A Method for Obtaining the Number of Maize Seedlings Based on the Improved YOLOv4 Lightweight Neural Network
Obtaining the number of plants is the key to evaluating the effect of maize mechanical sowing, and is also a reference for subsequent statistics on the number of missing seedlings. When the existing model is used for plant number detection, the recognition accuracy is low, the model parameters are l...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-10-01
|
| Series: | Agriculture |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2077-0472/12/10/1679 |
| _version_ | 1797476264798322688 |
|---|---|
| author | Jiaxin Gao Feng Tan Jiapeng Cui Bo Ma |
| author_facet | Jiaxin Gao Feng Tan Jiapeng Cui Bo Ma |
| author_sort | Jiaxin Gao |
| collection | DOAJ |
| description | Obtaining the number of plants is the key to evaluating the effect of maize mechanical sowing, and is also a reference for subsequent statistics on the number of missing seedlings. When the existing model is used for plant number detection, the recognition accuracy is low, the model parameters are large, and the single recognition area is small. This study proposes a method for detecting the number of maize seedlings based on an improved You Only Look Once version 4 (YOLOv4) lightweight neural network. First, the method uses the improved Ghostnet as the model feature extraction network, and successively introduces the attention mechanism and k-means clustering algorithm into the model, thereby improving the detection accuracy of the number of maize seedlings. Second, using depthwise separable convolutions instead of ordinary convolutions makes the network more lightweight. Finally, the multi-scale feature fusion network structure is improved to further reduce the total number of model parameters, pre-training with transfer learning to obtain the optimal model for prediction on the test set. The experimental results show that the harmonic mean, recall rate, average precision and accuracy rate of the model on all test sets are 0.95%, 94.02%, 97.03% and 96.25%, respectively, the model network parameters are 18.793 M, the model size is 71.690 MB, and frames per second (FPS) is 22.92. The research results show that the model has high recognition accuracy, fast recognition speed, and low model complexity, which can provide technical support for corn management at the seedling stage. |
| first_indexed | 2024-03-09T20:55:31Z |
| format | Article |
| id | doaj.art-7724bf10877048cdb42babcabb1b991e |
| institution | Directory Open Access Journal |
| issn | 2077-0472 |
| language | English |
| last_indexed | 2024-03-09T20:55:31Z |
| publishDate | 2022-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Agriculture |
| spelling | doaj.art-7724bf10877048cdb42babcabb1b991e2023-11-23T22:22:21ZengMDPI AGAgriculture2077-04722022-10-011210167910.3390/agriculture12101679A Method for Obtaining the Number of Maize Seedlings Based on the Improved YOLOv4 Lightweight Neural NetworkJiaxin Gao0Feng Tan1Jiapeng Cui2Bo Ma3College of Electrical and Information, Heilongjiang Bayi Agricultural University, Daqing 163000, ChinaCollege of Electrical and Information, Heilongjiang Bayi Agricultural University, Daqing 163000, ChinaCollege of Agricultural Engineering, Heilongjiang Bayi Agricultural University, Daqing 163000, ChinaQiqihar Branch of Heilongjiang Academy of Agricultural Sciences, Qiqihar 161006, ChinaObtaining the number of plants is the key to evaluating the effect of maize mechanical sowing, and is also a reference for subsequent statistics on the number of missing seedlings. When the existing model is used for plant number detection, the recognition accuracy is low, the model parameters are large, and the single recognition area is small. This study proposes a method for detecting the number of maize seedlings based on an improved You Only Look Once version 4 (YOLOv4) lightweight neural network. First, the method uses the improved Ghostnet as the model feature extraction network, and successively introduces the attention mechanism and k-means clustering algorithm into the model, thereby improving the detection accuracy of the number of maize seedlings. Second, using depthwise separable convolutions instead of ordinary convolutions makes the network more lightweight. Finally, the multi-scale feature fusion network structure is improved to further reduce the total number of model parameters, pre-training with transfer learning to obtain the optimal model for prediction on the test set. The experimental results show that the harmonic mean, recall rate, average precision and accuracy rate of the model on all test sets are 0.95%, 94.02%, 97.03% and 96.25%, respectively, the model network parameters are 18.793 M, the model size is 71.690 MB, and frames per second (FPS) is 22.92. The research results show that the model has high recognition accuracy, fast recognition speed, and low model complexity, which can provide technical support for corn management at the seedling stage.https://www.mdpi.com/2077-0472/12/10/1679maize seedlingsdetectionYOLOv4improved Ghostnetk-means clusteringattention mechanism |
| spellingShingle | Jiaxin Gao Feng Tan Jiapeng Cui Bo Ma A Method for Obtaining the Number of Maize Seedlings Based on the Improved YOLOv4 Lightweight Neural Network Agriculture maize seedlings detection YOLOv4 improved Ghostnet k-means clustering attention mechanism |
| title | A Method for Obtaining the Number of Maize Seedlings Based on the Improved YOLOv4 Lightweight Neural Network |
| title_full | A Method for Obtaining the Number of Maize Seedlings Based on the Improved YOLOv4 Lightweight Neural Network |
| title_fullStr | A Method for Obtaining the Number of Maize Seedlings Based on the Improved YOLOv4 Lightweight Neural Network |
| title_full_unstemmed | A Method for Obtaining the Number of Maize Seedlings Based on the Improved YOLOv4 Lightweight Neural Network |
| title_short | A Method for Obtaining the Number of Maize Seedlings Based on the Improved YOLOv4 Lightweight Neural Network |
| title_sort | method for obtaining the number of maize seedlings based on the improved yolov4 lightweight neural network |
| topic | maize seedlings detection YOLOv4 improved Ghostnet k-means clustering attention mechanism |
| url | https://www.mdpi.com/2077-0472/12/10/1679 |
| work_keys_str_mv | AT jiaxingao amethodforobtainingthenumberofmaizeseedlingsbasedontheimprovedyolov4lightweightneuralnetwork AT fengtan amethodforobtainingthenumberofmaizeseedlingsbasedontheimprovedyolov4lightweightneuralnetwork AT jiapengcui amethodforobtainingthenumberofmaizeseedlingsbasedontheimprovedyolov4lightweightneuralnetwork AT boma amethodforobtainingthenumberofmaizeseedlingsbasedontheimprovedyolov4lightweightneuralnetwork AT jiaxingao methodforobtainingthenumberofmaizeseedlingsbasedontheimprovedyolov4lightweightneuralnetwork AT fengtan methodforobtainingthenumberofmaizeseedlingsbasedontheimprovedyolov4lightweightneuralnetwork AT jiapengcui methodforobtainingthenumberofmaizeseedlingsbasedontheimprovedyolov4lightweightneuralnetwork AT boma methodforobtainingthenumberofmaizeseedlingsbasedontheimprovedyolov4lightweightneuralnetwork |