http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-110020674-A
Outgoing Links
| Predicate | Object |
|---|---|
| assignee | http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_0ee33b0e3e7398cf6fc957eeee59d510 |
| classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-213 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F16-55 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-2411 |
| classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F16-55 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06K9-62 |
| filingDate | 2019-03-13-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| inventor | http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_0db57ee30d1766c06b02ee0e6a232d29 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_e6ae2aadcb111bfdd61594bbcb2cce83 |
| publicationDate | 2019-07-16-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationNumber | CN-110020674-A |
| titleOfInvention | A cross-domain adaptive image classification method to improve local class discrimination |
| abstract | The invention relates to a cross-domain adaptive image classification method for improving the discrimination degree of local categories, and belongs to the technical field of image processing. In the cross-domain adaptive image classification method of the present invention, when an image content classification model is trained, the images in the training set and the test set are different in color tone, angle, sharpness, etc., and the data representing the image information obeys different probability distributions. Learn the features shared by the two image distributions, and at the same time, for each image, among other images with similar styles, mine the common points between images with the same content and the differences between images with different content, so that the same The content images are more aggregated, and the mutual interference between different types of content images is reduced, thereby improving the local category discrimination of the image data set, thereby improving the accuracy of image classification. |
| isCitedBy | http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-112199572-A http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-112199572-B http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-112861929-A http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-112861929-B http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-112426161-A |
| priorityDate | 2019-03-13-04:00^^<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/SID426113427 http://rdf.ncbi.nlm.nih.gov/pubchem/compound/CID6476031 |
Total number of triples: 22.