http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-114973152-B
Outgoing Links
| Predicate | Object |
|---|---|
| classificationCPCAdditional | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/Y02W90-00 |
| classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06V20-52 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/H04L67-12 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06V10-762 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06V10-774 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-045 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06V10-82 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-08 |
| classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06V10-82 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-08 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06V20-52 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06V10-762 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/H04L67-12 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06V10-774 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-04 |
| filingDate | 2022-07-27-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| grantDate | 2022-11-04-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationDate | 2022-11-04-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationNumber | CN-114973152-B |
| titleOfInvention | Monitoring method, device and medium for small molecule recyclable fracturing fluid storage tank based on neural network |
| abstract | The present invention provides a method, device and storage medium for monitoring a small molecule recirculating fracturing fluid storage tank based on a neural network. The method includes: acquiring a plurality of images with a target through a plurality of AI cameras arranged around the storage tank image of the target object, and send the plurality of images with the target object to the server; in the server, the plurality of images with the target object are sorted according to the generation time, and then a first image sequence is generated, and the first image sequence is generated. The images in the first image sequence are clustered using the DBSCAN algorithm to obtain N clusters, and the representative images of each selected cluster are sorted based on the generation time to obtain the second image sequence. The second image sequence is processed to determine whether the target object is a dangerous target, and the first neural network model is trained and generated using a loss function based on image quality as a weight. The present invention can accurately identify images of various qualities. |
| priorityDate | 2022-07-27-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Total number of triples: 32.