Combining MobileNetV1 and Depthwise Separable convolution bottleneck with Expansion for classifying the freshness of fish eyes
Image classification using Convolutional Neural Network (CNN) achieves optimal performance with a particular strategy. MobileNet reduces the parameter number for learning features by switching from the standard convolution paradigm to the depthwise separable convolution (DSC) paradigm. However, ther...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-12-01
|
| Series: | Information Processing in Agriculture |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2214317322000026 |
| _version_ | 1827882328754487296 |
|---|---|
| author | Eko Prasetyo Rani Purbaningtyas Raden Dimas Adityo Nanik Suciati Chastine Fatichah |
| author_facet | Eko Prasetyo Rani Purbaningtyas Raden Dimas Adityo Nanik Suciati Chastine Fatichah |
| author_sort | Eko Prasetyo |
| collection | DOAJ |
| description | Image classification using Convolutional Neural Network (CNN) achieves optimal performance with a particular strategy. MobileNet reduces the parameter number for learning features by switching from the standard convolution paradigm to the depthwise separable convolution (DSC) paradigm. However, there are not enough features to learn for identifying the freshness of fish eyes. Furthermore, minor variances in features should not require complicated CNN architecture. In this paper, our first contribution proposed DSC Bottleneck with Expansion for learning features of the freshness of fish eyes with a Bottleneck Multiplier. The second contribution proposed Residual Transition to bridge current feature maps and skip connection feature maps to the next convolution block. The third contribution proposed MobileNetV1 Bottleneck with Expansion (MB-BE) for classifying the freshness of fish eyes. The result obtained from the Freshness of the Fish Eyes dataset shows that MB-BE outperformed other models such as original MobileNet, VGG16, Densenet, Nasnet Mobile with 63.21% accuracy. |
| first_indexed | 2024-03-12T18:56:08Z |
| format | Article |
| id | doaj.art-22b1ec4ba78a48ebb05ce84686f4dee9 |
| institution | Directory Open Access Journal |
| issn | 2214-3173 |
| language | English |
| last_indexed | 2024-03-12T18:56:08Z |
| publishDate | 2022-12-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Information Processing in Agriculture |
| spelling | doaj.art-22b1ec4ba78a48ebb05ce84686f4dee92023-08-02T06:53:21ZengElsevierInformation Processing in Agriculture2214-31732022-12-0194485496Combining MobileNetV1 and Depthwise Separable convolution bottleneck with Expansion for classifying the freshness of fish eyesEko Prasetyo0Rani Purbaningtyas1Raden Dimas Adityo2Nanik Suciati3Chastine Fatichah4Department of Informatics, Engineering Faculty, Universitas Bhayangkara Surabaya, Jl. Ahmad Yani 114, Surabaya 60234, Indonesia; Department of Informatics, Faculty of Intelligent Electrical and Informatics Technology, Institut Teknologi Sepuluh Nopember, Jl. Raya ITS, Surabaya 60111, Indonesia; Corresponding author at: Department of Informatics, Engineering Faculty, Universitas Bhayangkara Surabaya, Jl. Ahmad Yani 114, Surabaya 60234, Indonesia.Department of Informatics, Engineering Faculty, Universitas Bhayangkara Surabaya, Jl. Ahmad Yani 114, Surabaya 60234, IndonesiaDepartment of Informatics, Engineering Faculty, Universitas Bhayangkara Surabaya, Jl. Ahmad Yani 114, Surabaya 60234, IndonesiaDepartment of Informatics, Faculty of Intelligent Electrical and Informatics Technology, Institut Teknologi Sepuluh Nopember, Jl. Raya ITS, Surabaya 60111, IndonesiaDepartment of Informatics, Faculty of Intelligent Electrical and Informatics Technology, Institut Teknologi Sepuluh Nopember, Jl. Raya ITS, Surabaya 60111, IndonesiaImage classification using Convolutional Neural Network (CNN) achieves optimal performance with a particular strategy. MobileNet reduces the parameter number for learning features by switching from the standard convolution paradigm to the depthwise separable convolution (DSC) paradigm. However, there are not enough features to learn for identifying the freshness of fish eyes. Furthermore, minor variances in features should not require complicated CNN architecture. In this paper, our first contribution proposed DSC Bottleneck with Expansion for learning features of the freshness of fish eyes with a Bottleneck Multiplier. The second contribution proposed Residual Transition to bridge current feature maps and skip connection feature maps to the next convolution block. The third contribution proposed MobileNetV1 Bottleneck with Expansion (MB-BE) for classifying the freshness of fish eyes. The result obtained from the Freshness of the Fish Eyes dataset shows that MB-BE outperformed other models such as original MobileNet, VGG16, Densenet, Nasnet Mobile with 63.21% accuracy.http://www.sciencedirect.com/science/article/pii/S2214317322000026Depthwise separable convolutionBottleneckClassificationFreshnessFish eyeResidual transition |
| spellingShingle | Eko Prasetyo Rani Purbaningtyas Raden Dimas Adityo Nanik Suciati Chastine Fatichah Combining MobileNetV1 and Depthwise Separable convolution bottleneck with Expansion for classifying the freshness of fish eyes Information Processing in Agriculture Depthwise separable convolution Bottleneck Classification Freshness Fish eye Residual transition |
| title | Combining MobileNetV1 and Depthwise Separable convolution bottleneck with Expansion for classifying the freshness of fish eyes |
| title_full | Combining MobileNetV1 and Depthwise Separable convolution bottleneck with Expansion for classifying the freshness of fish eyes |
| title_fullStr | Combining MobileNetV1 and Depthwise Separable convolution bottleneck with Expansion for classifying the freshness of fish eyes |
| title_full_unstemmed | Combining MobileNetV1 and Depthwise Separable convolution bottleneck with Expansion for classifying the freshness of fish eyes |
| title_short | Combining MobileNetV1 and Depthwise Separable convolution bottleneck with Expansion for classifying the freshness of fish eyes |
| title_sort | combining mobilenetv1 and depthwise separable convolution bottleneck with expansion for classifying the freshness of fish eyes |
| topic | Depthwise separable convolution Bottleneck Classification Freshness Fish eye Residual transition |
| url | http://www.sciencedirect.com/science/article/pii/S2214317322000026 |
| work_keys_str_mv | AT ekoprasetyo combiningmobilenetv1anddepthwiseseparableconvolutionbottleneckwithexpansionforclassifyingthefreshnessoffisheyes AT ranipurbaningtyas combiningmobilenetv1anddepthwiseseparableconvolutionbottleneckwithexpansionforclassifyingthefreshnessoffisheyes AT radendimasadityo combiningmobilenetv1anddepthwiseseparableconvolutionbottleneckwithexpansionforclassifyingthefreshnessoffisheyes AT naniksuciati combiningmobilenetv1anddepthwiseseparableconvolutionbottleneckwithexpansionforclassifyingthefreshnessoffisheyes AT chastinefatichah combiningmobilenetv1anddepthwiseseparableconvolutionbottleneckwithexpansionforclassifyingthefreshnessoffisheyes |