http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-112541929-A
Outgoing Links
Predicate | Object |
---|---|
assignee | http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_a4f7bbb1e2ce1e94f80ec66ec20f1ba7 |
classificationCPCAdditional | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-20081 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-10012 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-20021 |
classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-08 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-045 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T7-11 |
classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-04 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T7-11 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_ab362721d10d4dd9566b5dc4f15443bc http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_21307dd7810cc807cfc1730a5dc87ac7 |
publicationDate | 2021-03-23-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber | CN-112541929-A |
titleOfInvention | An image processing method and system for convolutional neural network |
abstract | The present application discloses an image segmentation method for a convolutional neural network. The 3D image A is tiled in width. Store multiple image blocks in the same number of multiple buffers respectively. Sets the left and right increments of each image block and how they are read. When it is necessary to read the data of the left or right added part of a certain image block, it is read from a buffer corresponding to an image block on the left or right of the image block. When the application divides the image width into blocks, the overlapping parts between the image blocks are not repeatedly written into the storage, and the image data sharing of the overlapping parts between the image blocks is realized by setting the reading strategy of the left added part and the right added part. This method is especially suitable for quantization implementation of convolutional neural network on FPGA. |
isCitedBy | http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-112991144-A http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-113933111-B http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-113933111-A |
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/SID453034310 http://rdf.ncbi.nlm.nih.gov/pubchem/compound/CID516892 |
Total number of triples: 24.