http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-115100679-A
Outgoing Links
| Predicate | Object |
|---|---|
| assignee | http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_d6a6f422b091ba12ea61d4adbf1b0e8e |
| classificationCPCAdditional | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/Y02A40-81 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06V2201-07 |
| classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06V40-10 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06Q50-02 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06V10-774 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06V10-82 |
| classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06V10-82 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06Q50-02 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06V40-10 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-04 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06V10-774 |
| filingDate | 2022-06-17-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| inventor | http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_02a8054c9ac86a5c40abd296d2c5c8b9 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_add3a006cccda949d8048c8f30aded46 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_471d9bfe04872a1ea9e03343ee494d1f |
| publicationDate | 2022-09-23-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationNumber | CN-115100679-A |
| titleOfInvention | A method for detecting the posture of the head, tail, ventral and back of the fish body |
| abstract | The invention discloses a method for detecting the posture of the head, tail, ventral and back of a fish body. Obtaining images of different fish body postures, and labeling the posture information, and constructing data sets of different fish body postures; preprocessing the data sets, including offline and online data enhancement, and expanding the data sets; The data set is input to the YOLOv5 neural network for training, a new target loss function is constructed, and the parameters of the model are continuously optimized by continuous iterative updating, and the trained YOLOv5 neural network is obtained; the trained YOLOv5 neural network is used to process the image to be tested to obtain the fish head The detection results of the tail-vendor-dorsal posture. The invention solves the problems that manual assisted orientation is required before fish removal processing, and the traditional fish body attitude detection method has single feature and poor robustness, and realizes fish body attitude definition and multi-target online attitude detection. |
| priorityDate | 2022-06-17-04:00^^<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/SID457280508 |
Total number of triples: 23.