Sensing, storing and computing integrated visual system based on photoelectric synapse device and memristive device

Through the integrated visual system of sensor, memory and computing based on photoelectric synaptic devices and memristor devices, combined with multi-layer convolutional neural network, the delay and compatibility problems of traditional computer vision systems are solved, efficient image processing and recognition are realized, and intelligence is promoted Development of visual sensing technology.

CN119296068AActive Publication Date: 2025-01-10NANJING UNIV OF POSTS & TELECOMM +1
View PDF 2 Cites 5 Cited by

Patent Information

Application Number
CN202411833215.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-12-13
Publication Date
2025-01-10
Estimated Expiration
2044-12-13

AI Technical Summary

Technical Problem

Due to the limitations of transmission bandwidth and memory performance, traditional computer vision systems have large delays between sensors, memory and processors, which is difficult to meet the efficient data processing needs of intelligent systems, especially in the fields of autonomous driving and micro-robots, and traditional silicon-based Logic gate circuits are difficult to compatible with new biological neural devices.

Method used

Adopt an integrated visual system based on photoelectric synaptic devices and memristor devices, image sensing and preprocessing is performed through matrix arrangement of photoelectric synaptic devices, image recognition is used by array memristor devices, and multi-layer convolutional neural networks are combined. Perform deep processing to realize image feature extraction and recognition.

Benefits of technology

It realizes the conversion and noise reduction processing of image optical information to electrical signals, improves computing power and recognition accuracy, eliminates the delay and power consumption of data transmission between the computing unit and the storage unit, and promotes lightweight and intelligent visual transmission. The development of technology.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119296068A_ABST
    Figure CN119296068A_ABST
Patent Text Reader

Abstract

The invention discloses a sensing, storing and computing integrated visual system based on a photoelectric synapse device and a memristor device, and belongs to the field of intelligent vision and brain-like computing. The image sensing module is used for converting two-dimensional image information of an image to be detected into a two-dimensional electric signal by utilizing the matrix arrangement photoelectric synapse device, and preprocessing the two-dimensional electric signal; the array memristor-based image recognition module is used for receiving the preprocessed two-dimensional electric signals, standardizing the two-dimensional electric signals to obtain an image feature matrix by using a memristor array structure, inputting the feature matrix into a multi-layer convolutional neural network, and performing matrix operation on the input feature matrix for multiple times by the multi-layer convolutional neural network to obtain an image feature matrix; and further extracting image feature information, and finally outputting an image recognition result. The visual system can be widely applied to the field of image sensing and recognition, and has the characteristics of blurred image enhancement, high recognition rate and the like.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Method for realizing calculation of endosensory reserve pool based on photosynaptic device

    CN114723000A

  • Artificial photoelectric synapse device based on topological insulator and application thereof

    CN115548147A

Cited By

  • Biomimetic moving target sensing, storage and calculation integrated photoelectric detection system based on continuous photoconduction

    CN119863667A

  • Bio-inspired Moving Target Sensing and Computing-in-Memory Photodetector System Based on Persistent Photoconductivity

    CN119863667B

  • Direction-selective sensing-storage-calculation integrated chip based on heterogeneous integration and preparation method thereof

    CN121705237A

  • Directional selective spiking integrated chip based on heterogeneous integration and preparation method

    CN121705237B

  • Persistent-photoconductivity-based bio-inspired sensing-memory-computing integrated photoelectric detection system for moving target

    WO2026199857A1