Simulating biological vision with hybrid neural networks

Article Abstract:

We present an example of how vision systems can be modeled and designed by integrating a top-down computationally-based approach with a bottom-up biologically-motivated architecture. The specific visual processing task we address is occlusion-based object segmentation - the discrimination of objects using cues derived from object interposition. We construct a model of object segmentation using hybrid neural networks - distributed parallel systems consisting of neural units modeled at different levels of abstraction. We show that such networks are particularly useful for systems which can be modeled using the combined top-down/bottom-up approach. Our hybrid model is capable of discriminating objects and stratifying them in relative depth. In addition, our system can account for several classes of human perceptual phenomena, such as illusory contours. We conclude that hybrid systems serve as a powerful paradigm for understanding the information processing strategies of biological vision and for constructing artificial vision-based applications. (Reprinted with permission of the publisher.)

Technology, Biology, Perception (Psychology), Vision, Hybrid Circuits, Perception

---

Modeling with Imperfect Data: A Case Study Simulating a Biological System

Article Abstract:

Two of the three main activities involved in a simulation experiment are estimating input data and estimating design parameters. For many applications, this is straight forward; the data exist in controlled environments for easy estimation, or are otherwise available. A problem arises when these data are unavailable. The present paper proposes a statistical procedure for estimating these values and parameters. Estimations are made based on input and output data actually used in the system. An experiment involving pine beetle infestation of pine forests produced varied results. Graphs illustrating simulation results and actual observed data are included.

Agriculture, Statistics (Data), Parameter estimation, Models, Statistics, Case Study, Forestry

---

NEXUS: a simulation environment for large-scale neural systems

Article Abstract:

NEXUS is a novel simulation environment designed for modeling large neural systems. The simulator allows the user to incorporate complex functional properties and symbolic processing at the level of the individual unit. In addition, NEXUS takes advantage of the principle of topographic map organization, found throughout the mammalian nervous system, to facilitate the modeling and design process. An easy-to-use graphical interface allows the user to interactively build and test models. This paper describes the principles underlying the NEXUS design and its advantages over other current simulation approaches. (Reprinted by permission of the publisher.)

Biomedical engineering, Scientific Research

Similar abstracts:

• Abstracts: The potential in using backpropagation neural networks for facial verification systems. A new software technique for sub-model invocation
• Abstracts: The distributed earth model orbiter simulation (DEMOS) three dimensional system at Johnson Space Center. Real-time simulation of the space station Freedom attitude control system
• Abstracts: Implementing TQM in health care requires adaptation and innovation. Strategic leadership for quality in health care
• Abstracts: Category 3: strategic quality planning. Why do health care practitioners resist quality management?
• Abstracts: TQM: the only source of sustainable competitive advantage. Quality Management Benchmark Assessment

This website is not affiliated with document authors or copyright owners. This page is provided for informational purposes only. Unintentional errors are possible.