Osamu Saotome

Hardware implementation of type-2 programmable fuzzifier

The design of full programmable type-2 membership function circuit is presented in this paper. This circuit is used to implement the fuzzifier block of Type-2 Fuzzy Logic Controller chip. In this paper the type-2 fuzzy set was obtained by blurring the width of the type-1 fuzzy set. This circuit allows programming the height and the shape of the membership function. It operates in current mode, with supply voltage of 3.3V. The simulation results of interval type-2 membership function circuit have been done in CMOS 0.35µm technology using Mentor Graphics software.

A ultra low power CMOS pA/V transconductor and its application to wavelet filters

Two topologies of ultra low-power CMOS triode transconductor are proposed. Its input transistors are kept in the triode region to benefit from the lowest gm/ID ratio. The gm is adjusted by a well defined (W/L) and VDS, the latter a replica of the tuning voltage VTUNE. Transconductance down to hundreds of pA/V are obtained and used to implement a 6th order wavelet filter. The resulting design complies with a 1.5V supply and a 0.35µm CMOS process. The total power consumption of the wavelet filters using the first and second topology of transconductor equals 51nW and 114nW, respectively.

Comparing Source Codes Generated by CASE Tools with Hand Coded

This paper presents a comparison between C and C+ + source code automatically generated by CASE tools, and those hand made. In order to compare these two different kinds of implementation, some software static metrics were chosen, due to its soundness and confidence when applied to source code evaluation. A critical real time embedded software for a student satellite onboard computer was considered. Finally, this work points to some preliminary differences between those two implementations, related to software quality applied to an aerospace software prototype.

A comparison between automated generated code tools using model based development

The use of automated generated code tools has been increasing in the last years mainly because it helps engineers to faster and better develop documented software in comparison with hand coded development. Nowadays, there are many tools available from different vendors. However, the most used tools for critical environments and real-time applications are the Rational Rose Real Time (RRRT) and the Rational Rhapsody. Recently, National Instruments, the LabVIEW software vendor, has released a toolkit to develop C code from its visual programming language. This represented a breakthrough for LabVIEW developers, allowing verification of what is being properly developed. Mathworks, the Matlab software vendor, has provided a similar toolkit to the LabVIEW. A case study comparing source codes generated by CASE Tools with hand coded was presented at the ITNG 2008 Conference. In this case study, authors have concluded that hand-made source code is less complex than CASE tools source codes. However, the cost/benefit becomes better when using CASE tools. The main purpose of this paper is to provide an insight about automated generated code using Model Based Development (MBD) tools (LabVIEW, Matlab, and Rational Rose Real Time). The case study presents the modeling of a 100 points sine wave application. A comparison between code metrics is performed, in order to verify which tool best fits into your project.

A bit scalable architecture for fuzzy processors

Several hardware architectures to implement fuzzy processors have been proposed to satisfy real-time requirements, but very few of these are suitable for automatic synthesis. This paper presents bit scalable architecture that allows the automatic synthesis of fuzzy processors in different bit wide resolution. The synthesis is made from a VHDL description. The size of the internal units is defined from a small number of parameters in the highest level entity.

On improving sub-pixel accuracy by means of B-spline

Local perturbations nearby contours strongly perturb the final result of processing remotely sensed images (RSI). It is common to establish a priori data to aid the estimation process. One can move some steps forward by means of a deformable model, for example, the snake model. In up to date research, the deformable contour is represented via B-spline snakes, which allows local control, concise depiction, and the use of fewer parameters. The estimation of edges with sub-pixel accuracy via a global B-spline depiction depends on determining the edge according to a Maximum Likelihood (ML) agenda and using the observed information likelihood. This practice guarantees that outliers present in data will be cleaned out. The data likelihood is calculated as a result of the observation model comprising both orientation and position data. Experiments where this procedure and the traditional spline interpolation have revealed that the algorithm introduced outperforms the conventional method for Gaussian as well as Salt and Pepper noise.

Vehicle inductive signatures recognition using a Madaline neural network

In this paper, we report results obtained with a Madaline neural network trained to classify inductive signatures of two vehicles classes: trucks with one rear axle and trucks with double rear axle. In order to train the Madaline, the inductive signatures were pre-processed and both classes, named C2 and C3, were subdivided into four subclasses. Thus, the initial classification task was split into four smaller tasks (theoretically) easier to be performed. The heuristic adopted in the training attempts to minimize the effects of the input space non-linearity on the classifier performance by uncoupling the learning of the classes and, for this, we induce output Adalines to specialize in learning one of the classes. The percentages of correct classifications presented concern patterns which were not submitted to the neural network in the training process, and, therefore, they indicate the neural network generalization ability. The results are good and stimulate the maintenance of this research on the use of Madaline networks in vehicle classification tasks using not linearly separable inductive signatures.

Fuzzy-Based Automatic Landmark Recognition in Aerial Images Using ORB for Aerial Auto-localization

Aerial navigation based on computer vision is a subject in constant development. It aims to identify the localization of an Unmanned Aerial Vehicle based on aerial images captured during flight. This paper employs a fuzzy-based application to identify landmarks, using the ORB algorithm, which uses descriptors for the neighborhood of keypoints to identify specific registered objects on a scene. In Addition to the keypoint matching from ORB, a fuzzy system is used to analyze each match, in order to guarantee the proper identification of the landmark.

Emergency Response Cyber-Physical System for Disaster Prevention with Sustainable Electronics

An Emergency Response (ER) system can mitigate the risks of natural disasters in underprivileged regions. This article focuses on the instrumentation level of the sensor networks from an ER system using sustainable electronics (salvaged electronic components) and reengineering to assist distant to reach and economically deficient locations.

Total Variation Applications in Computer Vision

The objectives of this chapter are: (i) to introduce a concise overview of regularization; (ii) to define and to explain the role of a particular type of regularization called total variation norm (TV-norm) in computer vision tasks; (iii) to set up a brief discussion on the mathematical background of TV methods; and (iv) to establish a relationship between models and a few existing methods to solve problems cast as TV-norm. For the most part, image-processing algorithms blur the edges of the estimated images, however TV regularization preserves the edges with no prior information on the observed and the original images. The regularization scalar parameter {\lambda} controls the amount of regularization allowed and it is an essential to obtain a high-quality regularized output. A wide-ranging review of several ways to put into practice TV regularization as well as its advantages and limitations are discussed.