Noise reduction method and system for a pulse array signal

Abstract

The invention discloses a pulse array signal noise reduction method, which comprises the following steps of: modeling distribution characteristics of pulse array signals in a spatial neighborhood anda timestamp neighborhood of a pulse to be denoised, establishing a distribution characteristic model of the pulse array signals, and extracting space-time distribution characteristics of the pulse array signals; constructing a space-time filter according to noise distribution characteristics in the pulse array signal, and filtering the extracted characteristics; and restoring the filtered pulse space-time distribution characteristics into pulse array signals according to the distribution characteristic model. The space-time distribution characteristics of the noise in the pulse array signal are fully considered. The modeling is carried out on the transitivity of the noise in the time domain and the randomness of the noise in the space domain. The physical significance represented by the pulse is combined. The space-time filter is constructed to carry out efficient denoising on the pulse signal. The efficiency is high, and the denoising effect is good.

Description

technical field [0001] The invention relates to the technical field of signal processing, in particular to a noise reduction method and system for pulse array signals. Background technique [0002] Traditional vision sensors usually take frames as units, and perform complete sampling of the scene according to a preset fixed frequency. This kind of sampling based on a fixed frame rate cannot reflect the dynamic changes of the scene, and it is easy to oversample or undersample the current scene, resulting in large video data redundancy, low temporal resolution, and easy blurring under high-speed motion. Inspired by the visual sampling mechanism of the biological retina, new cameras that collect pulse array signals have entered people's field of vision, including sensors that emit pulse signals based on changes in light intensity, such as Dynamic Vision Sensor (DVS), image sensors based on asynchronous time (Asynchronous Time-based image Sensor, ATIS), dynamic active pixel vis...

AI Technical Summary

Problems solved by technology

This kind of sampling based on a fixed frame rate cannot reflect the dynamic changes of the scene, and it is easy to oversample or undersample the current scene, resulting in large video data redundancy, low temporal resolution, and easy blurring under high-speed motion.

Images