Multi-Scale Convolution-Capsule Network for Crop Insect Pest Recognition
Accurate crop insect pest identification in fields is useful to control pests and beneficial to agricultural yield and quality. However, it is a difficult and challenging problem due to the crop insect pests being small with various sizes, postures, shapes, and disorganized backgrounds. Multi-scale...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-05-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/11/10/1630 |
| _version_ | 1797500253963812864 |
|---|---|
| author | Cong Xu Changqing Yu Shanwen Zhang Xuqi Wang |
| author_facet | Cong Xu Changqing Yu Shanwen Zhang Xuqi Wang |
| author_sort | Cong Xu |
| collection | DOAJ |
| description | Accurate crop insect pest identification in fields is useful to control pests and beneficial to agricultural yield and quality. However, it is a difficult and challenging problem due to the crop insect pests being small with various sizes, postures, shapes, and disorganized backgrounds. Multi-scale convolution-capsule network (MSCCN) is constructed for crop insect pest identification. It consists of a multi-scale convolution module, capsule network (CapsNet) module, and SoftMax classification module. Multi-scale convolution is used to extract the multi-scale discriminative features, CapsNet is employed to encode the hierarchical structure of the size-variant insect pests in the crop images, and Softmax is adopted for insect pest identification. MSCCN combines the advantages of convolutional neural network (CNN), CapsNet, and multi-scale CNN, and can learn multi-scale robust features from pest images of different shapes and sizes for pest recognition and identify various morphed pests. Experimental results on the crop pest image dataset show that this method has a good recognition rate of 91.4%. |
| first_indexed | 2024-03-10T03:59:16Z |
| format | Article |
| id | doaj.art-b5c048130c3a496fbef71d6fb0a49745 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-10T03:59:16Z |
| publishDate | 2022-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-b5c048130c3a496fbef71d6fb0a497452023-11-23T10:48:01ZengMDPI AGElectronics2079-92922022-05-011110163010.3390/electronics11101630Multi-Scale Convolution-Capsule Network for Crop Insect Pest RecognitionCong Xu0Changqing Yu1Shanwen Zhang2Xuqi Wang3College of Information Engineering, Xijing University, Xi’an 710123, ChinaCollege of Information Engineering, Xijing University, Xi’an 710123, ChinaCollege of Information Engineering, Xijing University, Xi’an 710123, ChinaCollege of Information Engineering, Xijing University, Xi’an 710123, ChinaAccurate crop insect pest identification in fields is useful to control pests and beneficial to agricultural yield and quality. However, it is a difficult and challenging problem due to the crop insect pests being small with various sizes, postures, shapes, and disorganized backgrounds. Multi-scale convolution-capsule network (MSCCN) is constructed for crop insect pest identification. It consists of a multi-scale convolution module, capsule network (CapsNet) module, and SoftMax classification module. Multi-scale convolution is used to extract the multi-scale discriminative features, CapsNet is employed to encode the hierarchical structure of the size-variant insect pests in the crop images, and Softmax is adopted for insect pest identification. MSCCN combines the advantages of convolutional neural network (CNN), CapsNet, and multi-scale CNN, and can learn multi-scale robust features from pest images of different shapes and sizes for pest recognition and identify various morphed pests. Experimental results on the crop pest image dataset show that this method has a good recognition rate of 91.4%.https://www.mdpi.com/2079-9292/11/10/1630crop insect pest identificationconvolutional neural network (CNN)capsule network (CapsNet)multi-scale convolution-capsule network (MSCCN) |
| spellingShingle | Cong Xu Changqing Yu Shanwen Zhang Xuqi Wang Multi-Scale Convolution-Capsule Network for Crop Insect Pest Recognition Electronics crop insect pest identification convolutional neural network (CNN) capsule network (CapsNet) multi-scale convolution-capsule network (MSCCN) |
| title | Multi-Scale Convolution-Capsule Network for Crop Insect Pest Recognition |
| title_full | Multi-Scale Convolution-Capsule Network for Crop Insect Pest Recognition |
| title_fullStr | Multi-Scale Convolution-Capsule Network for Crop Insect Pest Recognition |
| title_full_unstemmed | Multi-Scale Convolution-Capsule Network for Crop Insect Pest Recognition |
| title_short | Multi-Scale Convolution-Capsule Network for Crop Insect Pest Recognition |
| title_sort | multi scale convolution capsule network for crop insect pest recognition |
| topic | crop insect pest identification convolutional neural network (CNN) capsule network (CapsNet) multi-scale convolution-capsule network (MSCCN) |
| url | https://www.mdpi.com/2079-9292/11/10/1630 |
| work_keys_str_mv | AT congxu multiscaleconvolutioncapsulenetworkforcropinsectpestrecognition AT changqingyu multiscaleconvolutioncapsulenetworkforcropinsectpestrecognition AT shanwenzhang multiscaleconvolutioncapsulenetworkforcropinsectpestrecognition AT xuqiwang multiscaleconvolutioncapsulenetworkforcropinsectpestrecognition |