http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-105046269-B
Outgoing Links
Predicate | Object |
---|---|
classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F16-35 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-2411 |
classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06K9-62 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F16-35 |
filingDate | 2015-06-19-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
grantDate | 2019-02-22-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationDate | 2019-02-22-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber | CN-105046269-B |
titleOfInvention | A multi-instance and multi-label scene classification method based on multi-kernel fusion |
abstract | The present invention relates to a multi-instance multi-label scene classification method based on multi-core fusion, comprising: inputting a multi-instance multi-label data set, splitting it into a multi-instance data set and a multi-label data set; using different thresholds to classify the multi-instance A correlation matrix is established for each package in the data set; the basic kernel function between each two multi-instance data packets under the same threshold is obtained according to the obtained correlation matrix, and the basic kernel function values form a basic kernel matrix; The element values at the same position in the kernel matrix are convexly combined to obtain a multi-kernel matrix; using multi-label dataset training, multiple multi-kernel SVM classifiers are obtained. The multi-kernel SVM classifier is used to predict the label set of unknown multi-instance data packets to achieve scene classification. A multi-instance multi-label scene classification method based on multi-core fusion of the present invention improves the accuracy of scene classification. The invention also relates to a multi-instance multi-label scene classification system based on multi-core fusion. |
priorityDate | 2015-06-19-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Predicate | Subject |
---|---|
isDiscussedBy | http://rdf.ncbi.nlm.nih.gov/pubchem/compound/CID25572 http://rdf.ncbi.nlm.nih.gov/pubchem/substance/SID415713197 |
Total number of triples: 14.