An Efficient Classification of Rice Variety with Quantized Neural Networks
Rice, as one of the significant grain products across the world, features a wide range of varieties in terms of usability and efficiency. It may be known with various varieties and regional names depending on the specific locations. To specify a particular rice type, different features are considere...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-05-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/12/10/2285 |
| _version_ | 1797600374754902016 |
|---|---|
| author | Mustafa Tasci Ayhan Istanbullu Selahattin Kosunalp Teodor Iliev Ivaylo Stoyanov Ivan Beloev |
| author_facet | Mustafa Tasci Ayhan Istanbullu Selahattin Kosunalp Teodor Iliev Ivaylo Stoyanov Ivan Beloev |
| author_sort | Mustafa Tasci |
| collection | DOAJ |
| description | Rice, as one of the significant grain products across the world, features a wide range of varieties in terms of usability and efficiency. It may be known with various varieties and regional names depending on the specific locations. To specify a particular rice type, different features are considered, such as shape and color. This study uses an available dataset in Turkey consisting of five different varieties: Ipsala, Arborio, Basmati, Jasmine, and Karacadag. The dataset introduces 75,000 grain images in total; each of the 5 varieties has 15,000 samples with a 256 × 256-pixel dimension. The main contribution of this paper is to create Quantized Neural Network (QNN) models to efficiently classify rice varieties with the purpose of reducing resource usage on edge devices. It is well-known that QNN is a successful method for alleviating high computational costs and power requirements in response to many Deep Learning (DL) algorithms. These advantages of the quantization process have the potential to provide an efficient environment for artificial intelligence applications on microcontroller-driven edge devices. For this purpose, we created eight different QNN networks using the MLP and Lenet-5-based deep learning models with varying quantization levels to be trained by the dataset. With the Lenet-5-based QNN network created at the W3A3 quantization level, a 99.87% classification accuracy level was achieved with only 23.1 Kb memory size used for the parameters. In addition to this tremendous benefit of memory usage, the number of billion transactions per second (GOPs) is 23 times less than similar classification studies. |
| first_indexed | 2024-03-11T03:47:13Z |
| format | Article |
| id | doaj.art-e65c44ae1ec2410588ddb7366be2d61b |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-11T03:47:13Z |
| publishDate | 2023-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-e65c44ae1ec2410588ddb7366be2d61b2023-11-18T01:10:16ZengMDPI AGElectronics2079-92922023-05-011210228510.3390/electronics12102285An Efficient Classification of Rice Variety with Quantized Neural NetworksMustafa Tasci0Ayhan Istanbullu1Selahattin Kosunalp2Teodor Iliev3Ivaylo Stoyanov4Ivan Beloev5Department of Computer Technologies, Gönen Vocational School, Bandırma Onyedi Eylül University, 10200 Bandırma, TürkiyeDepartment of Computer Engineering, Faculty of Engineering, Balıkesir University, 10145 Balıkesir, TürkiyeDepartment of Computer Technologies, Gönen Vocational School, Bandırma Onyedi Eylül University, 10200 Bandırma, TürkiyeDepartment of Telecommunication, University of Ruse, 7017 Ruse, BulgariaDepartment of Electrical and Power Engineering, University of Ruse, 7017 Ruse, BulgariaDepartment of Transport, University of Ruse, 7017 Ruse, BulgariaRice, as one of the significant grain products across the world, features a wide range of varieties in terms of usability and efficiency. It may be known with various varieties and regional names depending on the specific locations. To specify a particular rice type, different features are considered, such as shape and color. This study uses an available dataset in Turkey consisting of five different varieties: Ipsala, Arborio, Basmati, Jasmine, and Karacadag. The dataset introduces 75,000 grain images in total; each of the 5 varieties has 15,000 samples with a 256 × 256-pixel dimension. The main contribution of this paper is to create Quantized Neural Network (QNN) models to efficiently classify rice varieties with the purpose of reducing resource usage on edge devices. It is well-known that QNN is a successful method for alleviating high computational costs and power requirements in response to many Deep Learning (DL) algorithms. These advantages of the quantization process have the potential to provide an efficient environment for artificial intelligence applications on microcontroller-driven edge devices. For this purpose, we created eight different QNN networks using the MLP and Lenet-5-based deep learning models with varying quantization levels to be trained by the dataset. With the Lenet-5-based QNN network created at the W3A3 quantization level, a 99.87% classification accuracy level was achieved with only 23.1 Kb memory size used for the parameters. In addition to this tremendous benefit of memory usage, the number of billion transactions per second (GOPs) is 23 times less than similar classification studies.https://www.mdpi.com/2079-9292/12/10/2285rice classificationdeep learningQuantized Neural NetworkLeNet-5 |
| spellingShingle | Mustafa Tasci Ayhan Istanbullu Selahattin Kosunalp Teodor Iliev Ivaylo Stoyanov Ivan Beloev An Efficient Classification of Rice Variety with Quantized Neural Networks Electronics rice classification deep learning Quantized Neural Network LeNet-5 |
| title | An Efficient Classification of Rice Variety with Quantized Neural Networks |
| title_full | An Efficient Classification of Rice Variety with Quantized Neural Networks |
| title_fullStr | An Efficient Classification of Rice Variety with Quantized Neural Networks |
| title_full_unstemmed | An Efficient Classification of Rice Variety with Quantized Neural Networks |
| title_short | An Efficient Classification of Rice Variety with Quantized Neural Networks |
| title_sort | efficient classification of rice variety with quantized neural networks |
| topic | rice classification deep learning Quantized Neural Network LeNet-5 |
| url | https://www.mdpi.com/2079-9292/12/10/2285 |
| work_keys_str_mv | AT mustafatasci anefficientclassificationofricevarietywithquantizedneuralnetworks AT ayhanistanbullu anefficientclassificationofricevarietywithquantizedneuralnetworks AT selahattinkosunalp anefficientclassificationofricevarietywithquantizedneuralnetworks AT teodoriliev anefficientclassificationofricevarietywithquantizedneuralnetworks AT ivaylostoyanov anefficientclassificationofricevarietywithquantizedneuralnetworks AT ivanbeloev anefficientclassificationofricevarietywithquantizedneuralnetworks AT mustafatasci efficientclassificationofricevarietywithquantizedneuralnetworks AT ayhanistanbullu efficientclassificationofricevarietywithquantizedneuralnetworks AT selahattinkosunalp efficientclassificationofricevarietywithquantizedneuralnetworks AT teodoriliev efficientclassificationofricevarietywithquantizedneuralnetworks AT ivaylostoyanov efficientclassificationofricevarietywithquantizedneuralnetworks AT ivanbeloev efficientclassificationofricevarietywithquantizedneuralnetworks |