The application of visible near-infrared (VIS-NIR) analysis technology to quantify the nutrients in soil has been widely recognized. It is important to improve the performance of regression models that can predict the soil-available potassium concentration. This study collected soil samples from southern Anhui, China, and concentrated on the modelling methods by using 29 pretreatment methods. The results show that a combination of three methods, Savitzky–Golay, standard normal variate, and dislodge tendency, exhibited better stability than others because it was the most capable of achieving levels A and B of the ratio of performance of deviation. The boosting algorithms that form an ensemble of multiple weak predictors exhibited better performance than partial least square (PLS) regression and support vector regression (SVR) for the prediction of soil-available potassium. These regression models could be employed to precisely predict the soil-available potassium concentration.

Deep neural networks (DNNs) have become the de facto standard for image recognition tasks, and their applications with respect to plant diseases have also obtained remarkable results. However, the large number of parameters and high computational complexities of these network models make them difficult to deploy on farms in remote areas. In this paper, focusing on the problems of resource constraints and plant diseases, we propose a DNN-based compression method. In order to reduce computational burden, this method uses lightweight fully connected layers to accelerate reasoning, pruning to remove redundant parameters and reduce multiply–accumulate operations, knowledge distillation instead of retraining to restore the lost accuracy, and then quantization to compress the size of the model further. After compressing the mainstream VGGNet and AlexNet models, the compressed versions are applied to the Plant Village dataset of plant disease images, and a performance comparison of the models before and after compression is obtained to verify the proposed method. The results show that the model can be compressed to 0.04 Mb with an accuracy of 97.09%. This experiment also proves the effectiveness of knowledge distillation during the pruning process, and compressed models are more efficient than prevalent lightweight models.

Although the classification method based on the deep neural network has achieved excellent results in classification tasks, it is difficult to apply to real-time scenarios because of high memory footprints and prohibitive inference times. Compared to unstructured pruning, structured pruning techniques can reduce the computation cost of the model runtime more effectively, but inevitably reduces the precision of the model. Traditional methods use fine tuning to restore model damage performance. However, there is still a large gap between the pruned model and the original one. In this paper, we use progressive multi-level distillation learning to compensate for the loss caused by pruning. Pre-pruning and post-pruning networks serve as the teacher and student networks. The proposed approach utilizes the complementary properties of structured pruning and knowledge distillation, which allows the pruned network to learn the intermediate and output representations of the teacher network, thus reducing the influence of the model subject to pruning. Experiments demonstrate that our approach performs better on CIFAR-10, CIFAR-100, and Tiny-ImageNet datasets with different pruning rates. For instance, GoogLeNet can achieve near lossless pruning on the CIFAR-10 dataset with 60% pruning. Moreover, this paper also proves that using the proposed distillation learning method during the pruning process achieves more significant performance gains than after completing the pruning.

This paper proposes a data anomaly detection and correction algorithm for the tea plantation IoT system based on deep learning, aiming at the multi-cause and multi-feature characteristics of abnormal data. The algorithm is based on the Z-score standardization of the original data and the determination of sliding window size according to the sampling frequency. First, we construct a convolutional neural network (CNN) model to extract abnormal data. Second, based on the support vector machine (SVM) algorithm, the Gaussian radial basis function (RBF) and one-to-one (OVO) multiclassification method are used to classify the abnormal data. Then, after extracting the time points of abnormal data, a long short-term memory network is established for prediction with multifactor historical data. The predicted values are used to replace and correct the abnormal data. When multiple consecutive abnormal values are detected, a faulty sensor judgment is given, and the specific faulty sensor location is output. The results show that the accuracy rate and micro-specificity of abnormal data detection for the CNN-SVM model are 3–4% and 20–30% higher than those of the traditional CNN model, respectively. The anomaly detection and correction algorithm for tea plantation data established in this paper provides accurate performance.

The “Dangshan” pear woolliness response is a physiological disease that causes large losses for fruit farmers and nutrient inadequacies.The cause of this disease is predominantly a shortage of boron and calcium in the pear and water loss from the pear. This paper used the fusion of near-infrared Spectroscopy (NIRS) and Computer Vision Technology (CVS) to detect the woolliness response disease of “Dangshan” pears. This paper employs the merging of NIRS features and image features for the detection of “Dangshan” pear woolliness response disease. Near-infrared Spectroscopy (NIRS) reflects information on organic matter containing hydrogen groups and other components in various biochemical structures in the sample under test, and Computer Vision Technology (CVS) captures image information on the disease. This study compares the results of different fusion models. Compared with other strategies, the fusion model combining spectral features and image features had better performance. These fusion models have better model effects than single-feature models, and the effects of these models may vary according to different image depth features selected for fusion modeling. Therefore, the model results of fusion modeling using different image depth features are further compared. The results show that the deeper the depth model in this study, the better the fusion modeling effect of the extracted image features and spectral features. The combination of the MLP classification model and the Xception convolutional neural classification network fused with the NIR spectral features and image features extracted, respectively, was the best combination, with accuracy (0.972), precision (0.974), recall (0.972), and F1 (0.972) of this model being the highest compared to the other models. This article illustrates that the accuracy of the “Dangshan” pear woolliness response disease may be considerably enhanced using the fusion of near-infrared spectra and image-based neural network features. It also provides a theoretical basis for the nondestructive detection of several techniques of spectra and pictures.

The number and growth of new shoots are very important information for bamboo forest cultivation and management. At present, there is no real-time, efficient and accurate monitoring method. In this study, a fixed webcam was applied for image capture, optimized YOLOv4 was used to model the detection of moso bamboo shoots, and a strategy of sorting and screening was proposed to track each moso bamboo shoot. The change in the number and height of moso bamboo shoots was obtained according to the number and height of detection boxes. The experimental results show that the system can remotely and automatically obtain the number of moso bamboo shoots and the pixel height of each bamboo shoot at any given time. The average relative error and variance in the number of moso bamboo shoots were 1.28% and 0.016%, respectively, and those for the corresponding pixel height results were −0.39% and 0.02%. This system can be applied to a series of monitoring purposes, such as the daily or weekly growth rate of moso bamboo shoots at monitoring stations and trends in the height of selected bamboo shoots.