| Predicate |
Object |
| assignee |
http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_bd1f583871320ad42c3df038ce51c5a2 |
| classificationCPCInventive |
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-241
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-2136
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-214 |
| classificationIPCInventive |
http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06V10-764
http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06K9-62
http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06V20-40
http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06V10-77
http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06V10-774 |
| filingDate |
2021-10-19^^<http://www.w3.org/2001/XMLSchema#date> |
| inventor |
http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_fb4bb46ac6b5c63ecf0a65082c0d45e9
http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_edce2c60911119feb9630889a749acef
http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_6e97bae39d983b55b76f88bdf04d261d
http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_9768b478739d93d5b3fb0a23ab1b9717
http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_06f9769c7f2c1f5b94cf8050136d387f
http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_1b39389ee702ddae4d0b5897dd5f6c6b
http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_d89c20124ef9c211df58c8489c7ab301
http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_dbdfe316781313fd4fe31413f6e1cb8d |
| publicationDate |
2022-02-11^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationNumber |
CN-114037931-A |
| titleOfInvention |
A multi-view discrimination method with adaptive weights |
| abstract |
The invention discloses an adaptive weight multi-view discrimination method, comprising the following steps: constructing an undirected weight map of different views of a data set, and calculating a Laplacian matrix L s ; based on Hilbert-Schmidt The Hilbert-Schmidt Independence Criteria (HSIC) constrains the consistency of different view data, and calculates the constraint matrix T; combines the consensus low-rank sparse representation learning method to optimize the projection matrix P; introduces the weight parameter And according to the amount of information contained in each view data, the corresponding weights are adaptively assigned; the final multi-view discriminant model with adaptive weights is constructed; by solving the target model, the optimal multi-view projection matrix of the model is obtained; Perform discriminant analysis and use KNN algorithm to obtain the accuracy of image recognition. The present invention is aimed at maintaining the consistent structure of different views in noise-contaminated image data, and has strong accuracy and robustness. |
| isCitedBy |
http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-116402554-B
http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-116402554-A |
| priorityDate |
2021-10-19^^<http://www.w3.org/2001/XMLSchema#date> |
| type |
http://data.epo.org/linked-data/def/patent/Publication |