No description available.

No description available.

No description available.

No description available.

No description available.

This paper presents the theory behind a model for a two-stage analog network for edge detection and image reconstruction to be implemented in VLSI. Edges are detected in the first stage using the multi-scale veto rule, which eliminates candidates that do not pass a threshold test at each of a set of different spatial scales. The image is reconstructed in the second stage from the brightness values adjacent to edge locations. The MSV rule allows good localization and efficient noise removal. Since the reconstructed images are visually similar to the originals, the possibility exists of achieving significant bandwidth compression.

The biological sense organ contains infinite potential. The artificial neural structures have emulated the potential of the central nervous system; however, most of the researchers have been using the linear combination of synaptic operation. In this thesis, this neural structure is referred to as the neural unit with linear synaptic operation (LSO). The objective of the research reported in this thesis is to develop novel neural units with higher-order synaptic operations (HOSO), and to explore their potential applications. The neural units with quadratic synaptic operation (QSO) and cubic synaptic operation (CSO) are developed and reported in this thesis. A comparative analysis is done on the neural units with LSO, QSO, and CSO. It is to be noted that the neural units with lower order synaptic operations are the subsets of the neural units with higher-order synaptic operations. It is found that for much more complex problems the neural units with higher-order synaptic operations are much more efficient than the neural units with lower order synaptic operations. Motivated by the intensity of the biological neural systems, the dynamic nature of the neural structure is proposed and implemented using the neural unit with CSO. The dynamic structure makes the system response relatively insensitive to external disturbances and internal variations in system parameters. With the success of these dynamic structures researchers are inclined to replace the recurrent (feedback) neural networks (NNs) in their present systems with the neural units with CSO. Applications of these novel dynamic neural structures are gaining potential in the areas of image processing for the machine vision and motion controls. One of the machine vision emulations from the biological attribution is edge detection. Edge detection of images is a significant component in the field of computer vision, remote sensing and image analysis. The neural units with HOSO do replicate some of the biological attributes for edge detection. Further more, the developments in robotics are gaining momentum in neural control applications with the introduction of mobile robots, which in turn use the neural units with HOSO; a CCD camera for the vision is implemented, and several photo-sensors are attached on the machine. In summary, it was demonstrated that the neural units with HOSO present the advanced control capability for the mobile robot with neuro-vision and neuro-control systems.

The biological sense organ contains infinite potential. The artificial neural structures have emulated the potential of the central nervous system; however, most of the researchers have been using the linear combination of synaptic operation. In this thesis, this neural structure is referred to as the neural unit with linear synaptic operation (LSO). The objective of the research reported in this thesis is to develop novel neural units with higher-order synaptic operations (HOSO), and to explore their potential applications. The neural units with quadratic synaptic operation (QSO) and cubic synaptic operation (CSO) are developed and reported in this thesis. A comparative analysis is done on the neural units with LSO, QSO, and CSO. It is to be noted that the neural units with lower order synaptic operations are the subsets of the neural units with higher-order synaptic operations. It is found that for much more complex problems the neural units with higher-order synaptic operations are much more efficient than the neural units with lower order synaptic operations. Motivated by the intensity of the biological neural systems, the dynamic nature of the neural structure is proposed and implemented using the neural unit with CSO. The dynamic structure makes the system response relatively insensitive to external disturbances and internal variations in system parameters. With the success of these dynamic structures researchers are inclined to replace the recurrent (feedback) neural networks (NNs) in their present systems with the neural units with CSO. Applications of these novel dynamic neural structures are gaining potential in the areas of image processing for the machine vision and motion controls. One of the machine vision emulations from the biological attribution is edge detection. Edge detection of images is a significant component in the field of computer vision, remote sensing and image analysis. The neural units with HOSO do replicate some of the biological attributes for edge detection. Further more, the developments in robotics are gaining momentum in neural control applications with the introduction of mobile robots, which in turn use the neural units with HOSO; a CCD camera for the vision is implemented, and several photo-sensors are attached on the machine. In summary, it was demonstrated that the neural units with HOSO present the advanced control capability for the mobile robot with neuro-vision and neuro-control systems.

No description available.

Yes / Edge detection is widely believed to be an important early stage in human visual processing. However, there have been relatively few attempts to map human edge detection filters. In this study, observers had to locate a randomly placed step edge in brown noise (the integral of white noise) with a 1/𝑓2 power spectrum. Their responses were modelled by assuming the probability the observer chose an edge location depended on the response of their own edge detection filter to that location. The observer’s edge detection filter was then estimated by maximum likelihood methods. The filters obtained were odd-symmetric and similar to a derivative of Gaussian, with a peak-to-trough width of 0.1–0.15 degrees. These filters are compared with previous estimates of edge detectors in humans, and with neurophysiological receptive fields and theoretical edge detectors.