Pinit Kumhom

Power harvester design for semi-passive UHF RFID Tag using a tunable impedance transformation

This paper proposes a design of a power harvester for a semi-passive RFID Tag. A tunable impedance transformation circuit is inserted between the antenna and the rectifier unit to maximize the power delivered to the rectifier of a tag and minimize reflections from the antenna input port. A power harvester for the UHF band of 950 MHz has been designed based on a 0.35 µm CMOS technology. This impedance matching is designed to maximize efficiency by characterizing both the tunable impedance transformation circuit and the conventional rectifier circuit. The antenna is modeled as an RF source with series impedance ZS of 50Ω while the rectifier is modeled an impedance ZL, which is approximated from the process and design parameters of the NMOS transistors used in the conventional NMOS rectifier. These two impedances are used to find optimal matching parameters of the chosen lowpass LL matching circuit. The simulation results show that the required voltage for a 1.2V secondary battery can be achieved with the tunable impedance matching circuit and conventional 2-stage rectifier. The DC output of the rectifier yields the maximum DC voltage of 1.77V, the maximum average current of 4.72µA, and the power conversion efficiency of 15.85% for the input RF power of −14.47dBm at the battery voltage of 1.2V.

Hand posture classification using wavelet moment invariant

In this paper, we present a wavelet moment invariants for classification a small change in rotation and subtle difference of hand posture causes misclassifying to other postures. The method combined zernike moment to capture global features and wavelet moment to differentiate between subtle variations in description can be utilized at the same time. Then, a fuzzy classification algorithm is used to classify hand posture. The classification rate obtained is 72% with of Thai sign language.

Nondestructive maturity determination of durian by using microwave moisture sensing

Since durian has specific properties, such as the thickness of peel, non-uniform shape, rough skin, etc., the nondestructive determination of the maturity is hard to perform. Microwaves are high frequency waves which have better penetration than other lower frequency waves. The paper proposes a simple and inexpensive microwave technique by using an optimal open-ended coaxial probe which operates at 3 GHz which matches the design of a coaxial probe and durian. In this experiment, 3GHz of microwave with low power gets the best response output. The system operates with a network analyzer whose the magnitude of reflection coefficient shows linear correlation to moisture content of durian. The result of % moisture content is compared to % dry-weight of durian in an oven. This nondestructive technique shows great promise for maturity determination of durian.

Visual Hand Gesture Interface for Computer Board Game Control

In computer vision interface for interactive computer game, gestures can be used as commands for the game control instead of keyboard or mouse. The utilization of hand gesture in human-computer interaction enables human operators to interact with computer environments in a natural and intuitive manner. This paper proposes visual gesture recognition for manipulating objects in a board game such as chess. The distance between a players thumb and index finger is used to determine whether the player is in the state of holding a playing object or not. A hand holding an object is tracked until it is stopped and put down the object by widening the distance between the thumb and the finger. The thumb-index distance is measured using the distance between two consecutive peaks of hand contour. The Kalman-filter-based prediction is applied for tracking the finger. The results show high accuracy in objecting picking, hand tracking, and object placing

An Efficient Implementation of the Nearest Neighbor Based Visual Objects Tracking

An independent visual objects tracking is less reliable than the data association of visual objects tracking. This paper describes a tracking method based on the nearest neighbor (NN) data association, which serves lower computational than do the multiple hypothesis tracking (MHT) or the joint probabilistic data association filter (JPDAF) but gives low reliability, if the number of targets is increased. This reliability can be increased by selecting appropriate visual object model. To obtain low computation while capable of handling non-rigid object, we propose an object model which combines the threshold of accumulated object region and the object bounding box. The elements of the association matrix are the distance function that is proposed as a mixture of object models of distance function. The combinations of object models of distance function are important mechanism for determining appropriate state of object correspondence which can be divided into six groups: updated track, missing track, newly track, grouped track, merged track and complex track. The missing track is solved by the track life time criterion while the grouping, the merged and the complex track are resolved by using the proposed NN algorithm again. The experimental results are correctly shown on various situations of correspondence problem from surveillance image sequences

Nondestructive durian maturity determination by using microwave free space measurement

Since durian has specific properties such as big peel, non-uniform shape, rough skin, etc. the nondestructive maturity determination is hard to perform. This paper proposes microwave free-space measurement by using 2 waveguide antennas as transmitter and receiver operating in frequency range of 2-3 GHz to determine the insertion loss (S21 ) after passing through durian which can be calculated in term of complex permittivity, the results are compared with percentage dry-weight as an standard durian maturity determination. The results show that this technique can provide fast durian maturity determination with 90% accuracy

Energy and peak load forecast models using neural network for fast developing area

In electric power systems, energy and peak load forecasting plays an important role in term of policy planning and budget allocation. An accurate forecast method can be helpful in developing a power supply strategy, and development plan, especially for developing countries where the demand is increased with dynamic and high growth rate. This paper proposes energy and peak load forecast models using artificial neural networks (ANN) with multilayer perceptron (MLP) and backpropagation (BP) learning algorithm. This method can forecast the monthly energy and peak load in a year from one month to two years. Factors reflecting weather and economic variation are selected based on correlation coefficients. In addition, a new training technique is introduced. A case study is performed on energy and peak load historical data of Vientiane Municipality system. It demonstrates the accuracy of the proposed method and shows that the forecast model is simple with high accuracy.

A digital image watermarking technique using prediction method and Weber ratio

Most violations of digital image copyrights are the cropping of some parts of the image such as human face for applying to other images without permission. Seriously cropped watermarked images which have various shapes, sizes and locations make it difficult to verify the copyright. This paper proposes a digital image watermarking technique in the spatial domain using prediction method and the Weber ratio. Pixels for embedding the watermarks are altered based on the binary watermark image using the Weber ratio to control the strength of the watermark. Our technique does not require the host image to extract the watermarks. An extracted watermark is computed by subtracting the watermarked pixel with its predicted pixel The experimental results showed that our technique yielded high similarity of the watermarks when a watermarked image was attacked by JPEG compression and the cropping attack.

A method of glove tracking for amateur boxing refereeing

Amateur boxing is one of the most exciting and interesting sports. Since referee of amateur boxing detects and counts number of punches that hit the opponents as scores for refereeing, an automatic glove tracking and punch detecting functions can help the judgment system because a machine has no bias. In this paper, we propose the method of glove tracking for amateur boxing refereeing. In this method, color-based segmentation is used to determine white region of glove in the first couple of frames and the glove in the next frames is tracked by opening window according to the movement vector of previous frames. To evaluate performance of proposed method, the experiments with videos of amateur boxer are performed. The results showed a glove could be tracked accurately.

A computer-aided-diagnosis of tonsillitis using tonsil size and color

Tonsillitis disease is a major cause of serious diseases such as heart attack and pneumonia. This paper proposes a computer-aided-diagnosis of tonsillitis using tonsil size and color. In this method, we employ red color and tonsil size as features. In order to detect the abnormal features, the proposed idea considers a trend of changing the features in each day. The change of tonsillitis features in each day would be computed in order to input into a decision model. As the features (color and size) are so difficult to exactly divide into each range, this paper applies a fuzzy logic as a decision unit. To evaluate the performance of the proposed method, the experiments with tonsillitis images are performed. The results reveal about 84 % accuracy for tonsillitis detection