http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-107424147-A

Outgoing Links

Predicate Object
assignee http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_2d09298f58671950aeb56a0514d3e5e2
classificationCPCAdditional http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-20081
classificationCPCInventive http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T7-0002
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-23213
classificationIPCInventive http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06K9-62
http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T7-00
filingDate 2017-07-03-04:00^^<http://www.w3.org/2001/XMLSchema#date>
inventor http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_183ccc655f423133105e76b69cfbf038
http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_5bc3b8f32f89c269250f031638de4116
publicationDate 2017-12-01-04:00^^<http://www.w3.org/2001/XMLSchema#date>
publicationNumber CN-107424147-A
titleOfInvention Pattern Defect Recognition and Location Method Based on Latent Dirichlet Distribution Model
abstract The invention discloses a method for identifying and locating graphic defects based on a hidden Dirichlet distribution model, which includes a training phase and a testing phase. Features, forming a chaotic feature vector, a training image is represented by a chaotic feature vector matrix; all trained chaotic feature vector matrices are clustered by the k-means clustering method to form a codebook, and then a primary histogram is formed through the codebook; The high-level histogram is obtained by learning the primary histogram through the hidden Dirichlet distribution model, and the training image is represented by the high-level histogram; the test phase includes the following steps: the test image is represented by the high-level histogram; the above-mentioned high-level histogram is calculated The similarity between the graph and the advanced histogram is used to judge the defect category and location of the test image.
isCitedBy http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-111258905-A
priorityDate 2017-07-03-04:00^^<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/compound/CID25572
http://rdf.ncbi.nlm.nih.gov/pubchem/substance/SID415713197

Total number of triples: 18.