A Multiview Recognition Method of Predefined Objects for Robot Assembly Using Deep Learning and Its Implementation on an FPGA
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 from a wide range of angles in a three-dimensional space. These descriptors define the datase...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-02-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/11/5/696 |
| _version_ | 1797475331210215424 |
|---|---|
| author | Victor Lomas-Barrie Ricardo Silva-Flores Antonio Neme Mario Pena-Cabrera |
| author_facet | Victor Lomas-Barrie Ricardo Silva-Flores Antonio Neme Mario Pena-Cabrera |
| author_sort | Victor Lomas-Barrie |
| collection | DOAJ |
| description | 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 from a wide range of angles in a three-dimensional space. These descriptors define the dataset, which allows for the training and further validation of a convolutional neural network. The classification is implemented in reconfigurable hardware in an embedded system with an RGB sensor and the processing unit. The system achieved an accuracy of 96.67% and a speed <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>2.25</mn><mo>×</mo></mrow></semantics></math></inline-formula> faster than the results reported for state-of-the-art solutions. Our proposal is 655 times faster than implementation on a PC. The presented embedded system meets the criteria of real-time video processing and it is suitable as an enhancement for the hand of a robotic arm in an intelligent manufacturing cell. |
| first_indexed | 2024-03-09T20:43:34Z |
| format | Article |
| id | doaj.art-4480e464e7f446858ebb0648b9b54631 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-09T20:43:34Z |
| publishDate | 2022-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-4480e464e7f446858ebb0648b9b546312023-11-23T22:52:38ZengMDPI AGElectronics2079-92922022-02-0111569610.3390/electronics11050696A Multiview Recognition Method of Predefined Objects for Robot Assembly Using Deep Learning and Its Implementation on an FPGAVictor Lomas-Barrie0Ricardo Silva-Flores1Antonio Neme2Mario Pena-Cabrera3Instituto de Investigaciones en Matematicas Aplicadas y en Sistemas, Universidad Nacional Autónoma de México, Mexico City 04510, MexicoInstituto de Investigaciones en Matematicas Aplicadas y en Sistemas, Universidad Nacional Autónoma de México, Mexico City 04510, MexicoUnidad Académica Mérida, IIMAS, Universidad Nacional Autónoma de México, Mérida 97290, MexicoInstituto de Investigaciones en Matematicas Aplicadas y en Sistemas, Universidad Nacional Autónoma de México, Mexico City 04510, MexicoThe 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 from a wide range of angles in a three-dimensional space. These descriptors define the dataset, which allows for the training and further validation of a convolutional neural network. The classification is implemented in reconfigurable hardware in an embedded system with an RGB sensor and the processing unit. The system achieved an accuracy of 96.67% and a speed <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>2.25</mn><mo>×</mo></mrow></semantics></math></inline-formula> faster than the results reported for state-of-the-art solutions. Our proposal is 655 times faster than implementation on a PC. The presented embedded system meets the criteria of real-time video processing and it is suitable as an enhancement for the hand of a robotic arm in an intelligent manufacturing cell.https://www.mdpi.com/2079-9292/11/5/696robot visionFPGACNNobject detectionhardware implementationLeNET-5 |
| spellingShingle | Victor Lomas-Barrie Ricardo Silva-Flores Antonio Neme Mario Pena-Cabrera A Multiview Recognition Method of Predefined Objects for Robot Assembly Using Deep Learning and Its Implementation on an FPGA Electronics robot vision FPGA CNN object detection hardware implementation LeNET-5 |
| title | A Multiview Recognition Method of Predefined Objects for Robot Assembly Using Deep Learning and Its Implementation on an FPGA |
| title_full | A Multiview Recognition Method of Predefined Objects for Robot Assembly Using Deep Learning and Its Implementation on an FPGA |
| title_fullStr | A Multiview Recognition Method of Predefined Objects for Robot Assembly Using Deep Learning and Its Implementation on an FPGA |
| title_full_unstemmed | A Multiview Recognition Method of Predefined Objects for Robot Assembly Using Deep Learning and Its Implementation on an FPGA |
| title_short | A Multiview Recognition Method of Predefined Objects for Robot Assembly Using Deep Learning and Its Implementation on an FPGA |
| title_sort | multiview recognition method of predefined objects for robot assembly using deep learning and its implementation on an fpga |
| topic | robot vision FPGA CNN object detection hardware implementation LeNET-5 |
| url | https://www.mdpi.com/2079-9292/11/5/696 |
| work_keys_str_mv | AT victorlomasbarrie amultiviewrecognitionmethodofpredefinedobjectsforrobotassemblyusingdeeplearninganditsimplementationonanfpga AT ricardosilvaflores amultiviewrecognitionmethodofpredefinedobjectsforrobotassemblyusingdeeplearninganditsimplementationonanfpga AT antonioneme amultiviewrecognitionmethodofpredefinedobjectsforrobotassemblyusingdeeplearninganditsimplementationonanfpga AT mariopenacabrera amultiviewrecognitionmethodofpredefinedobjectsforrobotassemblyusingdeeplearninganditsimplementationonanfpga AT victorlomasbarrie multiviewrecognitionmethodofpredefinedobjectsforrobotassemblyusingdeeplearninganditsimplementationonanfpga AT ricardosilvaflores multiviewrecognitionmethodofpredefinedobjectsforrobotassemblyusingdeeplearninganditsimplementationonanfpga AT antonioneme multiviewrecognitionmethodofpredefinedobjectsforrobotassemblyusingdeeplearninganditsimplementationonanfpga AT mariopenacabrera multiviewrecognitionmethodofpredefinedobjectsforrobotassemblyusingdeeplearninganditsimplementationonanfpga |