• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.3 no.1
• /
• pp.27-31
• /
• 2003

This paper proposes a robust on-line signature verification system based on a new segmentation method and fusion scheme. The proposed segmentation method resolves the problem of segment-to-segment comparison where the variation between reference signature and input signature causes the errors in the location and the number of segments. In addition, the fusion scheme is adopted, which discriminates genuineness by calculating each feature vector's fuzzy membership degree yielded from the proposed segmentation method. Experimental results show that the proposed signature verification system has lower False Reject Rate(FRR) for genuine signature and False Accept Rate(FAR) for forgery signature.

• Journal of information and communication convergence engineering
• /
• v.4 no.1
• /
• pp.36-40
• /
• 2006

A new method of digital image analysis technique for medical images of cancer cell is presented. This paper deals with the cancer cell discrimination. The object images were the Thyroid Gland cell images that were diagnosed as normal and abnormal. This paper proposes a new discrimination method based on fuzzy logic algorithm. The focus of this paper is an automatic discrimination of cells into normal and abnormal of medical images by dominant feature parameters method with fuzzy algorithm. As a consequence of using fuzzy logic algorithm, the nucleus were successfully diagnosed as normal and abnormal. As for the experimental result, average recognition rate of 64.66% was obtained by applying single parameter of 16 feature parameters at a time. The discrimination rate of 93.08% was obtained by proposed method.

• Proceedings of the KIPE Conference
• /
• 2005.07a
• /
• pp.487-489
• /
• 2005

This paper presents a wavelet-based neural network technology for the detection and classification of the short durations type of power quality disturbances. Transients happen during very short durations to the nano- and microsecond. Thus, a method for detecting and classifying transient signals at the same time and In an automatic combines the properties of the wavelet transform and the advantages of neural networks. Especially, the additional feature extraction to improve the recognition rate is considered. The configuration of the hardware of TMS320C6711 DSP based with 16 channel 20Mhz sampling rate A/D(Analog to Digital) converter and some case studies are described.

• ETRI Journal
• /
• v.27 no.4
• /
• pp.461-464
• /
• 2005

Recently, a variety of confidence measures for utterance verification has been studied to improve speech recognition performance by rejecting out-of-vocabulary inputs. Most of the conventional confidence measures for utterance verification are based primarily on hypothesis testing or an approximated posterior probability, and their performances depend on the robustness of an alternative hypothesis or the prior probability. We introduce a novel confidence measure called a search confusion rate (SCR), which does not require an alternative hypothesis or the approximation of posterior probability. Our confusion-based approach shows better performance in additive noise-corrupted speech as well as in clean speech.

• KIEE International Transactions on Electrophysics and Applications
• /
• v.5C no.3
• /
• pp.138-142
• /
• 2005

In this paper, we compared recognition rates between NN(neural networks) and clustering method as a scheme of off-line PD(partial discharge) diagnosis which occurs at the stator coil of traction motor. To acquire PD data, three defective models are made. PD data for classification were acquired from PD detector. And then statistical distributions are calculated to classify model discharge sources. These statistical distributions were applied as input data of two classification tools, BP(Back propagation algorithm) and ANFIS(adaptive network based fuzzy inference system) pre-processed FCM(fuzzy c-means) clustering method. So, classification rate of BP were somewhat higher than ANFIS. But other items of ANFIS were better than BP; learning time, parameter number, simplicity of algorithm.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.43.2-43
• /
• 2001

In this paper, we consider the problems of the evolutionary designed neural networks structures by genetic algorithm. Neural networks has been applied to various application fields since back-propagation algorithm was proposed, e.g. function approximation, pattern or character recognition and so on. However, one of difficulties to use the neural networks. It is how to design the structure of the neural network. Researchers and users design networks structures and training parameters such as learning rate and momentum rate and so on, by trial and error based on their experiences. In the case of designing large scales neural networks, it is very hard work for manually design by try and error. For this difficulty, various structural learning algorithms have been proposed. Especially, the technique of using genetic algorithm for networks structures design has been ...

• Journal of the Korea Safety Management & Science
• /
• v.19 no.4
• /
• pp.53-62
• /
• 2017

It has come to attention that a risk-assessing organization, that will benchmark a company's safety department, is imperative, following an increase in large-scale SOC-business project, construction of higher-raised buildings, development of underground space; all that have increase accident rates. Having faced problems that arise in firms that demand diversity, complexity and instantaneity, the purpose of the thesis is to arrive at efficient and practical problem-solving means. In order to solve the problems that would surface theoretically during an actual risk assessment, the state of the operation systems of the top five national construction firms having a hazard rate of 0.25 times less than the average rate have been analyzed, while a hierarchal recognition research of the employees who not only function at the operating level but are the practice subjects of a firm, has also been conducted, bringing the main text.

• Journal of the Korean Society of Industry Convergence
• /
• v.24 no.1
• /
• pp.1-7
• /
• 2021

In this paper, we presented the experimental basis for the theoretical background and robustness of the Convolutional Neural Network for object recognition based on artificial intelligence. An experimental result was performed to visualize the weighting filters and feature maps for each layer to determine what characteristics CNN is automatically generating. experimental results were presented on the trend of learning error and identification error rate by checking the relevance of the weight filter and feature map for learning error and identification error. The weighting filter and characteristic map are presented as experimental results. The automatically generated characteristic quantities presented the results of error rates for moving and rotating robustness to geometric changes.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.4
• /
• pp.35-41
• /
• 2001

This paper describes Linear Discriminant Analysis and common vector extraction for speech recognition. Voice signal contains psychological and physiological properties of the speaker as well as dialect differences, acoustical environment effects, and phase differences. For these reasons, the same word spelled out by different speakers can be very different heard. This property of speech signal make it very difficult to extract common properties in the same speech class (word or phoneme). Linear algebra method like BT (Karhunen-Loeve Transformation) is generally used for common properties extraction In the speech signals, but common vector extraction which is suggested by M. Bilginer et at. is used in this paper. The method of M. Bilginer et al. extracts the optimized common vector from the speech signals used for training. And it has 100％ recognition accuracy in the trained data which is used for common vector extraction. In spite of these characteristics, the method has some drawback-we cannot use numbers of speech signal for training and the discriminant information among common vectors is not defined. This paper suggests advanced method which can reduce error rate by maximizing the discriminant information among common vectors. And novel method to normalize the size of common vector also added. The result shows improved performance of algorithm and better recognition accuracy of 2％ than conventional method.

• Journal of Trauma and Injury
• /
• v.19 no.2
• /
• pp.113-120
• /
• 2006

Purpose: The aim of this study was to evaluate the clinical characteristics of classic heat stroke in Korea and to identify factors of prognosis for heat stroke by comparing a survival group with a non-survival group. Methods: We retrospectively analyzed 27 patients with heat stroke who visited the Emergency Department of Kyungpook National University Hospital from March 2001 to February 2005. First, we divided the patients into two groups, the classic heat stroke group and the exertional heat stroke group, and compared them. Second, we compared the survival group with the non-survival group. Age, sex, cause, place where patients were found, underlying diseases, cooling time, performance of endotracheal intubation, initial Glasgow Coma Scale, initial vital sign, and laboratory findings were reviewed. Results: Five of 27 patients in heat stroke died. The classic heat stroke group had 20 patients. They were old and had more patients in the bathroom than the exertional heat stroke group had. The non-survival group showed lower blood pressure, lower initial GCS score, and higher respiratory rate than the survival group. In laboratory findings, the non-survival group also showed lower$HCO_3-$ level, lower albumin level, lower glucose level, more prolonged PT, and higher CK-MB level than the survival group. Delay in recognition of heat stroke and cooling were poor prognostic factors in heat stroke. Conclusion: The classic heat stroke group had patients who were old and found in the bathroom. Early recognition and treatment of heat stroke is important to reduce mortality. Cooling time, initial GCS score, mean arterial pressure, resipratory rate, $HCO_3-$, PT, CK-MB, and albumin seem to be meaningful when forming a prognosis for heat stroke patients.