• Journal of Information Processing Systems
• /
• v.16 no.5
• /
• pp.1008-1014
• /
• 2020

Aiming at the defect detection quality of denim fabric, this paper designs an improved algorithm based on the optimized Gabor filter. Firstly, we propose an improved defect detection algorithm of jean fabric based on the maximum two-dimensional image entropy and the loss evaluation function. Secondly, 24 Gabor filter banks with 4 scales and 6 directions are created and the optimal filter is selected from the filter banks by the one-dimensional image entropy algorithm and the two-dimensional image entropy algorithm respectively. Thirdly, these two optimized Gabor filters are compared to realize the common defect detection of denim fabric, such as normal texture, miss of weft, hole and oil stain. The results show that the improved algorithm has better detection effect on common defects of denim fabrics and the average detection rate is more than 91.25%.

• Bulletin of the Society of Naval Architects of Korea
• /
• v.27 no.2
• /
• pp.87-95
• /
• 1990

Conventional methods to estimate natural frequencies and damping ratios of structures from measured response time series obtained during impact tests are reviewed. Maximum Entropy Method and Least Square Prony Method are introduced to alleviate the inherent limitation of the conventional methods. The performance of the methods are explored through computer simulation. As an example of application, they are applied to the time series obtained from an anchor drop-and-snup test of a container ship and the result is compared to that of conventional FFT method. As a result of the computer simulation, it is found that Maximum Entropy Method is very efficient to estimate natural frequencies of structures when two neighboring natural frequencies are close enough and short data records are only available, but it is not a reliable estimator for damping ratios. And it is also found that Least Square Prony Method is efficient to estimate the natural frequencies and damping ratios of highly damped structural system, but the estimation efficiency of damping ratios is significantly deteriorated in the presence of significant additive noise.

• The KIPS Transactions:PartB
• /
• v.15B no.1
• /
• pp.61-68
• /
• 2008

This paper presents a spam filter system using co-training with spamminess features and URL features based on the maximum entropy model. Spamminess features are the emphasizing patterns or abnormal patterns in spam messages used by spammers to express their intention and to avoid being filtered by the spam filter system. Since spammers use URLs to give the details and make a change to the URL format not to be filtered by the black list, normal and abnormal URLs can be key features to detect the spam messages. Co-training with spamminess features and URL features uses two different features which are independent each other in training. The filter system can learn information from them independently. Experiment results on TREC spam test collection shows that the proposed approach achieves 9.1% improvement and 6.9% improvement in accuracy compared to the base system and bogo filter system, respectively. The result analysis shows that the proposed spamminess features and URL features are helpful. And an experiment result of the co-training shows that two feature sets are useful since the number of training documents are reduced while the accuracy is closed to the batch learning.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.44 no.12
• /
• pp.102-107
• /
• 2007

This paper proposes a H.264/AVC entropy decoder without embedded processor nor memory fabrication process. Many researches on H.264/AVC entropy decoders require ROM or RAM fabrication process, which is difficult to be implemented in general digital logic fabrication process. Furthermore, many researches require embedded processors for bitstream manipulation, which increases area and power consumption. This papers proposes hardwired H.264/AVC entropy decoder without embedded processor, which improves data processing speed and reduces power consumption. Furthermore, its CAVLC decoder optimizes lookup table and internal buffer without embedded memory, which reduces hardware size and can be implemented in general digital logic fabrication process without ROM or RAM fabrication process. Designed entropy decoder was embedded in H.264/AVC video decoder, and it was verified to operate correctly in the system. Synthesized in TSMC 90nm fabrication process, its maximum operation frequency is 125MHz. It supports QCIF, CIF, and QVGA image format. Under slight modification of nC register and other blocks, it also support VGA image format.

• Journal of The Korean Astronomical Society
• /
• v.38 no.4
• /
• pp.445-462
• /
• 2005

In a companion paper, we have presented so-called Spatio-Spectral Maximum Entropy Method (SSMEM) particularly designed for Fourier-Transform imaging over a wide spectral range. The SSMEM allows simultaneous acquisition of both spectral and spatial information and we consider it most suitable for imaging spectroscopy of solar microwave emission. In this paper, we run the SSMEM for a realistic model of solar microwave radiation and a model array resembling the Owens Valley Solar Array in order to identify and resolve possible issues in the application of the SSMEM to solar microwave imaging spectroscopy. We mainly concern ourselves with issues as to how the frequency dependent noise in the data and frequency-dependent variations of source size and background flux will affect the result of imaging spectroscopy under the SSMEM. We also test the capability of the SSMEM against other conventional techniques, CLEAN and MEM.

• Journal of the Korean Physical Society
• /
• v.73 no.11
• /
• pp.1644-1649
• /
• 2018

Aiming at infrared thermal images with different buried depth defects, we study a variety of image segmentation algorithms based on the threshold to develop global search ability and the ability to find the defect area accurately. Firstly, the iterative thresholding method, the maximum entropy method, the minimum error method, the Ostu method and the minimum skewness method are applied to image segmentation of the same infrared thermal image. The study shows that the maximum entropy method and the minimum error method have strong global search capability and can simultaneously extract defects at different depths. However none of these five methods can accurately calculate the defect area at different depths. In order to solve this problem, we put forward a strategy of "divide and conquer". The infrared thermal image is divided into several local thermal maps, with each map containing only one defect, and the defect area is calculated after local image processing of the different buried defects one by one. The results show that, under the "divide and conquer" strategy, the iterative threshold method and the Ostu method have the advantage of high precision and can accurately extract the area of different defects at different depths, with an error of less than 5%.

• Smart Structures and Systems
• /
• v.33 no.3
• /
• pp.201-215
• /
• 2024

Spline chirplet transform and local maximum synchrosqueezing are introduced to present a novel structural instantaneous frequency (IF) identification method named local maximum synchrosqueezing spline chirplet transform (LMSSSCT). Namely spline chirplet transform (SCT), a transform is firstly introduced based on classic chirplet transform and spline interpolated kernel function. Applying SCT in association with local maximum synchrosqueezing, the LMSSSCT is then proposed. The index of accuracy and Rényi entropy show that LMSSSCT outperforms the other time-frequency analysis (TFA) methods in processing analytical signals, especially in the presence of noise. Numerical examples of a Duffing nonlinear system with single degree of freedom and a two-layer shear frame structure with time-varying stiffness are used to verify the effectiveness of structural IF identification. Moreover, a nonlinear supported beam structure test is conducted and the LMSSSCT is utilized for structural IF identification. Numerical simulation and experimental results demonstrate that the presented LMSSSCT can effectively identify the IFs of nonlinear structures and time-varying structures with good accuracy and stability.

• Journal of Information Processing Systems
• /
• v.13 no.6
• /
• pp.1554-1574
• /
• 2017

This paper introduces a new type of determining factor for Pseudo Random Strings (PRS). This classification depends upon a mathematical property called Finite Induction (FI). FI is similar to a Markov Model in that it presents a model of the sequence under consideration and determines the generating rules for this sequence. If these rules obey certain criteria, then we call the sequence generating these rules FI a PRS. We also consider the relationship of these kinds of PRS's to Good/deBruijn graphs and Linear Feedback Shift Registers (LFSR). We show that binary sequences from these special graphs have the FI property. We also show how such FI PRS's can be generated without consideration of the Hamiltonian cycles of the Good/deBruijn graphs. The FI PRS's also have maximum Shannon entropy, while sequences from LFSR's do not, nor are such sequences FI random.

• Journal of Advanced Marine Engineering and Technology
• /
• v.40 no.5
• /
• pp.437-446
• /
• 2016

Feature selection has become an essential technique to reduce the dimensionality of data sets. Many features are frequently irrelevant or redundant for the classification tasks. The purpose of feature selection is to select relevant features and remove irrelevant and redundant features. Applications of the feature selection range from text processing, face recognition, bioinformatics, speaker verification, and medical diagnosis to financial domains. In this study, we focus on filter methods based on information entropy : IG (Information Gain), FCBF (Fast Correlation Based Filter), and mRMR (minimum Redundancy Maximum Relevance). FCBF has the advantage of reducing computational burden by eliminating the redundant features that satisfy the condition of approximate Markov blanket. However, FCBF considers only the relevance between the feature and the class in order to select the best features, thus failing to take into consideration the interaction between features. In this paper, we propose an improved FCBF to overcome this shortcoming. We also perform a comparative study to evaluate the performance of the proposed method.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39A no.5
• /
• pp.237-243
• /
• 2014

In this paper, a calculation-efficient method for weight update in the algorithm based on maximization of the zero-error probability (MZEP) is proposed. This method is to utilize the current slope value in calculation of the next slope value, replacing the block processing that requires a summation operation in a sample time period. The simulation results shows that the proposed method yields the same performance as the original MZEP algorithm while significantly reducing the computational time and complexity with no need for a buffer for error samples. Also the proposed algorithm produces faster convergence speed than the algorithm that is based on the error-entropy minimization.