A Fast Neural Network Based on Attention Mechanisms for Detecting Field Flat Jujube
The efficient identification of the field flat jujube is the first condition to realize its automated picking. Consequently, a lightweight algorithm of target identification based on improved YOLOv5 (you only look once) is proposed to meet the requirements of high-accuracy and low-complexity. At fir...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-05-01
|
| Series: | Agriculture |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2077-0472/12/5/717 |
| _version_ | 1797502293401141248 |
|---|---|
| author | Shilin Li Shujuan Zhang Jianxin Xue Haixia Sun Rui Ren |
| author_facet | Shilin Li Shujuan Zhang Jianxin Xue Haixia Sun Rui Ren |
| author_sort | Shilin Li |
| collection | DOAJ |
| description | The efficient identification of the field flat jujube is the first condition to realize its automated picking. Consequently, a lightweight algorithm of target identification based on improved YOLOv5 (you only look once) is proposed to meet the requirements of high-accuracy and low-complexity. At first, the proposed method solves the imbalance of data distribution by improving the methods of data enhancement. Then, to improve the accuracy of the model, we adjust the structure and the number of the Concentrated-Comprehensive Convolution Block modules in the backbone network, and introduce the attention mechanisms of Efficient Channel Attention and Coordinate Attention. On this basis, this paper makes lightweight operations by using the Deep Separable Convolution to reduce the complexity of the model. Ultimately, the Complete Intersection over Union loss function and the non-maximum suppression of Distance Intersection over Union are used to optimize the loss function and the post-processing process, respectively. The experimental results show that the mean average precision of improved network reaches 97.4%, which increases by 1.7% compared with the original YOLOv5s network; and, the parameters, floating point of operations, and model size are compressed to 35.39%, 51.27%, and 37.5% of the original network, respectively. The comparison experiments are conducted around the proposed method and the common You Only Look Once target detection algorithms. The experimental results show that the mean average precision of the proposed method is 97.4%, which is higher than the 90.7%, 91.7%, and 88.4% of the YOLOv3, YOLOv4, and YOLOx-s algorithms, and the model size decreased to 2.3%, 2.2%, and 15.7%, respectively. The improved algorithm realizes a reduction of complexity and an increase in accuracy, it can be suitable for lightweight deployment to a mobile terminal at a later stage, and it provides a certain reference for the visual detection of picking robots. |
| first_indexed | 2024-03-10T03:31:01Z |
| format | Article |
| id | doaj.art-17f5095afaa546d4b5666cdafe7e6e47 |
| institution | Directory Open Access Journal |
| issn | 2077-0472 |
| language | English |
| last_indexed | 2024-03-10T03:31:01Z |
| publishDate | 2022-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Agriculture |
| spelling | doaj.art-17f5095afaa546d4b5666cdafe7e6e472023-11-23T09:40:39ZengMDPI AGAgriculture2077-04722022-05-0112571710.3390/agriculture12050717A Fast Neural Network Based on Attention Mechanisms for Detecting Field Flat JujubeShilin Li0Shujuan Zhang1Jianxin Xue2Haixia Sun3Rui Ren4College of Agricultural Engineering, Shanxi Agricultural University, Jinzhong 030800, ChinaCollege of Agricultural Engineering, Shanxi Agricultural University, Jinzhong 030800, ChinaCollege of Agricultural Engineering, Shanxi Agricultural University, Jinzhong 030800, ChinaCollege of Agricultural Engineering, Shanxi Agricultural University, Jinzhong 030800, ChinaCollege of Agricultural Engineering, Shanxi Agricultural University, Jinzhong 030800, ChinaThe efficient identification of the field flat jujube is the first condition to realize its automated picking. Consequently, a lightweight algorithm of target identification based on improved YOLOv5 (you only look once) is proposed to meet the requirements of high-accuracy and low-complexity. At first, the proposed method solves the imbalance of data distribution by improving the methods of data enhancement. Then, to improve the accuracy of the model, we adjust the structure and the number of the Concentrated-Comprehensive Convolution Block modules in the backbone network, and introduce the attention mechanisms of Efficient Channel Attention and Coordinate Attention. On this basis, this paper makes lightweight operations by using the Deep Separable Convolution to reduce the complexity of the model. Ultimately, the Complete Intersection over Union loss function and the non-maximum suppression of Distance Intersection over Union are used to optimize the loss function and the post-processing process, respectively. The experimental results show that the mean average precision of improved network reaches 97.4%, which increases by 1.7% compared with the original YOLOv5s network; and, the parameters, floating point of operations, and model size are compressed to 35.39%, 51.27%, and 37.5% of the original network, respectively. The comparison experiments are conducted around the proposed method and the common You Only Look Once target detection algorithms. The experimental results show that the mean average precision of the proposed method is 97.4%, which is higher than the 90.7%, 91.7%, and 88.4% of the YOLOv3, YOLOv4, and YOLOx-s algorithms, and the model size decreased to 2.3%, 2.2%, and 15.7%, respectively. The improved algorithm realizes a reduction of complexity and an increase in accuracy, it can be suitable for lightweight deployment to a mobile terminal at a later stage, and it provides a certain reference for the visual detection of picking robots.https://www.mdpi.com/2077-0472/12/5/717target detectionidentifying the field flat jujubeYOLOv5convolutional neural network |
| spellingShingle | Shilin Li Shujuan Zhang Jianxin Xue Haixia Sun Rui Ren A Fast Neural Network Based on Attention Mechanisms for Detecting Field Flat Jujube Agriculture target detection identifying the field flat jujube YOLOv5 convolutional neural network |
| title | A Fast Neural Network Based on Attention Mechanisms for Detecting Field Flat Jujube |
| title_full | A Fast Neural Network Based on Attention Mechanisms for Detecting Field Flat Jujube |
| title_fullStr | A Fast Neural Network Based on Attention Mechanisms for Detecting Field Flat Jujube |
| title_full_unstemmed | A Fast Neural Network Based on Attention Mechanisms for Detecting Field Flat Jujube |
| title_short | A Fast Neural Network Based on Attention Mechanisms for Detecting Field Flat Jujube |
| title_sort | fast neural network based on attention mechanisms for detecting field flat jujube |
| topic | target detection identifying the field flat jujube YOLOv5 convolutional neural network |
| url | https://www.mdpi.com/2077-0472/12/5/717 |
| work_keys_str_mv | AT shilinli afastneuralnetworkbasedonattentionmechanismsfordetectingfieldflatjujube AT shujuanzhang afastneuralnetworkbasedonattentionmechanismsfordetectingfieldflatjujube AT jianxinxue afastneuralnetworkbasedonattentionmechanismsfordetectingfieldflatjujube AT haixiasun afastneuralnetworkbasedonattentionmechanismsfordetectingfieldflatjujube AT ruiren afastneuralnetworkbasedonattentionmechanismsfordetectingfieldflatjujube AT shilinli fastneuralnetworkbasedonattentionmechanismsfordetectingfieldflatjujube AT shujuanzhang fastneuralnetworkbasedonattentionmechanismsfordetectingfieldflatjujube AT jianxinxue fastneuralnetworkbasedonattentionmechanismsfordetectingfieldflatjujube AT haixiasun fastneuralnetworkbasedonattentionmechanismsfordetectingfieldflatjujube AT ruiren fastneuralnetworkbasedonattentionmechanismsfordetectingfieldflatjujube |