http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-107680081-B
Outgoing Links
Predicate | Object |
---|---|
classificationCPCAdditional | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-10036 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-20084 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-20081 |
classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T7-0002 |
classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T7-00 |
filingDate | 2017-09-08-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
grantDate | 2020-06-23-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationDate | 2020-06-23-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber | CN-107680081-B |
titleOfInvention | Hyperspectral image unmixing method based on convolutional neural network |
abstract | The invention discloses a hyperspectral image unmixing method based on a convolutional neural network, which mainly solves the problems of low unmixing precision of hyperspectral images, complex model, long time consumption and low efficiency in the hyperspectral image unmixing process in the prior art. The problem. The steps of the present invention are: acquiring a data matrix, preprocessing the data matrix, constructing a convolutional neural network with a 10-layer structure containing a pixel-based fuzzy classification structure, training the convolutional neural network, fuzzy classification, and outputting the convolutional neural network The results are normalized to obtain unmixed results. The present invention introduces a convolutional neural network model based on a pixel-based fuzzy classification structure, and has the advantages of high unmixing accuracy, simple model, low calculation amount and easy implementation. |
priorityDate | 2017-09-08-04:00^^<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/substance/SID419512635 http://rdf.ncbi.nlm.nih.gov/pubchem/compound/CID962 |
Total number of triples: 15.