http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-109558888-A
Outgoing Links
Predicate | Object |
---|---|
assignee | http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_fa04e128d163cc429a0fc5fb3c4f653b |
classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-2411 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-24147 |
classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06K9-62 |
filingDate | 2017-09-27-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
inventor | http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_c86e2409ebc5654fb420ca31a9779082 |
publicationDate | 2019-04-02-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber | CN-109558888-A |
titleOfInvention | A Parallelized Acceleration Algorithm for Hyperspectral Remote Sensing Image Classification |
abstract | The invention belongs to the field of remote sensing image processing, in particular to a parallelized acceleration algorithm for hyperspectral remote sensing image classification. The invention firstly uses the manifold learning algorithm to reduce the dimension of the image to obtain the feature matrix to be decomposed; then uses the implicit restart Lanczos method to decompose the feature, wherein by calling the GPU to accelerate the time-consuming matrix operation module, the operation consumption is greatly reduced Finally, use the learned embedding vector as the input of the classifier to classify the hyperspectral image. The invention makes full use of the algorithm characteristics combined with the GPU device to accelerate the image classification processing process, and cleverly uses the compressed sparse storage format to greatly save the memory space. real-time application requirements. |
isCitedBy | http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-111488502-A http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-113034343-A http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-111428787-A http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-113034343-B |
priorityDate | 2017-09-27-04:00^^<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/SID419701332 http://rdf.ncbi.nlm.nih.gov/pubchem/compound/CID22978774 |
Total number of triples: 18.