Publication
Real-time image detection for edge devices: a peach fruit detection application
| dc.contributor.author | Assunção, Eduardo | |
| dc.contributor.author | Gaspar, Pedro Dinis | |
| dc.contributor.author | Alibabaei, Khadijeh | |
| dc.contributor.author | Simões, M.P. | |
| dc.contributor.author | Proença, Hugo | |
| dc.contributor.author | Soares, V.N.G.J. | |
| dc.contributor.author | Caldeira, J.M.L.P. | |
| dc.date.accessioned | 2022-11-09T09:20:44Z | |
| dc.date.available | 2022-11-09T09:20:44Z | |
| dc.date.issued | 2022 | |
| dc.description.abstract | Within the scope of precision agriculture, many applications have been developed to support decision making and yield enhancement. Fruit detection has attracted considerable attention from researchers, and it can be used offline. In contrast, some applications, such as robot vision in orchards, require computer vision models to run on edge devices while performing inferences at high speed. In this area, most modern applications use an integrated graphics processing unit (GPU). In this work, we propose the use of a tensor processing unit (TPU) accelerator with a Raspberry Pi target device and the state-of-the-art, lightweight, and hardware-aware MobileDet detector model. Our contribution is the extension of the possibilities of using accelerators (the TPU) for edge devices in precision agriculture. The proposed method was evaluated using a novel dataset of peaches with three cultivars, which will be made available for further studies. The model achieved an average precision (AP) of 88.2% and a performance of 19.84 frames per second (FPS) at an image size of 640 × 480. The results obtained show that the TPU accelerator can be an excellent alternative for processing on the edge in precision agriculture. | pt_PT |
| dc.description.version | info:eu-repo/semantics/publishedVersion | pt_PT |
| dc.identifier.citation | ASSUNÇÃO, Eduardo [et al.] (2022) - Real-time image detection for edge devices: a peach fruit detection application. Future Internet. 14:11. DOI: https://doi.org/10.3390/fi14110323. | |
| dc.identifier.doi | 10.3390/fi14110323 | pt_PT |
| dc.identifier.uri | http://hdl.handle.net/10400.11/8159 | |
| dc.language.iso | eng | pt_PT |
| dc.peerreviewed | yes | pt_PT |
| dc.subject | Deep learning | pt_PT |
| dc.subject | Edge device | pt_PT |
| dc.subject | Object detection | pt_PT |
| dc.subject | Precision agriculture | pt_PT |
| dc.subject | TPU accelerator | pt_PT |
| dc.title | Real-time image detection for edge devices: a peach fruit detection application | pt_PT |
| dc.type | journal article | |
| dspace.entity.type | Publication | |
| oaire.citation.issue | 11 | pt_PT |
| oaire.citation.startPage | 323 | pt_PT |
| oaire.citation.title | Future Internet | pt_PT |
| oaire.citation.volume | 14 | pt_PT |
| person.familyName | Gaspar | |
| person.familyName | Simões | |
| person.familyName | Caldeira | |
| person.givenName | Pedro Dinis | |
| person.givenName | Maria Paula | |
| person.givenName | João | |
| person.identifier | a4GD8aoAAAAJ | |
| person.identifier.ciencia-id | 6111-9F05-2916 | |
| person.identifier.ciencia-id | 5215-A196-0362 | |
| person.identifier.ciencia-id | 5B19-E130-E382 | |
| person.identifier.ciencia-id | A91B-85B8-C27E | |
| person.identifier.orcid | 0000-0003-1691-1709 | |
| person.identifier.orcid | 0000-0002-6599-0688 | |
| person.identifier.orcid | 0000-0002-8057-5474 | |
| person.identifier.orcid | 0000-0001-5830-3790 | |
| person.identifier.rid | N-3016-2013 | |
| person.identifier.scopus-author-id | 57419570900 | |
| person.identifier.scopus-author-id | 36504886200 | |
| person.identifier.scopus-author-id | 27067580500 | |
| rcaap.rights | openAccess | pt_PT |
| rcaap.type | article | pt_PT |
| relation.isAuthorOfPublication | ebfd94b1-21cd-4670-8626-e82f2b1c3436 | |
| relation.isAuthorOfPublication | c1c2eaaf-223e-4152-9245-04303ee41d75 | |
| relation.isAuthorOfPublication | a17d4ff5-1ff3-4dcc-b180-319e7ff3961d | |
| relation.isAuthorOfPublication | 8eebc97c-5334-4f29-b7ee-71c4c436aa69 | |
| relation.isAuthorOfPublication.latestForDiscovery | ebfd94b1-21cd-4670-8626-e82f2b1c3436 |