Dorra Sellami Masmoudi

New electronic cane for visually impaired people for obstacle detection and recognition

In this paper, we describe a new electronic white cane for visually impaired people. Our device may enable blind individuals to see the world with their ears. In fact, we used different types of sensors to detect and recognize obstacles. Our choice of sensors is based not only on system requirement but also on technology cost. Accordingly, we used two ultrasonic sensors and one monocular camera. With such low-cost measurement units, we rely on fusing data obtained by our complementary sensors in further steps. An alert speech message is then sent to the user making him know about the presence and nature of potential encountered obstacle by means of a Bluetooth module.

Hardware implementation of pulse mode RBF neural network-based image denoising

In this paper, we propose a very compact implementation of a pulse mode radial basis activation function (RBF). The main idea is to make use of the powerful means of RBF neural networks in function approximation and implement a reconfigurable hardware, achieving different image processing tasks with on-chip learning. The proposed network is applied here as illustration in image denoising, which is a very important step in image processing. The efficiency of the proposed pulse mode RBF neural network in image denoising versus other conventional filtering techniques is demonstrated. Moreover, it has been shown that the best strategy, leading to better learning generalisation performances, is to apply in the learning steps all kinds of noise in a random way. In such a way, generalisation is better with respect to each kind of noise. The corresponding design was implemented on a Virtex II PRO FPGA platform and synthesis results are presented.

A radio frequency CMOS current controlled oscillator based on a new low parasitic resistance CCII

In this paper we present a design of a current controlled oscillator in 0.35pm CMOS process. Owing to their high degree of controllability, the translinear second generation current conveyer is used as a basic block for our oscillator. Thus, the first step in our design was to improve static and dynamic behaviour of second generation current conveyors. The translinear implementation in CMOS technology was first studied. Then a considerable improvement of the parasitic series resistance on port X is done by proposing a new structure. A reduction of RX by nearly a factor of 10 is observed leading to a notable improvement of the frequency behaviour of the proposed oscillator. As an application example, a current controlled oscillator covering [IOOMHz600MHzI frequency range is proposed. Pspice simulation results are performed using CMOS 0.35 pm process of AMs.

LBPV descriptors-based automatic ACR/BIRADS classification approach

Mammogram tissue density has been found to be a strong indicator for breast cancer risk. Efforts in computer vision of breast parenchymal pattern have been made in order to improve the diagnostic accuracy by radiologists. Motivated by recent results in mammogram tissue density classification, a novel methodology for automatic American College of Radiology Breast Imaging Reporting and Data System classification using local binary pattern variance descriptor is presented in this article. The proposed approach characterizes the local density in different types of breast tissue patterns information into the LBP histogram. The performance of macro-calcification detection methods is developed using FARABI database. Performance results are given in terms of receiver operating characteristic. The area under curve of the corresponding approach has been found to be 79%.

A new human identification based on fusion fingerprints and faces biometrics using LBP and GWN descriptors

Single modality biometric recognition system is often not able to meet the desired system performance requirements. Several studies have shown that multimodal biometric identification systems improve the recognition accuracy and allow performances that are required for many security applications. In this paper, we have developed a multimodal biometric recognition system which combines two modalities: face and fingerprint. For face trait, we build features based on Gabor Wavelet Networks (GWNs), while Local Binary Patterns (LBP) is used for fingerprint trait. Experimental results affirm that a weighted sum based fusion achieves excellent recognition performances, which out performs both single biometric systems.

Neural network based edge detection with pulse mode operations and floating point format precision

This paper proposes a pulse mode neural network based edge detection system. Edge detection of an image reduces significantly the amount of data and filters out information that may be regarded as less irrelevant. Edge detection is efficient in medical imaging. Known by their significant compactness pulse mode neural networks are becoming an attractive solution for function approximation based on frequency modulation. Early pulse mode implementation suffers from some network constraints due to weight range limitations. To provide the best edge detection, the backpropagation algorithm is modified to have pulse mode operations for effective hardware implementation. Here, we undergo these limitations with a new pulse mode network architecture using floating point operations in the activation function. By using floating point number system for synapse weight value representation, any function can be approximated by the network. The proposed pulse mode MNN is used to detect the edges in images forming a heterogenous data base. It shows good learning capability. The corresponding design was implemented into a virtex II PRO XC2VP7 Xilinx FPGA.

Hand vein recognition system with circular difference and statistical directional patterns based on an artificial neural network

In this article, a novel hand vein pattern recognition process for human identification is presented. Hand vein characteristics can be considered as more reliable in biometric domain compared with other biometric characteristics, such as palmprint and fingerprint, because veins are located in volume, making features more robust to test conditions. In this paper, a rotation invariant texture descriptor called Circular Difference and Statistical Directional Patterns (CDSDP) is proposed to extract hand vein patterns. Its histogram is considered as attribute vector. The CDSDP is a surrounding circular difference with weights incorporating the statistical directional information of vessels. Experimental results show that the proposed descriptor based on CDSDP has better performance than the previous descriptors used in local binary patterns (LBP). The proposed method gives an Identification Rate (IR) of 99.8 % and an Error Equal Rate (EER) of 0.01 %. Furthermore, the average processing time of the proposed method is 5.2ms for one hand vein posture, which satisfies the criterion of a real time hand vein recognition system.

A New High Frequency Second Generation Current Conveyor Based Chaos Generator

In this paper, a new RLC low voltage chaos generator using second-generation current conveyor (CCII) as active building block is presented. The applied CCII was optimized with respect to static and dynamic performances thanks to an optimizing heuristic. The resultant high performances CCII has 1.406 GHz GHz and 1.53 GHz as current and voltage bandwidth respectively, and 180Omega as Rx input port resistance value. PSPICE simulations are presented to depict these results.

An optimized low voltage and High Frequency CCII based multifunction Filters

In this paper, a low voltage current conveyor (CCII) based multifunction filter is presented. Firstly, thanks to an optimizing heuristic, an optimal sizing of a low voltage low power CMOS current conveyor (CCII) was done. Hence, we improve static and dynamic performances of this configuration. The optimized CCII configuration has a current bandwidth of 1.103 GHz and a voltage bandwidth of 1.18 GHz and 33.4 Omega as RX parasitic resistance value. Secondly, implementation of a multifunction filter based on this configuration was done. The current mode Alter has a tunable central frequency in the range [50MHz-800MHz]. PSPICE simulations are presented to demonstrate these results.

A non linear stretching image enhancement technique for microcalcification detection

Microcalcifications are small deposits of calcium accumulated in breast tissue. They are situated in higher gray level regions, with a very short gray level dynamic range. Besides, they have a small size with low contrast compared to the surrounding tissue. All these characteristics make micro-calcifications preprocessing a crucial task. In this paper, we propose a new approach for enhancing microcalcifications in digitized mammogramdigitized mammogram, emphasizing corresponding gray level details. Accordingly, we propose an adaptive exponential function which creates a local modification of gray level to highlight details which are potential carriers of microcalcifications. We have applied the NLS twice: locally and globally. The performance of microcalcifications enhancement method is developing using Farabi Digital Database of Screening Mammography (FDDSM). Performance results are given in terms of the Seconde-Derivative-like Measure of enhancement (SDME). Our proposed approach achieve 118 of the local NLS and 115 of the Global NLS.