address was designed to be derived from IP address. In general, our system is safe from the ARP-based attacks even the attacker has a strong power. Moreover, we saved network bandwidth by disabling the ARP messages.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11b
• /
• pp.259-261
• /
• 2005

분류 방법으로서의 SVM(Support Vector Machine)은 커널 방법과 함께 사용됨으로써 그 유용성을 크게 향상시켰다. 커널 방법은 일반적으로 입력 데이터의 자질(feature)로 나타내는 공간으로부터 높은 차원의 공간으로 데이터를 사상(mapping)시키는 역할을 하게 되나, 기본적으로는 데이터간에 새로운 거리(metric)를 부설해주는 역할을 하는 것이다. 지금까지 나온 다양한 커널 방법은 구조화된(structured) 데이터에 대해 커널 형태로 거리를 부여하는 방법을 제시한다. 본 논문에서는 DNA의 작용을 모델링하여 만든 새로운 커널이 miRNA(micro RNA)와 mRNA(messenger RNA)쌍에 대한 발현 여부를 분류해 내기 위해 커널 형식으로 거리를 부여하는 방법을 보인다. 이 방법은 실리콘 컴퓨터가 아닌 실제 DNA분자로 실험할 수 있도록 설계된 것을 고려할 때 여러 종류의 DNA 코드를 분석하는 데 사용될 수 있는 새로운 분자컴퓨팅 방법이다.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.39 no.9
• /
• pp.985-996
• /
• 1990

An inverse kinematic solution of a robot manipulator using multilayer perceptrons is proposed. Neural networks allow the solution of some complex nonlinear equations such as the inverse kinematics of a robot manipulator without the need for its model. However, the back-propagation (BP) learning rule for multilayer perceptrons has the major limitation of being too slow in learning to be practical. In this paper, a new algorithm named Dynamically Reconfiguring BP is proposed to improve its learning speed. It uses a modified version of Kohonen's Self-Organizing Feature Map (SOFM) to partition the input space and for each input point, select a subset of the hidden processing elements or neurons. A subset of the original network results from these selected neuron which learns the desired mapping for this small input region. It is this selective property that accelerates convergence as well as enhances resolution. This network was used to learn the parity function and further, to solve the inverse kinematic problem of a robot manipulator. The results demonstrate faster learning than the BP network.