Deep learning is thought of as a promising mean to identify maize diseases. However, the drawback of deep learning is the huge sample data and low accuracy. In this paper, we proposed a multi-scale convolutional global pooling neural network to improve the accuracy of maize diseases identification. Firstly, on the basis of the AlexNet model, a convolutional layer and new Inception module are added to enhance the ability of AlexNet features extraction. Then, in order to avoid the over-fitting problem caused by too many parameters, we use the global pooling layer to replace the original fully-connected layer. Besides, we also adopt the transfer learning method to solve the over-fitting problem caused by insufficient sample data. The improved model can reduce over-fitting and epochs to enhance the performance of maize diseases recognition. From the considerable experimental results, we can conclude that the proposed model has better performance compared with convolutional neural network models VGGNet-16, DenseNet, ResNet-50 and AlexNet in recognition accuracy.

For improving the accuracy of weed recognition under complex field conditions, a weed recognition method using depthwise separable convolutional neural network based on deep transfer learning was proposed in this study. To improve the model classification accuracy, the Xception model was refined by using model transferring and fine-tuning. Specifically, the weight parameters trained by ImageNet data set were transferred to the Xception model. Then a global average pooling layer replaced the full connection layer of the Xception model. Finally, the XGBoost classifier was added to the top layer of the model to output results. The performance of the proposed model was validated using the digital field weed images. The experimental results demonstrated that the proposed method had significant improvement in both classification accuracy and training speed in comparison of VGG16, ResNet50, and Xception depth models. The test recognition accuracy of the proposed model reached to 99.63%. Further, the training of each round time cost was 208 s, less than VGG16, ResNet50 and Xception models’, which were 248 s, 245 s and 217 s, respectively. Therefore, the proposed model has a promising ability to process image detection and output more accurate recognition results, which can be used for other crops’ precision management.

Tomato is generally cultivated by transplanting seedlings in ridges and furrows. During growth, there are various types of tomato pests and diseases, making it challenging to identify them simultaneously. To address this issue, conventional convolutional neural networks have been investigated, but they have a large number of parameters and are time-consuming. In this paper, we proposed a lightweight multi-scale tomato pest and disease classification network, called CNNA. Firstly, we constructed a dataset of tomato diseases and pests consisting of 27,193 images with 18 categories. Then, we compressed and optimized the ConvNeXt-Tiny network structure to maintain accuracy while significantly reducing the number of parameters. In addition, we proposed a multi-scale feature fusion module to improve the feature extraction ability of the model for different spot sizes and pests, and we proposed a global channel attention mechanism to enhance the sensitivity of the network model to spot and pest features. Finally, the model was trained and deployed to the Jetson TX2 NX for inference of tomato pests and diseases in video stream data. The experimental results showed that the proposed CNNA model outperformed the pre-trained lightweight models such as MobileNetV3, MobileVit, and ShuffleNetV2 in terms of accuracy and all parameters, with a recognition accuracy of 98.96%. Meanwhile, the error rate, inference time for a single image, network parameters, FLOPs, and model size were only 1%, 47.35 ms, 0.37 M, 237.61 M, and 1.47 MB, respectively.

Early blight and late blight are two of the most prevalent and severe diseases affecting potato crops. Efficient and accurate grading of their severity is crucial for effective disease management. However, existing grading methods are limited to assessing the severity of each disease independently, often resulting in low recognition accuracy and slow grading processes. To address these challenges, this study proposes a novel two-stage approach for the rapid severity grading of both early blight and late blight in potato plants. In this research, two lightweight models were developed: Coformer and SegCoformer. In the initial stage, Coformer efficiently categorizes potato leaves into three classes: those afflicted by early blight, those afflicted by late blight, and healthy leaves. In the subsequent stage, SegCoformer accurately segments leaves, lesions, and backgrounds within the images obtained from the first stage. Furthermore, it assigns severity labels to the identified leaf lesions. To validate the accuracy and processing speed of the proposed methods, we conduct experimental comparisons. The experimental results indicate that Coformer achieves a classification accuracy as high as 97.86%, while SegCoformer achieves an mIoU of 88.50% for semantic segmentation. The combined accuracy of this method reaches 84%, outperforming the Sit + Unet_V accuracy by 1%. Notably, this approach achieves heightened accuracy while maintaining a faster processing speed, completing image processing in just 258.26 ms. This research methodology effectively enhances agricultural production efficiency.

Crop seedlings are similar in appearance to weeds, making crop detection extremely difficult. To solve the problem of detecting crop seedlings in complex field environments, a seedling dataset with four crops was constructed in this study. The single leaf labeling method was proposed as an alternative to conventional labeling approaches to improve the detection accuracy for dense planting crops. Second, a seedling detection network based on YOLOv5 and a transformer mechanism was proposed, and the effects of three features (query, key and value) in the transformer mechanism on the detection accuracy were explored in detail. Finally, the seedling detection network was optimized into a lightweight network. The experimental results show that application of the single leaf labeling method could improve the mAP0.5 of the model by 1.2% and effectively solve the problem of missed detection. By adding the transformer mechanism module, the mAP0.5 was improved by 1.5%, enhancing the detection capability of the model for dense and obscured targets. In the end, this study found that query features had the least impact on the transformer mechanism, and the optimized model improved the computation speed by 23 ms·frame⁻¹ on the intelligent computing platform Jetson TX2, providing a theoretical basis and technical support for real-time seedling management.