http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-112767274-A
Outgoing Links
| Predicate | Object |
|---|---|
| assignee | http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_cb351a5a4e03ee3773bd0df5dbae9ac3 |
| classificationCPCAdditional | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-20084 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-20081 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-10052 |
| classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-08 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T5-003 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-045 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T5-005 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-044 |
| classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T5-00 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-04 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-08 |
| filingDate | 2021-01-25-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| inventor | http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_0be04b2ac6e1ce4e13ee8c4cde730112 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_d738da1dff2603b430fc0181b28bc711 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_c000c6922c8e21f09484df79fcdb3c5f |
| publicationDate | 2021-05-07-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationNumber | CN-112767274-A |
| titleOfInvention | A method for detecting and removing rain streaks in light field images based on transfer learning |
| abstract | The invention discloses a method for detecting and removing rain streaks in light field images based on migration learning. The method includes a depth map calculation module, a rain streak detection module and a rainwater removal module. The present invention first obtains the depth map by using the depth map calculation module, then detects the synthetic data rain stripe map through the rain streak detection module, and uses the Gaussian process module to self-supervisely detects the real scene data rain stripe map, and finally combines the rainy 3DEPI volume blocks, After the obtained depth map and the extracted rain streak map are concatenated, input the 3D recursive generative adversarial network to remove the rain, and so on repeatedly until a high-quality rain-free map is obtained. Compared with the same type of work, the present invention uses transfer learning to build a self-supervised network, which can more accurately extract rain stripes in real scenes and obtain high-quality rain-free maps, and has good generalization ability. |
| priorityDate | 2021-01-25-04:00^^<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: 24.