http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-112767413-B
Outgoing Links
| Predicate | Object |
|---|---|
| classificationCPCAdditional | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-10032 |
| classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T5-002 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T7-11 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T7-187 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-2411 |
| classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T5-00 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06K9-62 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T7-187 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06V10-764 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T7-11 |
| filingDate | 2021-01-06-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| grantDate | 2022-03-15-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationDate | 2022-03-15-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationNumber | CN-112767413-B |
| titleOfInvention | A deep semantic segmentation method for remote sensing images based on the constraints of regional connectivity and co-occurrence knowledge |
| abstract | The invention discloses a deep semantic segmentation method of remote sensing images that integrates regional connectivity and co-occurrence knowledge constraints. This method adds target integrity constraints based on regional connectivity and target spatial distribution constraints based on spatial co-occurrence knowledge into the loss function terms of the deep semantic segmentation network. The network adjusts the network model by optimizing the comprehensive loss function terms added to the constraints to learn autonomously. Object-level feature representation and the use of spatial co-occurrence knowledge to automatically optimize the spatial distribution of segmentation objects. The invention innovatively proposes a deep semantic segmentation method of remote sensing images that integrates regional connectivity constraints and spatial co-occurrence knowledge constraints, which is the first time in the field of semantic segmentation to achieve constraints on the entire target; Embedding unstructured knowledge into the data-driven deep semantic segmentation network can effectively improve the segmentation accuracy and segmentation results of the deep semantic segmentation network. |
| priorityDate | 2021-01-06-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Total number of triples: 21.