| Predicate |
Object |
| assignee |
http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_9b2144c45d4fe27551278e028fe6f18a |
| classificationCPCInventive |
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T7-0004
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T5-40
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/G06T7-00
http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T5-40 |
| filingDate |
2017-12-22-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| inventor |
http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_87ecc0f57decd8cf172bf2180f5931d8
http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_4f6f78638d1302072b4c8ae063359b35
http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_3b4b2b907796b5a540c1d7332cb620ba
http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_1bdd04113447c8e2219758d2efc75433
http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_edcfe121a483001c715f4e05958bfb0b
http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_22bf95a21e6481bab658735843f43357
http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_2150750995a69644849afef4d5eca764
http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_bb6c116ec97cb47052ffafb3fe8ae59f |
| publicationDate |
2018-06-12-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationNumber |
CN-108154502-A |
| titleOfInvention |
A Convolutional Neural Network Based Solder Joint Recognition Method |
| abstract |
The invention provides a method for identifying through-hole solder joints based on a convolutional neural network. The identification method includes four steps: (1) performing a segmentation operation on the input image based on the similarity of the area pixels; (2) dividing the segmented area image Perform histogram image equalization; (3) serialize and normalize the input area images; (4) input the equalized area images into multi-layer convolution and pooling operations to extract convolution features; then These convolutional features are input into multi-layer fully connected layers for via solder joint classification, and output whether the region image is a via solder joint or not. The invention can realize the identification of through-hole solder joints in the PCB circuit board image, has the characteristics of high speed and high accuracy, and helps automatic soldering machines and other PCB welding equipment to automatically pick up spots. |
| isCitedBy |
http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-112036484-A
http://rdf.ncbi.nlm.nih.gov/pubchem/patent/WO-2023115409-A1
http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-109916921-A
http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-112285114-A
http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-114897889-B
http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-110929795-A
http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-110421283-A
http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-110929795-B
http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-112200164-A
http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-109877496-A
http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-109859164-A
http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-114897889-A
http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-111784777-A
http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-111504194-A
http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-109035182-B
http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-114299086-A
http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-109035182-A |
| priorityDate |
2017-12-22-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| type |
http://data.epo.org/linked-data/def/patent/Publication |