-
[show abstract]
[hide abstract]
ABSTRACT: This paper investigates the design curves of the class DE amplifier with nonlinear shunt capacitances for any output Q and any grading coefficient m of the diode junction in MOSFET. The design curves are derived by the numerical calculation. It is clarified that the nonlinearities of shunt capacitances affect the design curves of the class DE amplifier, especially, for low output Q. The waveforms from SPICE denote the validity of the design curves.
Circuits and Systems, 2008. ISCAS 2008. IEEE International Symposium on; 06/2008
-
[show abstract]
[hide abstract]
ABSTRACT: This paper presents an interleaved class E<sup>2</sup> dc/dc converter. In the proposed converter, N class E<sup>2</sup> dc/dc converters are connected in parallel at the class E rectifier. The proposed converter is suitable to the low-voltage/high-current applications. Since less current flows through the each converter, the amount of the power losses on the parasitic resistances are also reduced. In addition, the output filter of the rectifier is downsized because of higher frequency of the interleaved current. By carrying out the circuit experiments, we show that the experimental results agree with the numerical ones quantitatively. The laboratory measurement achieves the over 90% power conversion efficiency under 3.2W and 1MHz operation.
Circuits and Systems, 2008. ISCAS 2008. IEEE International Symposium on; 06/2008
-
[show abstract]
[hide abstract]
ABSTRACT: This paper studies a image show-through problem. It happens often when we copy or scan a paper document, in which the image from the back page shows through. The images obtained on both side of the paper can be considered as mixture components, which are nonlinear mixtures of original images. In this study, we propose to use self-organizing map (SOM) and fastICA to implement separation of the image mixtures. SOM is neural network-based technique using unsupervised learning and can provide useful data representations. The separation results show that the two blind separation methods are applicable to the problem.
Information Technology Applications in Biomedicine, 2007. ITAB 2007. 6th International Special Topic Conference on; 12/2007
-
[show abstract]
[hide abstract]
ABSTRACT: The increase in the number of medical exams, the need to store images resulting for those exams at a very high quality, and the introduction of teleradiology result in an overwhelm increase in storage and distribution costs of medical images. In this paper we propose a method using JPEG 2000 in the lossless or near lossless modes to encode medical images, in order to reduce image storage costs and in the same time store a very high image quality needed for medical applications. We use JPEG 2000 interactive protocols (JPIP) to allow interactive access to medical images in a way that minimizes the transmission bit rate by first displaying a reduced resolution or quality of the image, then allowing lossless display of regions of interest interactively specified by the image viewer. Our method secures medical images by applying JPEG 2000 specific security tools and is implemented in a way compliant to DICOM standard to ensure interoperability. Simulations and results verify the efficiency of the proposed method.
Information Technology Applications in Biomedicine, 2007. ITAB 2007. 6th International Special Topic Conference on; 12/2007
-
[show abstract]
[hide abstract]
ABSTRACT: This paper presents a method of continuous-time simple adaptive control (SAC) using multiple neural networks for a single-input single-output (SISO) nonlinear systems with unknown parameters and dynamics, bounded-input bounded- output, and bounded nonlinearities. The control input is given by the sum of the output of the simple adaptive controller and the sum of the outputs of the parallel small-scale neural networks. The parallel small-scale neural networks are used to compensate the nonlinearity of plant dynamics that is not taken into consideration in the usual SAC. The role of the parallel small- scale neural networks is to construct a linearized model by minimizing the output error caused by nonlinearities in the control systems. Finally, the stability analysis of the proposed method is carried out, and the effectiveness of this method is confirmed through computer simulations.
SICE, 2007 Annual Conference; 10/2007
-
[show abstract]
[hide abstract]
ABSTRACT: Sliding mode control (SMC) has a strong capability of controlling nonlinear systems with uncertainties. However, the discontinuous control signal causes the significant problem of chattering. Furthermore, it requires thorough knowledge of the parameters and dynamics of the controlled plant, which are difficult to be obtained or may be unknown, to calculate the equivalent control law of SMC. In this paper, a combination of SMC and neural network (NN) is proposed. The weights of NN are adjusted using a backpropagation algorithm. To construct corrective control law of SMC for overcoming the chattering problem, a new and simple approach using a simplified distance function with a modified sliding surface is utilized. Thus, the chattering is eliminated and the performance of SMC is improved. Finally, a brief stability analysis of the proposed method is carried out, and the effectiveness of this method is confirmed through computer simulations.
SICE, 2007 Annual Conference; 10/2007
-
[show abstract]
[hide abstract]
ABSTRACT: In this paper, a new type of multineural networks filter (MNNF) is presented that is trained for restoration and enhancement of the medical CR images. In medical CR image, noise has been categorized as quantum mottle, which is related to the incident X-ray exposure and artificial noise, which is caused by the grid, etc. MNNF consists of several neural network filters (NNFs). A novel analysis method is proposed to make the characteristics of the trained MNNF clearly. In the proposed method, a characteristics judgement system is presented to decide which NNF will be executed through the estimation of noise intensity calculated by maximum penalized likelihood estimator (MPLE). The new approach was tested on clinical medical X-ray image, synthesized noisy X-ray image and natural image. In all cases, the proposed MNNF produced better results in terms of mean square error (MSE) measure than MPLE, NNF and conventional wavelet BayesShrink (BS) methods.
Natural Computation, 2007. ICNC 2007. Third International Conference on; 09/2007
-
T. Yahagi
Electrical Engineering in Japan 04/2007; 95(3):115 - 121. · 0.09 Impact Factor
-
T. Yahagi
Electrical Engineering in Japan 04/2007; 93(6):116 - 123. · 0.09 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: This paper presents the fuzzy PID (FPID) controller using neural network (NN) for controlling the magnetic levitation system. Magnetic levitation systems are open loop unstable, uncertainly and inherently nonlinear systems. Consequently, controlling this kind of the system is very difficulty. The FPID controller is developed to provide nonlinear or linear control action that can improve performance of the controller in comparison with a conventional PID controller using only linear policy. Unfortunately, since FPID controller are nonlinear, it is more difficult to set the controller gains compared the linear PID controller. In this paper we propose a neural network to assist the FPID controller. The NN is added in parallel with FPID controller. The NN is used to compensate for inadequate FPID parameters and for stabilize the magnetic levitation system. The uniqueness our method is when the parameters of FPID are incorrect, then the NN takes over the controller, otherwise the NN does not operate. Online training and fast computing of the NN has been designed for that purposes. Finally, the experiment results showed the effectiveness of the proposed method
Intelligent Signal Processing and Communications, 2006. ISPACS '06. International Symposium on; 01/2007
-
[show abstract]
[hide abstract]
ABSTRACT: In most block-based video coding systems, the fast block matching algorithms (BMAs) use the origin as the initial search center, which may not track the motion very well. To improve the accuracy of the fast BMAs, a new adaptive motion tracking search algorithm is proposed in this paper. Based on the spatial correlation of motion blocks, a predicted starting search point, which reflects the motion trend of the current block, is adaptively chosen. It does not have the problem of being trapped by local minimum, and is characterized by finding the majority motion vector in one step. When compared with six other block-based search algorithms including the full-search and three-step-search, the new algorithm has an average PSNR very close to that of full search, yet an average search time faster than the three step-search
Intelligent Signal Processing and Communications, 2006. ISPACS '06. International Symposium on; 01/2007
-
[show abstract]
[hide abstract]
ABSTRACT: Motion segmentation is important in many computer vision application, which aims to detect motion regions such as moving vehicles and people in natural scenes. Detecting moving blobs provides a focus of attention for later processes such as tracking and activity analysis. However, changes from weather, illumination, shadow and repetitive motion from clutter make motion segmentation difficult to process quickly and reliably. In this paper we proposed a method using minimum graph cuts based on hybrid algorithm approach for moving object segmentation. Experiments are carried out to examine the efficiency of the proposed approach
Intelligent Signal Processing and Communications, 2006. ISPACS '06. International Symposium on; 01/2007
-
[show abstract]
[hide abstract]
ABSTRACT: When a signal is embedded in an additive Gaussian noise, its estimation is often done by finding a wavelet basis that concentrates the signal energy in few coefficients and then thresholding the noisy coefficients. However, in many practical problems such as medical X-ray image, astronomical and low-light images, the recorded data is not modeled by Gaussian noise but as the realization of a Poisson process. Multiwavelet is a new development to the body of wavelet theory. Multiwavelet simultaneously offers orthogonality, symmetry and short support which are not possible in scalar 2-channel wavelet systems. After reviewing this recently developed theory, a new theory and algorithm for denoising medical X-ray images using multiwavelet multiple resolution analysis (MRA) are presented and investigated in this paper. The proposed covariance shrink (CS) method is used to threshold wavelet coefficients. The form of thresholds is carefully formulated which is the key to more excellent results obtained in the extensive numerical simulations of medical image denoising compared to conventional methods
Intelligent Signal Processing and Communications, 2006. ISPACS '06. International Symposium on; 01/2007
-
[show abstract]
[hide abstract]
ABSTRACT: This paper considers an UWB (ultra-wideband) SC-FDE (single-carrier input-multiple output with frequency domain equalization) system for multiuser. The method exploits higher order statistics (HOS) in the received signals. It is followed by a method which solves permutation problem. The permutation problem is caused by the Fourier transform (FFT) operations. Simulations show that the proposed method performs close to the equalizers with perfect channel knowledge at the receiver
Intelligent Signal Processing and Communications, 2006. ISPACS '06. International Symposium on; 01/2007
-
[show abstract]
[hide abstract]
ABSTRACT: In this paper we propose a progressive encryption and controlled access scheme for JPEG 2000 encoded images. Our scheme applies SNOW 2 stream cipher to JPEG 2000 codestreams in a way that preserves most of the inherent flexibility of JPEG 2000 encoded images and enables untrusted intermediate network transcoders to downstream an encrypted JPEG 2000 image without access to decryption keys. Our scheme can also control access to various image resolutions or quality layers, by granting users different levels of access, using different decryption keys. Our scheme preservers most of the inherent flexibility, scalability, and transcodability of encrypted JPEG 2000 images and also preserves end-to-end security.
Intelligent Signal Processing and Communications, 2006. ISPACS '06. International Symposium on; 01/2007
-
[show abstract]
[hide abstract]
ABSTRACT: Compression artifacts are the results of an aggressive data compression scheme applied to an image that discards some data which is determined by an algorithm to be of lesser importance to the overall content but which is nonetheless discernible and objectionable to the user. In this paper we present a post-processing algorithm that focus on restoring the clarity of edges and removing the compression artifacts around edges at compression ratios of 200:1 and greater. At the decoder, the algorithm extracts the locations of edges from reconstructed image and classifies the pixels surrounding the edges for removing compression ringing artifacts, then applies a linear procedure to restore the clarity of the edges
Intelligent Signal Processing and Communications, 2006. ISPACS '06. International Symposium on; 01/2007
-
[show abstract]
[hide abstract]
ABSTRACT: Sliding mode control (SMC) has a strong capability in controlling nonlinear systems with uncertainties. However, SMC requires thorough knowledge of the controlled plant parameters and dynamics that is difficult to be obtained or may be unknown, which causes difficulties in calculating the equivalent control law of SMC. To overcome this problem, this paper proposes an adaptive SMC using simple adaptive control (SAC) developed for a class of single-input single-output (SISO) nonlinear systems with unknown parameters and dynamics, bounded-input bounded-output (BIBO), and bounded nonlinearity. The role of SAC is to construct an equivalent control input of adaptive SMC. To construct a corrective control input, this paper applies the sign function with a modified sliding surface. Finally, the stability analysis of the proposed method is carried out, and the effectiveness of this method is confirmed through computer simulations
Decision and Control, 2006 45th IEEE Conference on; 01/2007
-
[show abstract]
[hide abstract]
ABSTRACT: This paper presents design procedure and experimental results class DE inverter with band-pass filter. In the proposed inverter, band-pass filter is used instead of resonant filter in the class DE inverter presented at the previous papers. By using band-pass filter, the sensitivity of output voltage to the operating frequency can be suppressed. By carrying out circuit experiments, we show that the experimental results agree with numerical ones quantitatively. The laboratory experiments achieve 90.4% power conversion efficiency under 2.2 W output power and 1 MHz operation
IEEE Industrial Electronics, IECON 2006 - 32nd Annual Conference on; 12/2006
-
[show abstract]
[hide abstract]
ABSTRACT: This paper presents a new free-running-oscillation resonant DC-DC converter with the Class-E oscillator and the Class-E rectifier, along with its design procedure and design curves for any conditions. The proposed converter is a free-running-oscillation system and has no gate-drive circuit to drive the MOSFET. Therefore, the structure of the proposed converter is simpler than that of class E<sup>2</sup> dc-dc converter. The proposed converter is suitable for a constant voltage and constant power operation under the fixed designed condition. The proposed converter has the ability to achieve a high power-conversion efficiency and low noises under high-frequency operation because of Class-E switching at both the oscillator and the rectifier. We investigate operations of the proposed converter by circuit experiments and show the validity of the design curves. This paper denotes two experimental results. One achieves 79.3% power-conversion efficiency under 1.54-W output power at an operating frequency 2.00 MHz. The other does 80.5% under 1.60-W output power at an operating frequency 1.98 MHz
Circuits and Systems I: Regular Papers, IEEE Transactions on 10/2006; · 1.97 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: In this paper, the discrete cosine transform domain (DCT domain) watermarking technique for copyright protection of still digital images is analyze and new watermarking method with dual detection is proposed. The DCT is applied in blocks of 8 times 8 pixels as in the JPEG algorithm. Previous DCT domain watermarking schemes proved to be very resilient against many attacks except for the geometric attack, especially rotation. Thus, these problems were tackled in our research to be overcome by using dual detection, which uses an equivalent form of the spatial version of the watermark embedded in the DCT domain as detection technique. In order to test the robustness of the algorithms, simple attacks such as filtering, JPEG compression, rotation, resizing and cropping were exploited
Industrial Electronics, 2006 IEEE International Symposium on; 08/2006
