http://rdf.ncbi.nlm.nih.gov/pubchem/patent/JP-H0528245-A
Outgoing Links
Predicate | Object |
---|---|
assignee | http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_fd221d21998567268ab28734915df324 |
classificationCPCAdditional | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/Y02T10-12 |
classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N99-00 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-00 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T1-00 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F15-18 |
filingDate | 1991-07-24-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
inventor | http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_411d3abb94cc257e1ff95657e0021ad9 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_06fd8dd25032632099971d2f885d8227 |
publicationDate | 1993-02-05-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber | JP-H0528245-A |
titleOfInvention | Image recognition device |
abstract | (57) [Abstract] [Purpose] In the image recognition device, the feature on the boundary of the divided area is extracted, and it is not affected by the density unevenness at the time of printing, and the misidentification due to the difference in the character shape does not occur. To [Structure] The image dividing unit 3 includes a camera 1 through an image converting unit 2. Based on the grayscale image data from, the grayscale image is divided into a plurality of areas by vertical and horizontal lines, and the positions of the vertical lines and the horizontal lines are shifted, and divided into both divided areas. Image data is output respectively. The feature amount extraction unit 4 calculates feature amounts such as the density average value, the density variance value, the horizontal gravity center value, and the vertical gravity center value from the divided image data, and outputs the calculated feature amounts for each divided area, and the feature amount normalization is performed. The unit 5 normalizes each feature amount, and the recognition unit 6 recognizes the image of the inspection target object by learning back propagation using a neural network based on the normalized feature amounts. |
isCitedBy | http://rdf.ncbi.nlm.nih.gov/pubchem/patent/JP-2008228344-A http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-111044993-A http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-110622177-A |
priorityDate | 1991-07-24-04:00^^<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/SID413372539 http://rdf.ncbi.nlm.nih.gov/pubchem/compound/CID26001 |
Total number of triples: 20.