Pulse graph neural network tactile object identification method based on graph learning

The invention provides a pulse graph neural network tactile object recognition method based on graph learning, and the algorithm comprises the steps: obtaining a tactile graph, and constructing an M-tree tactile graph or a Z-tree tactile graph based on the tactile graph; an object recognition model...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: HU BINGQI, YU ZUKUN, ZHANG JUNXING, LI HUA, YANG JING, RUAN XIAOLI
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator HU BINGQI
YU ZUKUN
ZHANG JUNXING
LI HUA
YANG JING
RUAN XIAOLI
description The invention provides a pulse graph neural network tactile object recognition method based on graph learning, and the algorithm comprises the steps: obtaining a tactile graph, and constructing an M-tree tactile graph or a Z-tree tactile graph based on the tactile graph; an object recognition model is established, the M-tree touch image or the Z-tree touch image is used for training the object recognition model, and the object recognition model comprises LIF spiking neurons, a topological adaptive image convolution layer, a full connection layer and a final voting layer; performing weighting by using Gaussian prior distribution loss and back propagation training loss of an object recognition model of LIF spiking neurons, and then optimizing the object recognition model; and judging the category of the touch object by using the optimized object recognition model. According to the method, the Gaussian prior distribution loss and the back propagation training loss of the LIF spiking neuron object recognition mod
format Patent
fullrecord <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_CN118038236A</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>CN118038236A</sourcerecordid><originalsourceid>FETCH-epo_espacenet_CN118038236A3</originalsourceid><addsrcrecordid>eNrjZAgPKM0pTlVIL0osyFDISy0tSswBUiXl-UXZCiWJySWZOakK-UlZqcklCpkpqXklmWmZyYklmfl5CrmpJRn5KQpJicWpKQpAPsSInNTEorzMvHQeBta0RKDJvFCam0HRzTXE2UM3tSA_PrW4IDE5FWhNvLOfoaGFgbGFkbGZozExagCGMDow</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>Pulse graph neural network tactile object identification method based on graph learning</title><source>esp@cenet</source><creator>HU BINGQI ; YU ZUKUN ; ZHANG JUNXING ; LI HUA ; YANG JING ; RUAN XIAOLI</creator><creatorcontrib>HU BINGQI ; YU ZUKUN ; ZHANG JUNXING ; LI HUA ; YANG JING ; RUAN XIAOLI</creatorcontrib><description>The invention provides a pulse graph neural network tactile object recognition method based on graph learning, and the algorithm comprises the steps: obtaining a tactile graph, and constructing an M-tree tactile graph or a Z-tree tactile graph based on the tactile graph; an object recognition model is established, the M-tree touch image or the Z-tree touch image is used for training the object recognition model, and the object recognition model comprises LIF spiking neurons, a topological adaptive image convolution layer, a full connection layer and a final voting layer; performing weighting by using Gaussian prior distribution loss and back propagation training loss of an object recognition model of LIF spiking neurons, and then optimizing the object recognition model; and judging the category of the touch object by using the optimized object recognition model. According to the method, the Gaussian prior distribution loss and the back propagation training loss of the LIF spiking neuron object recognition mod</description><language>chi ; eng</language><subject>CALCULATING ; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS ; COMPUTING ; COUNTING ; PHYSICS</subject><creationdate>2024</creationdate><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&amp;date=20240514&amp;DB=EPODOC&amp;CC=CN&amp;NR=118038236A$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,308,780,885,25563,76418</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&amp;date=20240514&amp;DB=EPODOC&amp;CC=CN&amp;NR=118038236A$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>HU BINGQI</creatorcontrib><creatorcontrib>YU ZUKUN</creatorcontrib><creatorcontrib>ZHANG JUNXING</creatorcontrib><creatorcontrib>LI HUA</creatorcontrib><creatorcontrib>YANG JING</creatorcontrib><creatorcontrib>RUAN XIAOLI</creatorcontrib><title>Pulse graph neural network tactile object identification method based on graph learning</title><description>The invention provides a pulse graph neural network tactile object recognition method based on graph learning, and the algorithm comprises the steps: obtaining a tactile graph, and constructing an M-tree tactile graph or a Z-tree tactile graph based on the tactile graph; an object recognition model is established, the M-tree touch image or the Z-tree touch image is used for training the object recognition model, and the object recognition model comprises LIF spiking neurons, a topological adaptive image convolution layer, a full connection layer and a final voting layer; performing weighting by using Gaussian prior distribution loss and back propagation training loss of an object recognition model of LIF spiking neurons, and then optimizing the object recognition model; and judging the category of the touch object by using the optimized object recognition model. According to the method, the Gaussian prior distribution loss and the back propagation training loss of the LIF spiking neuron object recognition mod</description><subject>CALCULATING</subject><subject>COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS</subject><subject>COMPUTING</subject><subject>COUNTING</subject><subject>PHYSICS</subject><fulltext>true</fulltext><rsrctype>patent</rsrctype><creationdate>2024</creationdate><recordtype>patent</recordtype><sourceid>EVB</sourceid><recordid>eNrjZAgPKM0pTlVIL0osyFDISy0tSswBUiXl-UXZCiWJySWZOakK-UlZqcklCpkpqXklmWmZyYklmfl5CrmpJRn5KQpJicWpKQpAPsSInNTEorzMvHQeBta0RKDJvFCam0HRzTXE2UM3tSA_PrW4IDE5FWhNvLOfoaGFgbGFkbGZozExagCGMDow</recordid><startdate>20240514</startdate><enddate>20240514</enddate><creator>HU BINGQI</creator><creator>YU ZUKUN</creator><creator>ZHANG JUNXING</creator><creator>LI HUA</creator><creator>YANG JING</creator><creator>RUAN XIAOLI</creator><scope>EVB</scope></search><sort><creationdate>20240514</creationdate><title>Pulse graph neural network tactile object identification method based on graph learning</title><author>HU BINGQI ; YU ZUKUN ; ZHANG JUNXING ; LI HUA ; YANG JING ; RUAN XIAOLI</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_CN118038236A3</frbrgroupid><rsrctype>patents</rsrctype><prefilter>patents</prefilter><language>chi ; eng</language><creationdate>2024</creationdate><topic>CALCULATING</topic><topic>COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS</topic><topic>COMPUTING</topic><topic>COUNTING</topic><topic>PHYSICS</topic><toplevel>online_resources</toplevel><creatorcontrib>HU BINGQI</creatorcontrib><creatorcontrib>YU ZUKUN</creatorcontrib><creatorcontrib>ZHANG JUNXING</creatorcontrib><creatorcontrib>LI HUA</creatorcontrib><creatorcontrib>YANG JING</creatorcontrib><creatorcontrib>RUAN XIAOLI</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>HU BINGQI</au><au>YU ZUKUN</au><au>ZHANG JUNXING</au><au>LI HUA</au><au>YANG JING</au><au>RUAN XIAOLI</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>Pulse graph neural network tactile object identification method based on graph learning</title><date>2024-05-14</date><risdate>2024</risdate><abstract>The invention provides a pulse graph neural network tactile object recognition method based on graph learning, and the algorithm comprises the steps: obtaining a tactile graph, and constructing an M-tree tactile graph or a Z-tree tactile graph based on the tactile graph; an object recognition model is established, the M-tree touch image or the Z-tree touch image is used for training the object recognition model, and the object recognition model comprises LIF spiking neurons, a topological adaptive image convolution layer, a full connection layer and a final voting layer; performing weighting by using Gaussian prior distribution loss and back propagation training loss of an object recognition model of LIF spiking neurons, and then optimizing the object recognition model; and judging the category of the touch object by using the optimized object recognition model. According to the method, the Gaussian prior distribution loss and the back propagation training loss of the LIF spiking neuron object recognition mod</abstract><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier
ispartof
issn
language chi ; eng
recordid cdi_epo_espacenet_CN118038236A
source esp@cenet
subjects CALCULATING
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
COMPUTING
COUNTING
PHYSICS
title Pulse graph neural network tactile object identification method based on graph learning
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-08T14%3A36%3A22IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-epo_EVB&rft_val_fmt=info:ofi/fmt:kev:mtx:patent&rft.genre=patent&rft.au=HU%20BINGQI&rft.date=2024-05-14&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3ECN118038236A%3C/epo_EVB%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true