Handwritten Digits Recognition Based on a Parallel Optoelectronic Time-Delay Reservoir Computing System
In this work, the performance of an optoelectronic time-delay reservoir computing system for performing a handwritten digit recognition task is numerically investigated, and a scheme to improve the recognition speed using multiple parallel reservoirs is proposed. By comparing four image injection me...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-02-01
|
| Series: | Photonics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2304-6732/10/3/236 |
| _version_ | 1827748032364412928 |
|---|---|
| author | Dianzuo Yue Yushuang Hou Chunxia Hu Cunru Zang Yingzhe Kou |
| author_facet | Dianzuo Yue Yushuang Hou Chunxia Hu Cunru Zang Yingzhe Kou |
| author_sort | Dianzuo Yue |
| collection | DOAJ |
| description | In this work, the performance of an optoelectronic time-delay reservoir computing system for performing a handwritten digit recognition task is numerically investigated, and a scheme to improve the recognition speed using multiple parallel reservoirs is proposed. By comparing four image injection methods based on a single time-delay reservoir, we find that when injecting the histograms of oriented gradient (HOG) features of the digit image, the accuracy rate (AR) is relatively high and is less affected by the offset phase. To improve the recognition speed, we construct a parallel time-delay reservoir system including multi-reservoirs, where each reservoir processes part of the HOG features of one image. Based on 6 parallel reservoirs with each reservoir possessing 100 virtual nodes, the AR can reach about 97.8%, and the reservoir processing speed can reach about 1 × 10<sup>6</sup> digits per second. Meanwhile, the parallel reservoir system shows strong robustness to the parameter mismatch between multi-reservoirs. |
| first_indexed | 2024-03-11T06:01:36Z |
| format | Article |
| id | doaj.art-c63f19175c3d416989682f1e53d23799 |
| institution | Directory Open Access Journal |
| issn | 2304-6732 |
| language | English |
| last_indexed | 2024-03-11T06:01:36Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Photonics |
| spelling | doaj.art-c63f19175c3d416989682f1e53d237992023-11-17T13:18:12ZengMDPI AGPhotonics2304-67322023-02-0110323610.3390/photonics10030236Handwritten Digits Recognition Based on a Parallel Optoelectronic Time-Delay Reservoir Computing SystemDianzuo Yue0Yushuang Hou1Chunxia Hu2Cunru Zang3Yingzhe Kou4School of Mathematics and Information Science & Technology, Hebei Normal University of Science and Technology, Qinhuangdao 066004, ChinaSchool of Mathematics and Information Science & Technology, Hebei Normal University of Science and Technology, Qinhuangdao 066004, ChinaChongqing College of Electronic Engineering, Chongqing 401331, ChinaSchool of Mathematics and Information Science & Technology, Hebei Normal University of Science and Technology, Qinhuangdao 066004, ChinaSchool of Mathematics and Information Science & Technology, Hebei Normal University of Science and Technology, Qinhuangdao 066004, ChinaIn this work, the performance of an optoelectronic time-delay reservoir computing system for performing a handwritten digit recognition task is numerically investigated, and a scheme to improve the recognition speed using multiple parallel reservoirs is proposed. By comparing four image injection methods based on a single time-delay reservoir, we find that when injecting the histograms of oriented gradient (HOG) features of the digit image, the accuracy rate (AR) is relatively high and is less affected by the offset phase. To improve the recognition speed, we construct a parallel time-delay reservoir system including multi-reservoirs, where each reservoir processes part of the HOG features of one image. Based on 6 parallel reservoirs with each reservoir possessing 100 virtual nodes, the AR can reach about 97.8%, and the reservoir processing speed can reach about 1 × 10<sup>6</sup> digits per second. Meanwhile, the parallel reservoir system shows strong robustness to the parameter mismatch between multi-reservoirs.https://www.mdpi.com/2304-6732/10/3/236handwritten digits recognitionoptoelectronictime-delay reservoir |
| spellingShingle | Dianzuo Yue Yushuang Hou Chunxia Hu Cunru Zang Yingzhe Kou Handwritten Digits Recognition Based on a Parallel Optoelectronic Time-Delay Reservoir Computing System Photonics handwritten digits recognition optoelectronic time-delay reservoir |
| title | Handwritten Digits Recognition Based on a Parallel Optoelectronic Time-Delay Reservoir Computing System |
| title_full | Handwritten Digits Recognition Based on a Parallel Optoelectronic Time-Delay Reservoir Computing System |
| title_fullStr | Handwritten Digits Recognition Based on a Parallel Optoelectronic Time-Delay Reservoir Computing System |
| title_full_unstemmed | Handwritten Digits Recognition Based on a Parallel Optoelectronic Time-Delay Reservoir Computing System |
| title_short | Handwritten Digits Recognition Based on a Parallel Optoelectronic Time-Delay Reservoir Computing System |
| title_sort | handwritten digits recognition based on a parallel optoelectronic time delay reservoir computing system |
| topic | handwritten digits recognition optoelectronic time-delay reservoir |
| url | https://www.mdpi.com/2304-6732/10/3/236 |
| work_keys_str_mv | AT dianzuoyue handwrittendigitsrecognitionbasedonaparalleloptoelectronictimedelayreservoircomputingsystem AT yushuanghou handwrittendigitsrecognitionbasedonaparalleloptoelectronictimedelayreservoircomputingsystem AT chunxiahu handwrittendigitsrecognitionbasedonaparalleloptoelectronictimedelayreservoircomputingsystem AT cunruzang handwrittendigitsrecognitionbasedonaparalleloptoelectronictimedelayreservoircomputingsystem AT yingzhekou handwrittendigitsrecognitionbasedonaparalleloptoelectronictimedelayreservoircomputingsystem |