A Multiview Recognition Method of Predefined Objects for Robot Assembly Using Deep Learning and Its Implementation on an FPGA
RGB color model
DOI:
10.3390/electronics11050696
Publication Date:
2022-02-24T12:22:35Z
AUTHORS (4)
ABSTRACT
The process of recognizing manufacturing parts in real time requires fast, accurate, small, and low-power-consumption sensors. Here, we describe a method to extract descriptors from several objects observed wide range angles three-dimensional space. These define the dataset, which allows for training further validation convolutional neural network. classification is implemented reconfigurable hardware an embedded system with RGB sensor processing unit. achieved accuracy 96.67% speed 2.25× faster than results reported state-of-the-art solutions. Our proposal 655 times implementation on PC. presented meets criteria real-time video it suitable as enhancement hand robotic arm intelligent cell.
SUPPLEMENTAL MATERIAL
Coming soon ....
REFERENCES (27)
CITATIONS (3)
EXTERNAL LINKS
PlumX Metrics
RECOMMENDATIONS
FAIR ASSESSMENT
Coming soon ....
JUPYTER LAB
Coming soon ....