• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.68 no.1
• /
• pp.153-158
• /
• 2019

In this paper, we analyze the infrared feature for the small coast targets according to the surrounding environment for autonomous flight device equipped with an infrared imaging sensor and we propose Cross Duality of Neural Network (CR-DuNN) method which can classify the target and clutter in coastal environment. In coastal environment, there are various property according to diverse change of air temperature, sea temperature, deferent seasons. And small coast target have various infrared feature according to diverse change of environment. In this various environment, it is very important thing that we analyze and classify targets from the clutters to improve target detection accuracy. Thus, we propose infrared feature learning algorithm through LSTM neural network and also propose CR-DuNN algorithm that integrate LSTM prediction network with Du-CNN classification network to classify targets from the clutters.

• Journal of Korea Society of Industrial Information Systems
• /
• v.19 no.1
• /
• pp.77-85
• /
• 2014

Medicine-bottle classification algorithm to avoid medicine accidents must be robust to a geometric change such as rotation, size variation, location movement of the medicine bottles. In this paper, we propose an algorithm to classify the medicine bottles exactly in real-time by using SIFT(Scale Invariant Feature Transform) which is robust to the geometric change. In first, we classify medicine bottles by size using minimum boundary rectangle(MBR) of medicine bottles as a striking feature in order to classify the medicine bottles. We extract label region in the MBR and the region of interest(ROI) considering rotation. Then, we classify medicine bottles using SIFT for the extracted ROI. We also simplify the number of octave of SIFT in order to improve a process speed of SIFT. We confirm to classify all the medicine bottles exactly as a result of performance evaluation of the proposed algorithm about images of 250 medicine bottles. We also confirm to improve the process time more than twice the processing time by simplifying the number of octave of SIFT.

• Journal of Sasang Constitutional Medicine
• /
• v.22 no.1
• /
• pp.17-25
• /
• 2010

1. Objectives: Voice diagnosis has been used to classify individuals into the Sasang constitution in SCM(Sasang Constitution Medicine) and to recognize his/her health condition in TKM(Traditional Korean Medicine). In this paper, we purposed a new speech classification algorithm for Sasang constitution. 2. Methods: This algorithm is based on the SVM(Support Vector Machine) technique, which is a classification method to classify two distinct groups by finding voluntary nonlinear boundary in vector space. It showed high performance in classification with a few numbers of trained data set. We designed for this algorithm using 3 SVM classifiers to classify into 4 groups, which are composed of 3 constitutional groups and additional indecision group. 3. Results: For the optimal performance, we found that 32.2% of the voice data were classified into three constitutional groups and 79.8% out of them were grouped correctly. 4. Conclusions: This new classification method including indecision group appears efficient compared to the standard classification algorithm which classifies only into 3 constitutional groups. We find that more thorough investigation on the voice features is required to improve the classification efficiency into Sasang constitution.

• Journal of Information Technology Applications and Management
• /
• v.9 no.2
• /
• pp.1-16
• /
• 2002

Recently, neural network methods have been studied to find out more valuable information in data bases. But the supervised learning methods of neural networks have an overfitting problem, which leads to errors of target patterns. And the unsupervised learning methods can distort important information in the process of regularizing data. Thus they can't efficiently classify data, To solve the problems, this paper introduces a hybrid neural networks HACAB(Hybrid Algorithm combining a Competition learning model And BP Algorithm) combining a competition learning model and 8P algorithm. HACAB is designed for cases which there is no target patterns. HACAB makes target patterns by adopting a competition learning model and classifies input patterns using the target patterns by BP algorithm. HACAB is evaluated with random input patterns and Iris data In cases of no target patterns, HACAB can classify data more effectively than BP algorithm does.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.190-195
• /
• 2004

We propose a pattern classification algorithm using self-nonself discrimination principle of immune cells and apply it to DNA pattern classification problem. Pattern classification problem in bioinformatics is very important and frequent one. In this paper, we propose a classification algorithm based on the negative selection of the immune system to classify DNA patterns. The negative selection is the process to determine an antigenic receptor that recognize antigens, nonself cells. The immune cells use this antigen receptor to judge whether a self or not. If one composes ${\eta}$ groups of antigenic receptor for ${\eta}$ different patterns, these receptor groups can classify into ${\eta}$ patterns. We propose a pattern classification algorithm based on the negative selection in nucleotide base level and amino acid level. Also to show the validity of our algorithm, experimental results of RNA group classification are presented.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.47 no.6
• /
• pp.1-5
• /
• 2010

We propose an algorithm which can classify the system should use a PI controller, which have a weak high frequency attenuation characteristics near the critical frequency. To classify the system, we use a time delay element to calculate a gain attenuation rate near the critical frequency. The proposed algorithm also can design PI controller with the given magnitude margin and phase margin specification. The proposed algorithm uses time delay element and saturation function to identify the one point information in frequency domain. We justify the proposed algorithm via the simulation.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.6
• /
• pp.802-814
• /
• 1993

This paper propose an algorithm for preprocessing of character recognition, which classify characters into Hangeul and Hanja. In this study, we use the 9 structural chacteristics of Hanja which isn't affected by deformation of size and style of characters and rates based on character size to classify characters. Firstly, we process the blocking to segment each characters. Secondly, on this segmented characters, we apply algorithm proposed in this paper to classify Hangeul and Hanja. Finally, we classify characters into Hangeul and Hanja, respectively. An experiment with 2350 Hangeul and 4888 Hanja printed Gothic and Mincho style of KS-C 5601 are carried out. We experiment on typeface sample book, newspapers, academic society's papers, magazines, textbooks and documents written out word processor to obtain the classifying rates of 98.8%, 92%, 96%, 98% and 98%, respectively.

• Journal of Information Processing Systems
• /
• v.20 no.2
• /
• pp.215-225
• /
• 2024

With the rapid development of Internet of Things (IoT) and big data technology, a large amount of data will be generated during the operation of related industries. How to classify the generated data accurately has become the core of research on data mining and processing in IoT industry chain. This study constructs a classification model of IoT industry chain based on improved random forest algorithm and text analysis, aiming to achieve efficient and accurate classification of IoT industry chain big data by improving traditional algorithms. The accuracy, precision, recall, and AUC value size of the traditional Random Forest algorithm and the algorithm used in the paper are compared on different datasets. The experimental results show that the algorithm model used in this paper has better performance on different datasets, and the accuracy and recall performance on four datasets are better than the traditional algorithm, and the accuracy performance on two datasets, P-I Diabetes and Loan Default, is better than the random forest model, and its final data classification results are better. Through the construction of this model, we can accurately classify the massive data generated in the IoT industry chain, thus providing more research value for the data mining and processing technology of the IoT industry chain.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.6
• /
• pp.819-825
• /
• 2007

In this paper, we propose a fault diagnosis algorithm of induction motors by DFT and wavelet. We extract a feature vector using a fault pattern extraction method by DFT in frequency domain and wavelet transform in time-frequency domain. And then we deal with a fusion algorithm for the feature vectors extracted from DFT and wavelet to classify the faults of induction motors. Finally, we provide an experimental results that the proposed algorithm can be successfully applied to classify the several fault signals acquired from induction motors.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.31 no.3
• /
• pp.41-51
• /
• 2006

Clustering is to group similar objects into clusters. Until now there are a lot of approaches using Self-Organizing feature Maps (SOFMS) But they have problems with a small output-layer nodes and initial weight. For example, one of them is a one-dimension map of c output-layer nodes, if they want to make c clusters. This approach has problems to classify elaboratively. This Paper suggests one-dimensional output-layer nodes in SOFMs. The number of output-layer nodes is more than those of clusters intended to find and the order of output-layer nodes is ascending in the sum of the output-layer node's weight. We un find input data in SOFMs output node and classify input data in output nodes using Euclidean distance. The proposed algorithm was tested on well-known IRIS data and TSPLIB. The results of this computational study demonstrate the superiority of the proposed algorithm.