System for simulating decision-making process in brain of mammal with respect to visually observed movement of body

Abstract

The invention is a system (1) that simulates a decision process in a mammalian brain with respect to a motion characteristic relating to a visually observed body posture of a body by means of a simulated visual path comprising an interface towards a simulated neuronal structure, the system includes an interface that converts at least luminescence information of the observed body into an optical flow data stream that delivers information relating to the visually observed body and that can be processed in the simulated neuron structure, the system being a feed-forward system that can be coupled to the simulated neuron structure. And from the visual observation that the decision comprises hierarchically: the simulated visual path and its interface (3, 3L, 3R); a simulated local motion direction detection neuron structure (4, 4L, 4R) for detecting the motion direction by means of a receptive field; a simulated opposition motion detection neuron structure (5, 5L, 5R) for detecting opposition motion at least relating to expansion and contraction; a simulated complex pattern detection neuron structure (6, 6L, 6R) for globally detecting an optical flow pattern over the entire visual observation and according to the evolution of the entire visual observation during the time, the detectable pattern being a prototype pattern; and a simulated motion pattern detection neuron structure (7, 7LR) for detecting a motion pattern and providing a decision regarding a motion characteristic. According to the invention, the neurons of the simulated motion pattern detection neuron structure (7, 7LR) comprise a forgetting ability that is a function of the delay and for each neuron an activity of the neuron.

Description

technical field [0001] The present invention relates to a detection and decision-making system that simulates the decision-making process in the mammalian brain with respect to visually observed body movements. Background technique [0002] Humans' powerful ability to recover information about moving living organisms (e.g., identity or activity type) from sparse visual input is known as biokinesis. Biological motion perception has long been the subject of research. Motion perception is an important biological function, and it is useful in many activities ranging from basic survival activities of mammals (and especially humans) to social life. Biological models of the neuronal structures and their relationships involved in biological motion perception have been developed. [0003] So far, it has been identified that functions in biological motion perception involve the integration of local motion detection and recognition of dynamic shape cues. More precisely, this functio...

AI Technical Summary

Problems solved by technology

[0008] However, the results provided by their simulations in non-biological machines (especially computers), which attempted to mimic the biological structures involved, were not very efficient and inconsistent with the real biological neuronal structures involved in the perception of biological motion in living things. The result provided is not close enough

Images