http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-113435546-A
Outgoing Links
Predicate | Object |
---|---|
assignee | http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_a35112499840ad9299fa681b0bcc01e8 |
classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-214 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-2431 |
classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06K9-62 |
filingDate | 2021-08-26-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
inventor | http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_4de44f867a475a0112cf0f6f3886962a http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_edd4c6e47ed744ea26bf56145c075fad http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_a798fd2f3628d987931660d934eafad0 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_b898ee18c57e72217836f1603ee13018 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_b9a9b7dad6cabc75b96b3666fe8eb70e http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_39abd195c4671e061393731f1e5fc70e |
publicationDate | 2021-09-24-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber | CN-113435546-A |
titleOfInvention | Transferable Image Recognition Method and System Based on Discrimination Confidence Level |
abstract | The invention discloses a transferable image recognition method and system based on the discrimination confidence level, which firstly adopts the source domain data training to obtain the source domain pre-training model, and uses the parameters obtained by the source domain model training as the feature extraction parameters of the target domain model and classification parameters, so that the target domain model selects pseudo-label trusted samples from the target domain data based on the training parameters of the source domain model, and uses the selected trusted samples to assign pseudo-labels and weights to untrusted samples, which effectively reduces the The uncertainty of the pseudo-labels of all current target domain images; finally, the target domain model is optimized by training the target domain data with pseudo-labels together with the source domain data, so that the target image recognition performance of the final target domain model has been greatly improved , which can perform fast migration and effective image recognition; and effectively reduce the annotation of target image recognition, which greatly reduces manpower and material resources. |
isCitedBy | http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-115186773-B http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-115186773-A http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-114220016-B http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-114220016-A http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-114998602-A http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-114239753-A |
priorityDate | 2021-08-26-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Predicate | Subject |
---|---|
isCitedBy | http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-113326731-A |
isDiscussedBy | http://rdf.ncbi.nlm.nih.gov/pubchem/substance/SID456171974 |
Total number of triples: 25.