Title: Event-based Neuromorphic Systems, A way to Engineer the Neural Systems
عنوان: سيستمهای نرون-ريختی (Neuromorphic) بر اساس پردازش وقايع (Event ها)، تحولی در شبيه سازی سيستم های عصبی
Neuromorphic engineering is a new interdisciplinary field of science involving biologists, physicists, mathematicians, computer scientists, and engineers to design hardware/physical models of neural systems. It aims at designing silicon-based neural systems for computational or biomedical purposes. The term “neuromorphic” relates to the computational architecture, shaped to model biological neural systems, hence it is strongly linked to computational neuroscience.
In this talk we will give a general overview of “Neuromorphic Engineering”, with an emphasis on Vision. Then we will introduce a Vision-Sensor technology called Silicon-Retina or DVS that imitates some retinal functions. Since this new sensor demands a new style of information processing compared with conventional cameras (event-based dynamics vs. static frames), we will propose an asynchronous event-based stereo matching algorithm for Neuromorphic Silicon Retinas, and a spiking-neural structure for this task. The main idea of the proposed algorithm and the network is grounded in the cooperative computing principle, which was first proposed by David Marr in the mid-80s . The classic cooperative approach for stereoscopic fusion operates on static features in two images. The question we address in this research is how to formulate an event-driven cooperative process (vs. static cooperative) to deal with dynamic spatio-temporal features in Silicon Retinas (or in neuronal retinal activities) .
- Marr, “Vision, A computational Investigation into the Human Representation and Processing of Visual Information” edited by Lucia M. Vaina, MIT press, 2010, pp. 111-122 (2010)
- Firouzi, J. Conradt, “Asynchronous Event-based Cooperative Stereo Matching Using Neuromorphic Silicon Retinas”, Neural Processing Letters 43(2), pp 311-326 · April 2016