• Journal of KIISE:Software and Applications
• /
• v.31 no.7
• /
• pp.911-921
• /
• 2004

Accurate minutiae extraction from input fingerprint images is one of the critical modules in robust automatic fingerprint identification system. However, the performance of a minutiae extraction is heavily dependent on the quality of the input fingerprint images. If the preprocessing is performed according to the fingerprint image characteristics in the image enhancement step, the system performance will be more robust. In this paper, we propose a knowledge-based preprocessing method, which extracts S features (the mean and variance of gray values, block directional difference, orientation change level, and ridge-valley thickness ratio) from the fingerprint images and analyzes image quality with Ward's clustering algorithm, and enhances the images with respect to oily/neutral/dry characteristics. Experimental results using NIST DB 4 and Inha University DB show that clustering algorithm distinguishes the image Quality characteristics well. In addition, the performance of the proposed method is assessed using quality index and block directional difference. The results indicate that the proposed method improves both the quality index and block directional difference.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.14 no.5
• /
• pp.15-21
• /
• 1977

The syntactic pattern recognition system of Korean characters is composed of three main functional parts; Preprocessing, Graph-representation, and Segmentation. In preprocessing routine, the input pattern has been thinned using the Hilditch's thinning algorithm. The graph-representation is the detection of a number of nodes over the input pattern and codification of branches between nodes by 8 directional components. Next, segmentation routine which has been implemented by top down nondeterministic parsing under the control of tree grammar identifies parts of the graph-represented Pattern as basic components of Korean characters. The authors have made sure that this system is effective for recognizing Korean characters through the recognition simulations by digital computer.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.05a
• /
• pp.47-50
• /
• 2000

In this paper, an optimal identification method using Multi-FNN(Fuzzy-Neural Network) is proposed for model ins of nonlinear complex system. In order to control of nonlinear process with complexity and uncertainty of data, proposed model use a HCM clustering algorithm which carry out the input-output data preprocessing function and Genetic Algorithm which carry out optimization of model. The proposed Multi-FNN is based on Yamakawa's FNN and it uses simplified inference as fuzzy inference method and Error Back Propagation Algorithm as learning rules. HCM clustering method which carry out the data preprocessing function for system modeling, is utilized to determine the structure of Multi-FNN by means of the divisions of input-output space. Also, the parameters of Multi-FNN model such as apexes of membership function, learning rates and momentum coefficients are adjusted using genetic algorithms. Also, a performance index with a weighting factor is presented to achieve a sound balance between approximation and generalization abilities of the model, To evaluate the performance of the proposed model, we use the time series data for gas furnace and the numerical data of nonlinear function.

• Journal of KIISE:Computing Practices and Letters
• /
• v.9 no.5
• /
• pp.509-520
• /
• 2003

In this paper, we developed a new and robust algorithm for a practical and very efficient MPEG-4 video coding. The MPEG-4 video group has developed the video Verification Model(VM) which evolved through time by means of core experiments. And in the standardization process, MS-FDAM was developed based on the standard document of ISO/IEC 14496-2 and VM as a reference MPEG-4 coding system. But MS -FDAM has drawbacks in practical MPEG-4 coding and it does not have the VOP extraction functionality. In this research, we implemented a preprocessing system for a real-time input and the VOP extraction for a practical content-based MPEG-4 video coding and also implemented the motion detection to achieve the high compression rate of 180:1.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.5 no.1
• /
• pp.43-51
• /
• 1995

In this paper, we propose a neural network with fuzzy preprocessor not only for improving the classifi¬cation accuracy but also for being able to classify objects whose attribute values do not have clear bound¬aries. The fuzzy input signal representation scheme is included as a preprocessing module. It transforms imprecise input in linguistic form and precisely stated numerical input into multidimensional numerical values. 'The transformed input is processed in the postprocessing module. The experimental results indi-cate the superiority of fuzzy input signal representation scheme in comparison to binary input signal rep¬resentation scheme and decimal input signal representation scheme.

• International Journal of Automotive Technology
• /
• v.7 no.5
• /
• pp.637-644
• /
• 2006

The detection of engine misfire events is one of major concerns in engine control due to its negative effect on air pollution and engine performance. In this paper, a misfire detection system based on crankshaft angular speed fluctuation is developed. Synthetic variable method is adopted for the preprocessing of crankshaft angular speed. This method successfully estimates the work output of each cylinder by finding the effect of combustion energy on the crankshaft rotational speed or acceleration after virtually removing the effect of the internal inertia forces from the measured crankshaft speed signals. The detection system is developed using neural network with the revised synthetic angular acceleration as input which is derived from the preprocessing. Mathematical simulation is carried out for developing and verifying the misfire detection system. Finally, the reliability of the developed system is validated through an experiment.

• Proceedings of the KIEE Conference
• /
• 2000.11d
• /
• pp.666-668
• /
• 2000

In this paper, a fuzzy prediction method is proposed for time series data having uncertainty and non-stationary characteristics. Conventional methods, which use past data directly in prediction procedure, cannot properly handle non-stationary data whose long-term mean is floating. To cope with this problem, a data preprocessing technique utilizing the differences of original time series data is suggested. The difference sets are established from data. And the optimal difference set is selected for input of fuzzy predictor. The proposed method based the Takigi-Sugeno-Kang(TSK or TS) fuzzy rule. Computer simulations show improved results for various time series.

• Proceedings of the IEEK Conference
• /
• 1999.11a
• /
• pp.560-563
• /
• 1999

In this paper, we present an optimal hardware implementation for preprocessing of a person identification algorithm utilizing vein pattern of dorsal surface of hand. For the vein pattern recognition, the computational burden of the algorithm lies mainly in the preprocessing of the input images, especially in lowpass filtering. we could reduce the identification time to one tenth by hardware design of the lowpass filter compared to sequential computations. In terms of the computation accuracy, the simulation results show that the CSD code provided an optimized coefficient value with about 91.62% accuracy in comparison with the floating point implementation of current coefficient value of the lowpass filter. The post-simulation of a VHDL model has been performed by using the ModelSim$^{TM}$. The implemented chip operates at 20MHz and has the operational speed of 55.107㎳.㎳.

• Journal of Multimedia Information System
• /
• v.3 no.2
• /
• pp.1-8
• /
• 2016

In this paper, we propose a method to automatically diagnose various diseases. The input data consists of electrocardiograph (ECG) recordings. We extract R-to-R interval (RRI) signals from ECG recordings, which are preprocessed to remove trends and ectopic beats, and to keep the signal stationary. After that, we perform some prospective analysis to extract time-domain parameters, frequency-domain parameters, and nonlinear parameters of the signal. Those parameters are unique for each disease and can be used as the statistical symptoms for each disease. Then, we perform feature selection to improve the performance of the diagnosis classifier. We utilize the selected features to diagnose various diseases using machine learning. We subsequently measure the performance of the machine learning classifier to make sure that it will not misdiagnose the diseases. The first two steps, which are R-to-R extraction and preprocessing, have been successfully implemented with satisfactory results.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.3A
• /
• pp.260-268
• /
• 2010

Channel inversion is one of the simplest techniques for multiuser downlink systems with single-antenna users. In this paper, we extend the regularized channel inversion technique developed for the single-antenna user case to multiuser multiple-input multiple-output (MIMO) channels with multiple-antenna users. We first employ the multiuser preprocessing to project the multiuser signals near the null space of the unintended users based on the MMSE criterion, and then the single-user preprocessing is applied to the decomposed MIMO interference channels. In order to reduce the complexity, we focus on non-iterative solutions for the multiuser transmit beamforming and use a linear receiver based on an MMSE criterion. Simulation results show that the proposed scheme outperforms existing joint iterative algorithms in most multiuser configurations.