Correction: Recognition of mulberry leaf diseases based on multi-scale residual network fusion SENet.

[This corrects the article DOI: 10.1371/journal.pone.0298700.].

Recognition of mulberry leaf diseases based on multi-scale residual network fusion SENet.

Silkworms are insects with important economic value, and mulberry leaves are the food of silkworms. The quality and quantity of mulberry leaves have a direct impact on cocooning. Mulberry leaves are often infected with various diseases during the growth process. Because of the subjectivity and time-consuming problems in artificial identification of mulberry leaf diseases. In this work, a multi-scale residual network fusion Squeeze-and-Excitation Networks (SENet) is proposed for mulberry leaf disease recognition. The mulberry leaf disease dataset was expanded by performing operations such as brightness enhancement, contrast enhancement, level flipping and adding Gaussian noise. Multi-scale convolution was used instead of the traditional single-scale convolution, allowing the network to be widened to obtain more feature information and avoiding the overfitting phenomenon caused by the network piling up too deep. SENet was introduced into the residual network to enhance the extraction of key feature information of the model, thus improving the recognition accuracy of the model. The experimental results showed that the method proposed in this paper can effectively improve the recognition performance of the model. The recognition accuracy reached 98.72%. The recall and F1 score were 98.73% and 98.72% respectively. Compared with some other models, this model has better recognition effect and can provide technical reference for intelligent mulberry leaf disease detection.

Correction: Sugarcane stem node identification algorithm based on improved YOLOv5.

[This corrects the article DOI: 10.1371/journal.pone.0295565.].

Sugarcane stem node identification algorithm based on improved YOLOv5.

Identification of sugarcane stem nodes is generally dependent on high-performance recognition equipment in sugarcane seed pre-cutting machines and inefficient. Accordingly, this study proposes a novel lightweight architecture for the detection of sugarcane stem nodes based on the YOLOv5 framework, named G-YOLOv5s-SS. Firstly, the study removes the CBS and C3 structures at the end of the backbone network to fully utilize shallow-level feature information. This enhances the detection performance of sugarcane stem nodes. Simultaneously, it eliminates the 32 times down-sampled branches in the neck structure and the 20x20 detection heads at the prediction end, reducing model complexity. Secondly, a Ghost lightweight module is introduced to replace the conventional convolution module in the BottleNeck structure, further reducing the model's complexity. Finally, the study incorporates the SimAM attention mechanism to enhance the extraction of sugarcane stem node features without introducing additional parameters. This improvement aims to enhance recognition accuracy, compensating for any loss in precision due to lightweight modifications. The experimental results showed that the average precision of the improved network for sugarcane stem node identification reached 97.6%, which was 0.6% higher than that of the YOLOv5 baseline network. Meanwhile, a model size of 2.6MB, 1,129,340 parameters, and 7.2G FLOPs, representing respective reductions of 82%, 84%, and 54.4%. Compared with mainstream one-stage target detection algorithms such as YOLOv4-tiny, YOLOv4, YOLOv5n, YOLOv6n, YOLOv6s, YOLOv7-tiny, and YOLOv7, G-YOLOv5s-SS achieved respective average precision improvements of 12.9%, 5.07%, 3.6%, 2.1%, 1.2%, 3%, and 0.4% in sugarcane stem nodes recognition. Meanwhile, the model size was compressed by 88.9%, 98.9%, 33.3%, 72%, 92.9%, 78.8% and 96.3%, respectively. Compared with similar studies, G-YOLOv5s-SS not only enhanced recognition accuracy but also considered model size, demonstrating an overall excellent performance that aligns with the requirements of sugarcane seed pre-cutting machines.

Height detection of crop divider toes of sugarcane harvester based on Kalman adaptive adjustment.

Crop divider toes are an essential device of sugarcane harvester. Moving forward against the ground is a critical way to improve the harvesting rate of lodged sugarcane. Height detection is the basis for precise control of crop divider toes moving forward against the ground. Due to the current problem of operating difficulties in manually adjusting the height of crop divider, a height detection system based on a millimeter wave radar sensor was designed to detect the height of crop divider toes from the ground. This paper proposed a height detection method of crop divider toes for sugarcane harvester based on Kalman adaptive adjustment. The data measured by the sensor was pretreated to determine whether the height had changed. Reset the Kalman filter and adjust the parameters when changes occur to improve the filter response speed and ranging accuracy. To adapt to the scenario of quickly adjusting the height of crop divider during the traveling process of sugarcane harvester. A one-way ANOVA test and a two-way ANOVA test were conducted on a simulated test platform. The results of the one-way ANOVA test showed that both forward speed and vegetation cover thickness had a significant effect on height detection accuracy. The results of the two-way ANOVA test showed that the interaction of forward speed and vegetation cover thickness did not have a significant effect on ranging accuracy. It was verified through experiments that both the ranging accuracy and the response speed of height change were significantly improved after the processing of the method in this paper. The mean square error after processing was lower than 2.5 cm. The feasibility of the height detection system was verified by field trials. The results of this study will provide a reference for the design of automatic elevation of crop divider.

Application of improved YOLOv7-based sugarcane stem node recognition algorithm in complex environments.

Introduction: Sugarcane stem node detection is one of the key functions of a small intelligent sugarcane harvesting robot, but the accuracy of sugarcane stem node detection is severely degraded in complex field environments when the sugarcane is in the shadow of confusing backgrounds and other objects. Methods: To address the problem of low accuracy of sugarcane arise node detection in complex environments, this paper proposes an improved sugarcane stem node detection model based on YOLOv7. First, the SimAM (A Simple Parameter-Free Attention Module for Convolutional Neural Networks) attention mechanism is added to solve the problem of feature loss due to the loss of image global context information in the convolution process, which improves the detection accuracy of the model in the case of image blurring; Second, the Deformable convolution Network is used to replace some of the traditional convolution layers in the original YOLOv7. Finally, a new bounding box regression loss function WIoU Loss is introduced to solve the problem of unbalanced sample quality, improve the model robustness and generalization ability, and accelerate the convergence speed of the network. Results: The experimental results show that the mAP of the improved algorithm model is 94.53% and the F1 value is 92.41, which are 3.43% and 2.21 respectively compared with the YOLOv7 model, and compared with the mAP of the SOTA method which is 94.1%, an improvement of 0.43% is achieved, which effectively improves the detection performance of the target detection model. Discussion: This study provides a theoretical basis and technical support for the development of a small intelligent sugarcane harvesting robot, and may also provide a reference for the detection of other types of crops in similar environments.