SeedSortNet: a rapid and highly effificient lightweight CNN based on visual attention for seed sorting
Seed purity directly affects the quality of seed breeding and subsequent processing products. Seed sorting based on machine vision provides an effective solution to this problem. The deep learning technology, particularly convolutional neural networks (CNNs), have exhibited impressive performance in...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
PeerJ Inc.
2021-08-01
|
| Series: | PeerJ Computer Science |
| Subjects: | |
| Online Access: | https://peerj.com/articles/cs-639.pdf |
| _version_ | 1818885102662320128 |
|---|---|
| author | Chunlei Li Huanyu Li Zhoufeng Liu Bicao Li Yun Huang |
| author_facet | Chunlei Li Huanyu Li Zhoufeng Liu Bicao Li Yun Huang |
| author_sort | Chunlei Li |
| collection | DOAJ |
| description | Seed purity directly affects the quality of seed breeding and subsequent processing products. Seed sorting based on machine vision provides an effective solution to this problem. The deep learning technology, particularly convolutional neural networks (CNNs), have exhibited impressive performance in image recognition and classification, and have been proven applicable in seed sorting. However the huge computational complexity and massive storage requirements make it a great challenge to deploy them in real-time applications, especially on devices with limited resources. In this study, a rapid and highly efficient lightweight CNN based on visual attention, namely SeedSortNet, is proposed for seed sorting. First, a dual-branch lightweight feature extraction module Shield-block is elaborately designed by performing identity mapping, spatial transformation at higher dimensions and different receptive field modeling, and thus it can alleviate information loss and effectively characterize the multi-scale feature while utilizing fewer parameters and lower computational complexity. In the down-sampling layer, the traditional MaxPool is replaced as MaxBlurPool to improve the shift-invariant of the network. Also, an extremely lightweight sub-feature space attention module (SFSAM) is presented to selectively emphasize fine-grained features and suppress the interference of complex backgrounds. Experimental results show that SeedSortNet achieves the accuracy rates of 97.33% and 99.56% on the maize seed dataset and sunflower seed dataset, respectively, and outperforms the mainstream lightweight networks (MobileNetv2, ShuffleNetv2, etc.) at similar computational costs, with only 0.400M parameters (vs. 4.06M, 5.40M). |
| first_indexed | 2024-12-19T16:00:06Z |
| format | Article |
| id | doaj.art-a8fae9c668a04ced8d3aa97cad41c90f |
| institution | Directory Open Access Journal |
| issn | 2376-5992 |
| language | English |
| last_indexed | 2024-12-19T16:00:06Z |
| publishDate | 2021-08-01 |
| publisher | PeerJ Inc. |
| record_format | Article |
| series | PeerJ Computer Science |
| spelling | doaj.art-a8fae9c668a04ced8d3aa97cad41c90f2022-12-21T20:14:58ZengPeerJ Inc.PeerJ Computer Science2376-59922021-08-017e63910.7717/peerj-cs.639SeedSortNet: a rapid and highly effificient lightweight CNN based on visual attention for seed sortingChunlei Li0Huanyu Li1Zhoufeng Liu2Bicao Li3Yun Huang4School of Electrical and Information Engineering, Zhongyuan University of Technology, Zhengzhou, Henan, ChinaSchool of Electrical and Information Engineering, Zhongyuan University of Technology, Zhengzhou, Henan, ChinaSchool of Electrical and Information Engineering, Zhongyuan University of Technology, Zhengzhou, Henan, ChinaSchool of Electrical and Information Engineering, Zhongyuan University of Technology, Zhengzhou, Henan, ChinaXiamen Vision+ Technology Co. Ltd, Xiamen, Fujian, ChinaSeed purity directly affects the quality of seed breeding and subsequent processing products. Seed sorting based on machine vision provides an effective solution to this problem. The deep learning technology, particularly convolutional neural networks (CNNs), have exhibited impressive performance in image recognition and classification, and have been proven applicable in seed sorting. However the huge computational complexity and massive storage requirements make it a great challenge to deploy them in real-time applications, especially on devices with limited resources. In this study, a rapid and highly efficient lightweight CNN based on visual attention, namely SeedSortNet, is proposed for seed sorting. First, a dual-branch lightweight feature extraction module Shield-block is elaborately designed by performing identity mapping, spatial transformation at higher dimensions and different receptive field modeling, and thus it can alleviate information loss and effectively characterize the multi-scale feature while utilizing fewer parameters and lower computational complexity. In the down-sampling layer, the traditional MaxPool is replaced as MaxBlurPool to improve the shift-invariant of the network. Also, an extremely lightweight sub-feature space attention module (SFSAM) is presented to selectively emphasize fine-grained features and suppress the interference of complex backgrounds. Experimental results show that SeedSortNet achieves the accuracy rates of 97.33% and 99.56% on the maize seed dataset and sunflower seed dataset, respectively, and outperforms the mainstream lightweight networks (MobileNetv2, ShuffleNetv2, etc.) at similar computational costs, with only 0.400M parameters (vs. 4.06M, 5.40M).https://peerj.com/articles/cs-639.pdfSeed sortingComputer visionLightweight CNNAttention mechanism |
| spellingShingle | Chunlei Li Huanyu Li Zhoufeng Liu Bicao Li Yun Huang SeedSortNet: a rapid and highly effificient lightweight CNN based on visual attention for seed sorting PeerJ Computer Science Seed sorting Computer vision Lightweight CNN Attention mechanism |
| title | SeedSortNet: a rapid and highly effificient lightweight CNN based on visual attention for seed sorting |
| title_full | SeedSortNet: a rapid and highly effificient lightweight CNN based on visual attention for seed sorting |
| title_fullStr | SeedSortNet: a rapid and highly effificient lightweight CNN based on visual attention for seed sorting |
| title_full_unstemmed | SeedSortNet: a rapid and highly effificient lightweight CNN based on visual attention for seed sorting |
| title_short | SeedSortNet: a rapid and highly effificient lightweight CNN based on visual attention for seed sorting |
| title_sort | seedsortnet a rapid and highly effificient lightweight cnn based on visual attention for seed sorting |
| topic | Seed sorting Computer vision Lightweight CNN Attention mechanism |
| url | https://peerj.com/articles/cs-639.pdf |
| work_keys_str_mv | AT chunleili seedsortnetarapidandhighlyeffificientlightweightcnnbasedonvisualattentionforseedsorting AT huanyuli seedsortnetarapidandhighlyeffificientlightweightcnnbasedonvisualattentionforseedsorting AT zhoufengliu seedsortnetarapidandhighlyeffificientlightweightcnnbasedonvisualattentionforseedsorting AT bicaoli seedsortnetarapidandhighlyeffificientlightweightcnnbasedonvisualattentionforseedsorting AT yunhuang seedsortnetarapidandhighlyeffificientlightweightcnnbasedonvisualattentionforseedsorting |