Implementing a Compression Technique on the Progressive Contextual Excitation Network for Smart Farming Applications
The utilization of computer vision in smart farming is becoming a trend in constructing an agricultural automation scheme. Deep learning (DL) is famous for the accurate approach to addressing the tasks in computer vision, such as object detection and image classification. The superiority of the deep...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-12-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/24/9717 |
| _version_ | 1797455327594020864 |
|---|---|
| author | Setya Widyawan Prakosa Jenq-Shiou Leu He-Yen Hsieh Cries Avian Chia-Hung Bai Stanislav Vítek |
| author_facet | Setya Widyawan Prakosa Jenq-Shiou Leu He-Yen Hsieh Cries Avian Chia-Hung Bai Stanislav Vítek |
| author_sort | Setya Widyawan Prakosa |
| collection | DOAJ |
| description | The utilization of computer vision in smart farming is becoming a trend in constructing an agricultural automation scheme. Deep learning (DL) is famous for the accurate approach to addressing the tasks in computer vision, such as object detection and image classification. The superiority of the deep learning model on the smart farming application, called Progressive Contextual Excitation Network (PCENet), has also been studied in our recent study to classify cocoa bean images. However, the assessment of the computational time on the PCENet model shows that the original model is only 0.101s or 9.9 FPS on the Jetson Nano as the edge platform. Therefore, this research demonstrates the compression technique to accelerate the PCENet model using pruning filters. From our experiment, we can accelerate the current model and achieve 16.7 FPS assessed in the Jetson Nano. Moreover, the accuracy of the compressed model can be maintained at 86.1%, while the original model is 86.8%. In addition, our approach is more accurate than ResNet18 as the state-of-the-art only reaches 82.7%. The assessment using the corn leaf disease dataset indicates that the compressed model can achieve an accuracy of 97.5%, while the accuracy of the original PCENet is 97.7%. |
| first_indexed | 2024-03-09T15:52:49Z |
| format | Article |
| id | doaj.art-b453ce2556fe48a198fe651b21a1ce15 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T15:52:49Z |
| publishDate | 2022-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-b453ce2556fe48a198fe651b21a1ce152023-11-24T17:54:11ZengMDPI AGSensors1424-82202022-12-012224971710.3390/s22249717Implementing a Compression Technique on the Progressive Contextual Excitation Network for Smart Farming ApplicationsSetya Widyawan Prakosa0Jenq-Shiou Leu1He-Yen Hsieh2Cries Avian3Chia-Hung Bai4Stanislav Vítek5Department of Electronic and Computer Engineering (ECE), National Taiwan University of Science and Technology, Taipei 106335, TaiwanDepartment of Electronic and Computer Engineering (ECE), National Taiwan University of Science and Technology, Taipei 106335, TaiwanDepartment of Electronic and Computer Engineering (ECE), National Taiwan University of Science and Technology, Taipei 106335, TaiwanDepartment of Electronic and Computer Engineering (ECE), National Taiwan University of Science and Technology, Taipei 106335, TaiwanDepartment of Electronic and Computer Engineering (ECE), National Taiwan University of Science and Technology, Taipei 106335, TaiwanFaculty of Electrical Engineering, Czech Technical University in Prague, Technicka 2, 16627 Prague, Czech RepublicThe utilization of computer vision in smart farming is becoming a trend in constructing an agricultural automation scheme. Deep learning (DL) is famous for the accurate approach to addressing the tasks in computer vision, such as object detection and image classification. The superiority of the deep learning model on the smart farming application, called Progressive Contextual Excitation Network (PCENet), has also been studied in our recent study to classify cocoa bean images. However, the assessment of the computational time on the PCENet model shows that the original model is only 0.101s or 9.9 FPS on the Jetson Nano as the edge platform. Therefore, this research demonstrates the compression technique to accelerate the PCENet model using pruning filters. From our experiment, we can accelerate the current model and achieve 16.7 FPS assessed in the Jetson Nano. Moreover, the accuracy of the compressed model can be maintained at 86.1%, while the original model is 86.8%. In addition, our approach is more accurate than ResNet18 as the state-of-the-art only reaches 82.7%. The assessment using the corn leaf disease dataset indicates that the compressed model can achieve an accuracy of 97.5%, while the accuracy of the original PCENet is 97.7%.https://www.mdpi.com/1424-8220/22/24/9717deep learningmodel compressionprogressive contextual excitationpruning filters |
| spellingShingle | Setya Widyawan Prakosa Jenq-Shiou Leu He-Yen Hsieh Cries Avian Chia-Hung Bai Stanislav Vítek Implementing a Compression Technique on the Progressive Contextual Excitation Network for Smart Farming Applications Sensors deep learning model compression progressive contextual excitation pruning filters |
| title | Implementing a Compression Technique on the Progressive Contextual Excitation Network for Smart Farming Applications |
| title_full | Implementing a Compression Technique on the Progressive Contextual Excitation Network for Smart Farming Applications |
| title_fullStr | Implementing a Compression Technique on the Progressive Contextual Excitation Network for Smart Farming Applications |
| title_full_unstemmed | Implementing a Compression Technique on the Progressive Contextual Excitation Network for Smart Farming Applications |
| title_short | Implementing a Compression Technique on the Progressive Contextual Excitation Network for Smart Farming Applications |
| title_sort | implementing a compression technique on the progressive contextual excitation network for smart farming applications |
| topic | deep learning model compression progressive contextual excitation pruning filters |
| url | https://www.mdpi.com/1424-8220/22/24/9717 |
| work_keys_str_mv | AT setyawidyawanprakosa implementingacompressiontechniqueontheprogressivecontextualexcitationnetworkforsmartfarmingapplications AT jenqshiouleu implementingacompressiontechniqueontheprogressivecontextualexcitationnetworkforsmartfarmingapplications AT heyenhsieh implementingacompressiontechniqueontheprogressivecontextualexcitationnetworkforsmartfarmingapplications AT criesavian implementingacompressiontechniqueontheprogressivecontextualexcitationnetworkforsmartfarmingapplications AT chiahungbai implementingacompressiontechniqueontheprogressivecontextualexcitationnetworkforsmartfarmingapplications AT stanislavvitek implementingacompressiontechniqueontheprogressivecontextualexcitationnetworkforsmartfarmingapplications |