• Proceedings of the Korea Information Processing Society Conference
• /
• 2010.04a
• /
• pp.405-408
• /
• 2010

This paper presents a new feature extraction technique, correlation coefficient and Manhattan distance (MD) based method for recognition of rotated object in an image. This paper also represented a new concept of intensity invariant. We extracted global features of an image and converts a large size image into a one-dimensional vector called circular feature vector's (CFVs). An especial advantage of the proposed technique is that the extracted features are same even if original image is rotated with rotation angles 1 to 360 or rotated. The proposed technique is based on fuzzy sets and finally we have recognized the object by using histogram matching, correlation coefficient and manhattan distance of the objects. The proposed approach is very easy in implementation and it has implemented in Matlab7 on Windows XP. The experimental results have demonstrated that the proposed approach performs successfully on a variety of small as well as large scale rotated images.

• Proceedings of the IEEK Conference
• /
• 2001.09a
• /
• pp.669-672
• /
• 2001

We propose an efficient algorithm for an interactive content-based image retrieval using relevance feedback. In the proposed algorithm, a new query feature vector first is yielded from the average feature vector of the relevant images that is fed back from the result images of the previous retrieval. Each component weight of a feature vector is computed from an inverse of standard deviation for each component of the relevant images. The updated feature vector of the query and the component weights are used in the iterative retrieval process. In addition, the irrelevant images are excluded from object images in the next iteration to obtain additional performance improvement. In order to evaluate the retrieval performance of the proposed method, we experiment for three image databases, that is, Corel, Vistex, and Ultra databases. We have chosen wavelet moments, BDIP and BVLC, and MFS as features representing the visual content of an image. The experimental results show that the proposed method yields large precision improvement.

• Proceedings of the KIEE Conference
• /
• 1987.07b
• /
• pp.1434-1437
• /
• 1987

We present a very simple algorithm for reducing the encoding (codebook search) complexity of vector quantization (VQ), exploiting some features of a vector currently being encoded. A proposed VQ of 16 (=$4{\times}4$) vector dimension and 256 codewords shows a slight performance degradation of about 0.1-0.9 dB, however, with only 16 or 32 among 256 codeword searches, i.e., with just 1/16 or 1/8 search complexity compared to a full-search VQ. And the proposed VQ scheme is also compared to and shown to be a bit superior to tree-search VQ with regard to their SNR performance and memory requirement.

• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.4
• /
• pp.379-384
• /
• 2009

A human gaze detection and tracing method is importantly required for HMI(Human-Machine-Interface) like a Human-Serving robot. This paper proposed a novel three-dimension (3D) human gaze estimation method by using a face recognition, an orientation estimation and SVMs (Support Vector Machines). 2,400 images with the pan orientation range of $-90^{\circ}{\sim}90^{\circ}$ and tilt range of $-40^{\circ}{\sim}70^{\circ}$ with intervals unit of $10^{\circ}$ were used. A stereo camera was used to obtain the global coordinate of the center point between eyes and Gabor filter banks of horizontal and vertical orientation with 4 scales were used to extract the facial features. The experiment result shows that the error rate of proposed method is much improved than Liddell's.

• International journal of advanced smart convergence
• /
• v.5 no.3
• /
• pp.8-15
• /
• 2016

In recent years, one of deep learning models called Deep Belief Network (DBN) which formed by stacking restricted Boltzman machine in a greedy fashion has beed widely used for classification and recognition. With an ability to extracting features of high-level abstraction and deal with higher dimensional data structure, this model has ouperformed outstanding result on image and speech recognition. In this research, we assess the applicability of deep learning in dna classification level. Since the training phase of DBN is costly expensive, specially if deals with DNA sequence with thousand of variables, we introduce a new encoding method, using decimal-binary vector to represent the sequence as input to the model, thereafter compare with one-hot-vector encoding in two datasets. We evaluated our proposed model with different contrastive algorithms which achieved significant improvement for the training speed with comparable classification result. This result has shown a potential of using decimal-binary vector on DBN for DNA sequence to solve other sequence problem in bioinformatics.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.55 no.7
• /
• pp.306-312
• /
• 2006

Support Vector Machine(SVM), of which the foundations have been developed by Vapnik (1995), is gaining popularity thanks to many attractive features and promising empirical performance. In this paper, we propose a new short-term load forecasting technique based on SVM. We discuss the input vector selection of SVM for load forecasting and analyze the prediction performance for various SVM parameters such as kernel function, cost coefficient C, and $\varepsilon$ (the width of 8 $\varepsilon-tube$). The computer simulation shows that the prediction performance of the proposed method is superior to that of the conventional neural networks.

• Phonetics and Speech Sciences
• /
• v.1 no.4
• /
• pp.127-132
• /
• 2009

The feature vectors which are used in conventional speaker recognition (SR) systems may have many correlations between their neighbors. To improve the performance of the SR, many researchers adopted linear transformation method like principal component analysis (PCA). In general, the linear transformation of the feature vectors is based on concatenated form of the static features and their dynamic features. However, the linear transformation which based on both the static features and their dynamic features is more complex than that based on the static features alone due to the high order of the features. To overcome these problems, we propose an efficient method that applies linear transformation and temporal information of the features to reduce complexity and improve the performance in speaker verification (SV). The proposed method first performs a linear transformation by PCA coefficients. The delta parameters for temporal information are then obtained from the transformed features. The proposed method only requires 1/4 in the size of the covariance matrix compared with adding the static and their dynamic features for PCA coefficients. Also, the delta parameters are extracted from the linearly transformed features after the reduction of dimension in the static features. Compared with the PCA and conventional methods in terms of equal error rate (EER) in SV, the proposed method shows better performance while requiring less storage space and complexity.

• Computers and Concrete
• /
• v.25 no.6
• /
• pp.565-574
• /
• 2020

Over the years, several machine learning approaches have been proposed and utilized to create a prediction model for the high-performance concrete (HPC) slump flow. Despite HPC is a highly complex material, predicting its pattern is a rather ambitious process. Hence, choosing and applying the correct method remain a crucial task. Like some other problems, prediction of HPC slump flow suffers from abnormal attributes which might both have an influence on prediction accuracy and increases variance. In recent years, different studies are proposed to optimize the prediction accuracy for HPC slump flow. However, more state-of-the-art regression algorithms can be implemented to create a better model. This study focuses on several methods with different mathematical backgrounds to get the best possible results. Four well-known algorithms Support Vector Regression, M5P Trees, Random Forest, and MLPReg are implemented with optimum parameters as base learners. Also, redundant features are examined to better understand both how ingredients influence on prediction models and whether possible to achieve acceptable results with a few components. Based on the findings, the MLPReg algorithm with optimum parameters gives better results than others in terms of commonly used statistical error evaluation metrics. Besides, chosen algorithms can give rather accurate results using just a few attributes of a slump flow dataset.

• Asian Pacific Journal of Cancer Prevention
• /
• v.14 no.1
• /
• pp.97-102
• /
• 2013

Objective: The aim of this study was to develop an automated computer-aided diagnostic system for diagnosis of thyroid cancer pattern in fine needle aspiration cytology (FNAC) microscopic images with high degree of sensitivity and specificity using statistical texture features and a Support Vector Machine classifier (SVM). Materials and Methods: A training set of 40 benign and 40 malignant FNAC images and a testing set of 10 benign and 20 malignant FNAC images were used to perform the diagnosis of thyroid cancer. Initially, segmentation of region of interest (ROI) was performed by region-based morphology segmentation. The developed diagnostic system utilized statistical texture features derived from the segmented images using a Gabor filter bank at various wavelengths and angles. Finally, the SVM was used as a machine learning algorithm to identify benign and malignant states of thyroid nodules. Results: The SVMachieved a diagnostic accuracy of 96.7% with sensitivity and specificity of 95% and 100%, respectively, at a wavelength of 4 and an angle of 45. Conclusion: The results show that the diagnosis of thyroid cancer in FNAC images can be effectively performed using statistical texture information derived with Gabor filters in association with an SVM.

• The Journal of the Acoustical Society of Korea
• /
• v.10 no.1
• /
• pp.52-60
• /
• 1991

This paper describes the study on isolated word recognition by using DHMM(Dynamic Hidden Markov Model) which has dynamic feature of spectrum as a parameter. This paper discusses speech recognition experiment basedon HMM which can evaluate not only instantaneous spectral features but also dynamic spectral features. LPC cepstrum parameters is used as a static feature and LPC cepstrum's regression coefficient is used as a dynamic feature. These two features are quantized by each VQ codebook. DHMM is modeled by receiving static vector and dynamic vector by input. In the whole experiment, as recognition experiment using DHMM shows 92.7% of recognition rate while the experiment using conventional HMM shows 88.8% of recognition rate, DHMM proved to be a useful model.