• Journal of Korea Multimedia Society
• /
• v.8 no.11
• /
• pp.1556-1564
• /
• 2005

Various optimization techniques are deployed in the compilation process of a source program for improving the program's execution speed and reducing the size of the source code. Of the optimization pattern matching techniques, the string pattern matching technique involves finding an optimal pattern that corresponds to the intermediate code. However, it is deemed inefficient due to excessive time required for optimized pattern search. The tree matching pattern technique can result in many redundant comparisons for pattern determination, and there is also the disadvantage of high cost involved in constructing a code tree. The objective of this paper is to propose a table-driven code optimizer using the DFA(Deterministic Finite Automata) optimization table to overcome the shortcomings of existing optimization techniques. Unlike other techniques, this is an efficient method of implementing an optimizer that is constructed with the deterministic automata, which determines the final pattern, refuting the pattern selection cost and expediting the pattern search process.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.4 no.1
• /
• pp.25-44
• /
• 2010

With the increasing popularity of mobile devices, it has become necessary to protect private information and content in these devices. Face recognition has been favored over conventional passwords or security keys, because it can be easily implemented using a built-in camera, while providing user convenience. However, because mobile devices can be used both indoors and outdoors, there can be many illumination changes, which can reduce the accuracy of face recognition. Therefore, we propose a new face recognition method on a mobile device robust to illumination variations. This research makes the following four original contributions. First, we compared the performance of face recognition with illumination variations on mobile devices for several illumination normalization procedures suitable for mobile devices with low processing power. These include the Retinex filter, histogram equalization and histogram stretching. Second, we compared the performance for global and local methods of face recognition such as PCA (Principal Component Analysis), LNMF (Local Non-negative Matrix Factorization) and LBP (Local Binary Pattern) using an integer-based kernel suitable for mobile devices having low processing power. Third, the characteristics of each method according to the illumination va iations are analyzed. Fourth, we use two matching scores for several methods of illumination normalization, Retinex and histogram stretching, which show the best and $2^{nd}$ best performances, respectively. These are used as the inputs of an SVM (Support Vector Machine) classifier, which can increase the accuracy of face recognition. Experimental results with two databases (data collected by a mobile device and the AR database) showed that the accuracy of face recognition achieved by the proposed method was superior to that of other methods.

• Journal of KIISE
• /
• v.44 no.7
• /
• pp.668-673
• /
• 2017

Deep neural networks have shown remarkable performance in various fields of pattern recognition such as voice recognition, image recognition and object detection. However, underlying mechanisms of the network have not been fully revealed. In this paper, we focused on empirical analysis of the network parameters. The Faster R-CNN(region-based convolutional neural network) was used as a baseline network of our work and three important parameters were analyzed: the dropout ratio which prevents the overfitting of the neural network, the size of the anchor boxes and the activation function. We also compared the performance of dropout and batch normalization. The network performed favorably when the dropout ratio was 0.3 and the size of the anchor box had not shown notable relation to the performance of the network. The result showed that batch normalization can't entirely substitute the dropout method. The used leaky ReLU(rectified linear unit) with a negative domain slope of 0.02 showed comparably good performance.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.16 no.12
• /
• pp.1324-1335
• /
• 1991

In this paper, a phoneme separation and an establishment of discriminative pattern of Korean phonemes are studied on experiment. The separation uses parameters such as pitch extraction, glottal peak pulse width of each pitch. speech duration. envelope and amplitude bias. The first pitch is extracted by deviations of glottal peak and width. energy and normalization on a bias on the top of vowel envelope. And then, it traces adjacent pitch to vowel in whole. On vewel, amethod to be reduced gliding pattern and the possible of vowel distinction to be used just second formant are proposed, and shrinking pitch waveform has nothing to do with pitch length is estimated. A pattern of envelope, spectrum, shrinking waveform, and a method of analysis by mutual relation among phonemes and manners of articulation on consonant are detected. As experimental results, 90% on vowel phoneme, 80% and 60% on initial and final consonant are discriminated.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.15 no.2
• /
• pp.386-394
• /
• 1995

The fringes developed by misalignment, an application of Moire method measuring small deformation of specimen, was transformed to frequency function. After that, theory of normalization was applied to derive the equations explaining the relation of the deformation and Moire fringes. Above all, the equations were produced to explain the rotation and increasing of fringes. In addition to that, the relation of fringe number and strain was illustated with the equations deduced from frequency function and geometrical method respectively. These two expressions were more effective than the used ones owing to the one can accommodate the other.

• Korean Journal of Applied Biomechanics
• /
• v.15 no.2
• /
• pp.155-165
• /
• 2005

The purpose of this study was to compare the muscle activities of Double Legged Jump (DLJ) and Single Legged Jump (SLJ) by the normalization of muscle activity. Eight college students without the lower extremity injuries were selected as subjects for collecting EMG data of vastus medialis and gastrocnemius. The entire section of motion was established as eccentric and concentric contractions, and each of the contractions was divided into three sections with equal timing intervals, which becomes a total of 6 phases. The EMG data of each phase was integrated and normalized. The muscle activities of the vastus medialis for both eccentric and concentric contractions were significantly different between DLJ and SLJ(p<.05). The increase in overall muscle activity of SLJ was 33.6%. Approximately, there was an increase of 25.9% in eccentric contraction and 40% in concentric contraction. Moreover, the data of the muscle activity of gastrocnemius was similar to the data of the muscle activity of vastus medialis. In conclusion, this research suggests muscle activity of a certain motion can be normalized for an analysis of another motion.

• Journal of Korea Multimedia Society
• /
• v.23 no.12
• /
• pp.1486-1495
• /
• 2020

The analysis of digital microscopy images plays a vital role in computer-aided diagnosis (CAD) and prognosis. The main purpose of this paper is to develop a machine learning technique to predict the histological grades in prostate biopsy. To perform a multiclass classification, an AI-based deep learning algorithm, a multichannel convolutional neural network (MCCNN) was developed by connecting layers with artificial neurons inspired by the human brain system. The histological grades that were used for the analysis are benign, grade 3, grade 4, and grade 5. The proposed approach aims to classify multiple patterns of images extracted from the whole slide image (WSI) of a prostate biopsy based on the Gleason grading system. The Multichannel Convolution Neural Network (MCCNN) model takes three input channels (Red, Green, and Blue) to extract the computational features from each channel and concatenate them for multiclass classification. Stain normalization was carried out for each histological grade to standardize the intensity and contrast level in the image. The proposed model has been trained, validated, and tested with the histopathological images and has achieved an average accuracy of 96.4%, 94.6%, and 95.1%, respectively.

• The Journal of the Korea Contents Association
• /
• v.5 no.4
• /
• pp.94-102
• /
• 2005

In this paper, three axial accelerometer was used to develop a small sensor module, which was attached to human body to calculate the acceleration in gravity direction by human motion, when it was positioned in any direction. To measure its wearer's walking or running motion using the sensor module, the acquired sensor data was pre-processed to enable its quantitative analysis. The acquired digital data was transformed to orthogonal coordinate value in three dimension and calculated to be single scalar acceleration data in gravity direction and normalized to be physical unit value. The normalized sensor data was used to detect walking pattern and calculate their step counts. Developed algorithm was implemented in the form of PDA application. The accuracy of the developed sensor to detect step count was about 97% in laboratory experiment.

• Transactions on Control, Automation and Systems Engineering
• /
• v.2 no.1
• /
• pp.40-46
• /
• 2000

In this paper, a transient identification based on a Hidden Markov Model (HMM) has been suggested and evaluated experimentally for the classification of transients in the dynamic process. The transient can be identified by its unique time dependent patterns related to the principal variables. The HMM, a double stochastic process, can be applied to transient identification which is a spatial and temporal classification problem under a statistical pattern recognition framework. The HMM is created for each transient from a set of training data by the maximum-likelihood estimation method. The transient identification is determined by calculating which model has the highest probability for the given test data. Several experimental tests have been performed with normalization methods, clustering algorithms, and a number of states in HMM. Several experimental tests have been performed including superimposing random noise, adding systematic error, and untrained transients. The proposed real-time transient identification system has many advantages, however, there are still a lot of problems that should be solved to apply to a real dynamic process. Further efforts are being made to improve the system performance and robustness to demonstrate reliability and accuracy to the required level.

• The Journal of the Acoustical Society of Korea
• /
• v.18 no.1E
• /
• pp.27-31
• /
• 1999

Recent linguistic researches have brought into focus the relations between prosodic structures and syntactic, semantic or phonological structures. Most of them prove that prosodic information is available for understanding syntactic, semantic and discourse structures. But this result has not been integrated yet into recent Korean speech recognition or understanding systems. This study, as a part of integrating prosodic information into the speech recognition system, proposes an automatic detection technique of Korean accentual phrase boundaries by using one-stage DP, and the normalized pitch pattern. For making the normalized pitch pattern, this study proposes a method of modified normalization for Korean spoken language. For the experiment, this study employs 192 sentential speech data of 12 men's voice spoken in standard Korean, in which 720 accentual phrases are included, and 74.4% of the accentual phrase boundaries are correctly detected while 14.7% are the false detection rate.