Wild Mushroom Classification Based on Improved MobileViT Deep Learning
Wild mushrooms are not only tasty but also rich in nutritional value, but it is difficult for non-specialists to distinguish poisonous wild mushrooms accurately. Given the frequent occurrence of wild mushroom poisoning, we propose a new multidimensional feature fusion attention network (M-ViT) combi...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-04-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/13/8/4680 |
| _version_ | 1797606564620664832 |
|---|---|
| author | Youju Peng Yang Xu Jin Shi Shiyi Jiang |
| author_facet | Youju Peng Yang Xu Jin Shi Shiyi Jiang |
| author_sort | Youju Peng |
| collection | DOAJ |
| description | Wild mushrooms are not only tasty but also rich in nutritional value, but it is difficult for non-specialists to distinguish poisonous wild mushrooms accurately. Given the frequent occurrence of wild mushroom poisoning, we propose a new multidimensional feature fusion attention network (M-ViT) combining convolutional networks (ConvNets) and attention networks to compensate for the deficiency of pure ConvNets and pure attention networks. First, we introduced an attention mechanism Squeeze and Excitation (SE) module in the MobilenetV2 (MV2) structure of the network to enhance the representation of picture channels. Then, we designed a Multidimension Attention module (MDA) to guide the network to thoroughly learn and utilize local and global features through short connections. Moreover, using the Atrous Spatial Pyramid Pooling (ASPP) module to obtain longer distance relations, we fused the model features from different layers, and used the obtained joint features for wild mushroom classification. We validated the model on two datasets, mushroom and MO106, and the results showed that M-ViT performed the best on the two test datasets, with accurate dimensions of 96.21% and 91.83%, respectively. We compared the performance of our method with that of more advanced ConvNets and attention networks (Transformer), and our method achieved good results. |
| first_indexed | 2024-03-11T05:16:59Z |
| format | Article |
| id | doaj.art-80e024dcfeaa435095080a26f6901254 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-11T05:16:59Z |
| publishDate | 2023-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-80e024dcfeaa435095080a26f69012542023-11-17T18:07:39ZengMDPI AGApplied Sciences2076-34172023-04-01138468010.3390/app13084680Wild Mushroom Classification Based on Improved MobileViT Deep LearningYouju Peng0Yang Xu1Jin Shi2Shiyi Jiang3College of Big Data and Information Engineering, Guizhou University, Guiyang 550025, ChinaCollege of Big Data and Information Engineering, Guizhou University, Guiyang 550025, ChinaCollege of Big Data and Information Engineering, Guizhou University, Guiyang 550025, ChinaCollege of Big Data and Information Engineering, Guizhou University, Guiyang 550025, ChinaWild mushrooms are not only tasty but also rich in nutritional value, but it is difficult for non-specialists to distinguish poisonous wild mushrooms accurately. Given the frequent occurrence of wild mushroom poisoning, we propose a new multidimensional feature fusion attention network (M-ViT) combining convolutional networks (ConvNets) and attention networks to compensate for the deficiency of pure ConvNets and pure attention networks. First, we introduced an attention mechanism Squeeze and Excitation (SE) module in the MobilenetV2 (MV2) structure of the network to enhance the representation of picture channels. Then, we designed a Multidimension Attention module (MDA) to guide the network to thoroughly learn and utilize local and global features through short connections. Moreover, using the Atrous Spatial Pyramid Pooling (ASPP) module to obtain longer distance relations, we fused the model features from different layers, and used the obtained joint features for wild mushroom classification. We validated the model on two datasets, mushroom and MO106, and the results showed that M-ViT performed the best on the two test datasets, with accurate dimensions of 96.21% and 91.83%, respectively. We compared the performance of our method with that of more advanced ConvNets and attention networks (Transformer), and our method achieved good results.https://www.mdpi.com/2076-3417/13/8/4680attention mechanismfine-grainedfeature fusionMobileViT |
| spellingShingle | Youju Peng Yang Xu Jin Shi Shiyi Jiang Wild Mushroom Classification Based on Improved MobileViT Deep Learning Applied Sciences attention mechanism fine-grained feature fusion MobileViT |
| title | Wild Mushroom Classification Based on Improved MobileViT Deep Learning |
| title_full | Wild Mushroom Classification Based on Improved MobileViT Deep Learning |
| title_fullStr | Wild Mushroom Classification Based on Improved MobileViT Deep Learning |
| title_full_unstemmed | Wild Mushroom Classification Based on Improved MobileViT Deep Learning |
| title_short | Wild Mushroom Classification Based on Improved MobileViT Deep Learning |
| title_sort | wild mushroom classification based on improved mobilevit deep learning |
| topic | attention mechanism fine-grained feature fusion MobileViT |
| url | https://www.mdpi.com/2076-3417/13/8/4680 |
| work_keys_str_mv | AT youjupeng wildmushroomclassificationbasedonimprovedmobilevitdeeplearning AT yangxu wildmushroomclassificationbasedonimprovedmobilevitdeeplearning AT jinshi wildmushroomclassificationbasedonimprovedmobilevitdeeplearning AT shiyijiang wildmushroomclassificationbasedonimprovedmobilevitdeeplearning |