A target recognition method, device, system and computer-readable storage medium

Abstract

The invention discloses a target recognition method, device, system and computer-readable storage medium. The method includes the following steps: acquiring original pulse data, determining a pulse sampling window, inputting pulses in the pulse sampling window to a pulse neural network, and The pulses of the spiking neural network are mapped, and the pulses are sequentially transmitted along the excitation neurons of each layer other than the last layer of excitation neurons, the pulses are transmitted along the last layer of excitation neurons to the inference layer, and the pulses are transmitted along the last layer of reasoning neurons. The reasoning neurons of each layer are transmitted sequentially to determine the recognition result; the device includes a pulse acquisition module, a sampling window module, a pulse mapping module, a neuron excitation module, a neuron reasoning module and a recognition result determination module; the system includes the above-mentioned device; The invention can realize accurate recognition and fast recognition of the target to be recognized, and can be better applied to the target with high moving speed, and can give consideration to both recognition accuracy and calculation amount.

Description

technical field [0001] The present invention relates to the technical field of object recognition, and more specifically, the present invention is an object recognition method, device, system and computer-readable storage medium. Background technique [0002] In recent years, the field of artificial intelligence has developed rapidly, and breakthroughs have been made in algorithms, hardware, chips, etc., especially in the image field represented by computer vision, and many research results have been widely commercialized and civilianized. people's daily life. However, the current artificial intelligence algorithm still takes the second-generation artificial neural network as the main core. This type of algorithm was proposed as early as the 1980s. With years of continuous research, it has fallen into a research bottleneck in the academic community. Therefore, people have shifted their focus to the third-generation artificial neural network (that is, the pulse neural networ...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the problems widely existing in existing target recognition technologies, such as poor recognition effect and inability to effectively recognize high-speed moving targets, the present invention innovatively provides a target recognition method, device, system and computer-readable storage medium. The neural network performs noise filtering and pulse enhancement on the original input pulse, which can realize accurate and efficient recognition of the target to be recognized, and can also significantly reduce the amount of calculation and power consumption

Images