Open Access. Powered by Scholars. Published by Universities.®
Articles 1 - 10 of 10
Full-Text Articles in Engineering
Application Of Shunting Inhibitory Artificial Neural Networks To Medical Diagnosis, Ganesh Arulampalam, Abdesselam Bouzerdoum
Application Of Shunting Inhibitory Artificial Neural Networks To Medical Diagnosis, Ganesh Arulampalam, Abdesselam Bouzerdoum
Research outputs pre 2011
Shunting inhibitory artificial neural networks (SIANNs) are biologically inspired networks in which the neurons interact among each other via a nonlinear mechanism called shunting inhibition. Since they are high-order networks, SIANNs are capable of producing complex, nonlinear decision boundaries. In this article, feedforward SIANNs are applied to several medical diagnosis problems and the results are compared with those obtained using multilayer perceptrons (MLPs). First, the structure of feedforward SIANNs is presented. Then, these networks are applied to some standard medical classification problems, namely the Pima Indians diabetes and Wisconsin breast cancer classification problems. The SIANN performance compares favourably with that …
Knowledge-Based Genetic Algorithm For Layer Assignment, Maolin Tang, Kamran Eshraghian, Daryoush Habibi
Knowledge-Based Genetic Algorithm For Layer Assignment, Maolin Tang, Kamran Eshraghian, Daryoush Habibi
Research outputs pre 2011
Layer assignment is an important post-layout optimization technique in very large scale integrated circuit (VLSI) layout automation. It re-assigns wire segments in a routing solution to appropriate layers to achieve certain optimization objectives. The paper focuses on investigating the layer assignment problem with application to via minimization, which is known to be NP-complete. A knowledge based genetic algorithm for the layer assignment problem is proposed, with the aim of utilizing domain specific knowledge to speed up the process of evolution and to improve the quality of solutions. Experimental results show that this knowledge based genetic algorithm can consistently produce the …
A Wide Dynamic Range Cmos Imager With Extended Shunting Inhibition Image Processing Capabilities, Farid Boussaid, Amine Bermak, Abdesselam Bouzerdoum
A Wide Dynamic Range Cmos Imager With Extended Shunting Inhibition Image Processing Capabilities, Farid Boussaid, Amine Bermak, Abdesselam Bouzerdoum
Research outputs pre 2011
A CMOS imager based on a novel mixed-mode VLSI implementation of biologically inspired shunting inhibition vision models is presented. It can achieve a wide range of image processing tasks such as image enhancement or edge detection via a programmable shunting inhibition processor. Its most important feature is a gain control mechanism allowing local and global adaptation to the mean input light intensity. This feature is shown to be very suitable for wide dynamic range imagers
Skin Color Detection For Face Localization In Human-Machine Communications, Douglas Chai, Son Lam Phung, Abdesselam Bouzerdoum
Skin Color Detection For Face Localization In Human-Machine Communications, Douglas Chai, Son Lam Phung, Abdesselam Bouzerdoum
Research outputs pre 2011
This paper presents the proposed user interface design for computers whereby users can navigate in a 3D graphics scene and change camera viewpoint via head movement. This human-machine communication relies very much on the performance of its face localization module, which must determine head pose and track head movement. We have employed the skin color detection approach to face localization. The approach is studied and presented. The experimental results show that our chosen methodology is very effective. Furthermore, we demonstrate that skin color detection approach can cope with the variations of skin color and lighting conditions
An Intelligent Imaging Approach To The Identification Of Forensic Ballistics Specimens, Clifton L. Smith
An Intelligent Imaging Approach To The Identification Of Forensic Ballistics Specimens, Clifton L. Smith
Research outputs pre 2011
The characteristic markings on the cartridge and projectile of a bullet fired from a gun can be recognised as a "fingerprint" for identification of the firearm. Over thirty different features can be distinguished. The intelligent imaging of the class characteristics of ballistics specimens will provide the potential to identify the make and model of the firearm. Precise measurement of features allow ballistics metrics to be obtained for the identification of the weapon. This paper will describe progress in the development of a multidimensional cluster analysis model for forensic ballistics specimens. The cluster analysis will provide classification that is based on …
Classification Of Bandlimited Fsk4 And Fsk8 Signals, Visalakshi Ramakonar, Daryoush Habibi, Abdesselam Bouzerdoum
Classification Of Bandlimited Fsk4 And Fsk8 Signals, Visalakshi Ramakonar, Daryoush Habibi, Abdesselam Bouzerdoum
Research outputs pre 2011
This paper compares two types of classifiers applied to bandlimited FSK4 and FSK8 signals. The first classifier employs the decision-theoretic approach and the second classifier is a neural network structure. Key features are extracted using a zero crossing sampler. A novel decision tree is proposed and optimum threshold values are found for the decision theoretic approach. For the neural network, the optimum structure is found to be the smallest structure to give 100% overall success rate. The performance of the both classifiers has been evaluated by simulating bandlimited FSK4 and FSK8 signals corrupted by Gaussian noise. It is shown that …
Skin Colour Based Face Detection, Son Lam Phung, Douglas K. Chai, Abdesselam Bouzerdoum
Skin Colour Based Face Detection, Son Lam Phung, Douglas K. Chai, Abdesselam Bouzerdoum
Research outputs pre 2011
This paper describes a new approach to face detection. A colour input image is first processed using neural networks to detect skin regions in the image. Each neural network separates skin and non-skin pixels on the basis of chrominance information. The skin-colour classifier employs the committee machine technique, which improves skin colour detection by combining the classification results of a set of multilayer perceptrons (MLPs). The skin colour classifier achieves a classification rate of 84% compared to 81% for the best individual MLP classifier. The output of the committee machine is processed by a 2D smoothing filter before being converted …
Image Compression Using A Stochastic Competitive Learning Algorithm (Scola), Abdesselam Bouzerdoum
Image Compression Using A Stochastic Competitive Learning Algorithm (Scola), Abdesselam Bouzerdoum
Research outputs pre 2011
We introduce a new stochastic competitive learning algorithm (SCoLA) and apply it to vector quantization for image compression. In competitive learning, the training process involves presenting, simultaneously, an input vector to each of the competing neurons, which then compare the input vector to their own weight vectors and one of them is declared the winner based on some deterministic distortion measure. Here a stochastic criterion is used for selecting the winning neuron, whose weights are then updated to become more like the input vector. The performance of the new algorithm is compared to that of frequency-sensitive competitive learning (FSCL); it …
A Compact Multi-Chip-Module Implementation Of A Multi-Precision Neural Network Classifier, Amine Bermak, Dominique Martinez
A Compact Multi-Chip-Module Implementation Of A Multi-Precision Neural Network Classifier, Amine Bermak, Dominique Martinez
Research outputs pre 2011
This paper describes a novel MCM digital implementation of a reconfigurable multi-precision neural network classifier. The design is based on a scalable systolic architecture with a user defined topology and arithmetic precision of the neural network. Indeed, the MCM integrates 64/32/16 neurons with a corresponding accuracy of 4/8/16-bits. A prototype has been designed and successfully tested in CMOS 0.7 μm technology
Novel Image Enhancement Technique Using Shunting Inhibitory Cellular Neural Networks, Tarik Hammadou, Abdesselam Bouzerdoum
Novel Image Enhancement Technique Using Shunting Inhibitory Cellular Neural Networks, Tarik Hammadou, Abdesselam Bouzerdoum
Research outputs pre 2011
This paper describes a method for improving image quality in a color CMOS image sensor. The technique simultaneously acts to compress the dynamic range, reorganize the signal to improve visibility, suppress noise, identify local features, achieve color constancy, and lightness rendition. An efficient hardware architecture and a rigorous analysis of the different modules are presented to achieve high quality CMOS digital cameras