Sirindhorn International Institute of Technology

About: Sirindhorn International Institute of Technology is a based out in . It is known for research contribution in the topics: Supply chain & Combustion. The organization has 1048 authors who have published 1678 publications receiving 30067 citations.

Prediction of Raw Milk Microbial Quality Using Data Mining Techniques

Abstract: Data mining techniques were applied to predict raw milk quality in terms of methylene blue reduction time (MBRT) from the independent parameters of raw milk inspection parameters such as travel time, temperature of milk, solid-not-fat, %fat, acidity and specific gravity. Predictive models were developed and the performance of 3 data mining algorithms namely; Multiple Linear Regression (MLR), Artificial Neural Network (ANN) and K-Nearest neighbor (KNN), was measured in terms of average error and Root Mean Square Error (RMSE). MLR showed high and inconsistent RMS error in 3 randomly picked data partitions whereas KNN and ANN were able to predict the MBRT values from the physico-chemical quality parameters, KNN was the preferred algorithm (K=7, RMSE of 1.7). The models were applied to a new set of data (n=78) without showing them the output parameter (MBRT). The predicted values of MBRT were plotted against the actual observed values to classify milk into 4 quality grades.

FIR filters with punctured radix-8 symmetric coefficients: Design and multiplier-free realizations

Abstract: In recent years, several approaches have been investigated to obtain multiplier-free realizations of digital filters. One approach makes use of periodically time-varying (PTV) structures. The idea is to distribute the computation in time and in space. Distribution in time provides reuse of the same hardware by means of PTV coefficients. Distribution in space increases the number of coefficients but simplifies the values of the coefficients. Computation distribution is based on radix-r number representation, and it can be carried out to the extent that each computation involves a simple coefficient that can be realized using only addition and shift (no hardware multiplier). Previous PTV realizations could not exploit the coefficient symmetry of finite impulse response (FIR) filters to reduce the number of coefficients. This paper proposes design and realizations of FIR filters with punctured radix-8 coefficients belonging to the septuple set {0, ±1, ±2, ±4), which can be implemented using only a shift operation without requiring any hardware multiplier. The realizations exploit the coefficient symmetry to reduce the hardware by about one-half. Due to a non-uniform grid of representation, we apply a modified Karmarkar's linear programming algorithm to find the optimum set of discrete coefficients that minimizes the weighted peak ripple error. Comparison with a conventional FIR filter with sum-of-powers-of-two (SOPOT) coefficients shows that the proposed filter is faster and uses less hardware than one with SOPOT coefficients. However, the punctured radix-8 system has a limit on the achievable ripple.

A false alarm reduction method for a gas sensor based electronic nose.

Abstract: Electronic noses (E-Noses) are becoming popular for food and fruit quality assessment due to their robustness and repeated usability without fatigue, unlike human experts. An E-Nose equipped with classification algorithms and having open ended classification boundaries such as the k-nearest neighbor (k-NN), support vector machine (SVM), and multilayer perceptron neural network (MLPNN), are found to suffer from false classification errors of irrelevant odor data. To reduce false classification and misclassification errors, and to improve correct rejection performance; algorithms with a hyperspheric boundary, such as a radial basis function neural network (RBFNN) and generalized regression neural network (GRNN) with a Gaussian activation function in the hidden layer should be used. The simulation results presented in this paper show that GRNN has more correct classification efficiency and false alarm reduction capability compared to RBFNN. As the design of a GRNN and RBFNN is complex and expensive due to large numbers of neuron requirements, a simple hyperspheric classification method based on minimum, maximum, and mean (MMM) values of each class of the training dataset was presented. The MMM algorithm was simple and found to be fast and efficient in correctly classifying data of training classes, and correctly rejecting data of extraneous odors, and thereby reduced false alarms.

Actuation of Pneumatic Artificial Muscle via Hydrogen Absorption/Desorption of Metal Hydride-LaNi5:

Abstract: This paper presents experimental studies on mechanical actuations of a pneumatic artificial muscle (PAM), which is driven by hydrogen gas based metal hydride (MH). The dynamic performances of hydrogen absorption/desorption, taking place within a MH reactor, are controlled via implementing cooling/heating effects of a thermoelectric module (TEM). Hydrogen pressure is applied as a driving force to commanding work outputs of the PAM as desired mechanical actuations. Due to strong inherent nonlinearity, a conventional proportional integral derivative (PID) control law is not capable of regulating thermodynamic variables of the HM reaction according to desired performances of the PAM. In this study, the fuzzy adaptive PID control is proposed in manipulating the MH reaction via the TEM. This viability of the proposed methodology is confirmed by the fact that the gains of PID control law are adapted by fuzzy rule-based tuning scheme at various operating conditions of the MH reactor. The experimental results show...

Video shot classification using 2D motion histogram

Abstract: This paper describes a new histogram based approach for classifying camera motions or characterizing video shots. The proposed method utilizes both magnitudes and orientations of motion vectors simultaneously, rather than using them separately as same as the existing methods. A 2D histogram, namely 2D array motion vector histogram, that carries both magnitudes and orientations of motion vectors detected in video sequences. We have compared the proposed approach with existing methods. Experimental results show that the proposed method can classify more camera motions with better performance.