With the global population growth and increasing food demand, the development of precision agriculture has become particularly critical. In precision agriculture, accurately identifying areas of nitrogen stress in crops and planning precise fertilization paths are crucial. However, traditional coverage path-planning (CPP) typically considers only single-area tasks and overlooks the multi-area tasks CPP. To address this problem, this study proposed a Regional Framework for Coverage Path-Planning for Precision Fertilization (RFCPPF) for crop protection UAVs in multi-area tasks. This framework includes three modules: nitrogen stress spatial distribution extraction, multi-area tasks environmental map construction, and coverage path-planning. Firstly, Sentinel-2 remote-sensing images are processed using the Google Earth Engine (GEE) platform, and the Green Normalized Difference Vegetation Index (GNDVI) is calculated to extract the spatial distribution of nitrogen stress. A multi-area tasks environmental map is constructed to guide multiple UAV agents. Subsequently, improvements based on the Double Deep Q Network (DDQN) are introduced, incorporating Long Short-Term Memory (LSTM) and dueling network structures. Additionally, a multi-objective reward function and a state and action selection strategy suitable for stress area plant protection operations are designed. Simulation experiments verify the superiority of the proposed method in reducing redundant paths and improving coverage efficiency. The proposed improved DDQN achieved an overall step count that is 60.71% of MLP-DDQN and 90.55% of Breadth-First Search–Boustrophedon Algorithm (BFS-BA). Additionally, the total repeated coverage rate was reduced by 7.06% compared to MLP-DDQN and by 8.82% compared to BFS-BA.

Soil nutrient content is an important index to determine the amount of fertilizer. Traditional soil nutrient content is obtained by manual sampling, which greatly increases the cost of agricultural production input. In order to solve this problem, this paper studies the relationship between soil nutrients, fertilizer amount, yield and the next year’s soil nutrients, and establishes an optimized BP neural network model to realize the prediction of soil nutrient content. Aiming at the problem that the traditional BP neural network model will affect the prediction accuracy due to the different weights and thresholds, the Grey Wolf algorithm with reverse learning mechanism is introduced to obtain the optimal solution of BP neural network weights and thresholds. The soil nutrient content data collected in chenjiadian village, Nongan County, Jilin Province for five consecutive years were used to test the model. The data of the first four years were used as the training set and the data of the fifth year as the test set. The prediction results of BP neural network model, Grey Wolf algorithm optimized BP neural network model and Grey Wolf algorithm optimized BP neural network model with reverse learning mechanism were compared The prediction accuracy of BP neural network model optimized by Grey Wolf algorithm with reverse learning mechanism reached 88.7%, which was better than the first two models. It can provide basis for the fertilization decision of variable rate fertilization in the next year, so as to reduce the input of labor cost.

Applying the method of semantic segmentation to the segmentation of grape leaves is an important method to solve how to segment grape leaves from complex backgrounds. This article uses U-net++ convolutional neural network to segment grape leaves from complex backgrounds using MIOU, PA, and mPA as evaluation metrics. After the leaves are segmented, the OTSU threshold segmentation + EXG algorithm is used to extract the diseased spots of grape leaves and healthy grape leaves by increasing the proportion of green vectors. Grape leaf disease was automatically graded by the ratio of the healthy green part of the grape to the total leaf area.

Accurate postharvest quality assessment of specialty grains remains challenging due to complex quality variations and the limited deployment of multimodal sensing systems. RGB imaging provides efficient external phenotypic information, whereas visible–near-infrared (VNIR) sensing provides complementary wavelength-dependent reflectance information. However, the high cost and complexity of spectral acquisition restrict the practical application of multimodal models. This study proposes an RGB–VNIR cross-modal decision residual distillation framework for non-destructive quality recognition of Canadian wild rice kernels. The proposed Decision Spectral Residual Extractor (DSRE) models VNIR information as decision-level residual knowledge relative to RGB predictions, while the Cross-modal Residual Reliability Adaptation (CRRA) selectively transfers transferable spectral knowledge according to residual utility and recoverability. Experiments on eight quality categories of Canadian wild rice kernels demonstrated the effectiveness of the proposed framework. On the paired RGB–VNIR test set (n = 384), the DSRE-based multimodal teacher achieved 97.40% Accuracy, compared with 94.53% for the CRRA RGB-only student on the same samples, confirming the complementary contribution of VNIR information. In a separate evaluation on the held-out RGB-only test set (n = 1570), the CRRA student achieved 98.60% Accuracy and 98.62% Macro-F1, indicating effective multimodal knowledge transfer without VNIR acquisition during inference. The proposed framework establishes practical multimodal learning with a unimodal deployment paradigm, providing a promising solution for cost-effective and non-destructive quality evaluation of specialty grains.

This study presents a dehazing algorithm specifically tailored for improving the quality of surveillance images impaired by foggy weather. The core research areas encompass atmospheric light estimation and the dehazing process. First, a perfect white balance method is employed for preprocessing surveillance video frames to remove the color influence caused by fog in the images. Second, based on an in-depth analysis of surveillance scene characteristics, a depth-based atmospheric light estimation method is proposed to accurately obtain crucial atmospheric light information within fog. Subsequently, a cross Bilateral filter is introduced, carefully balancing filtering efficacy and edge preservation to effectively eliminate noise in foggy images. Finally, by subtracting the estimated atmospheric light component from surveillance video frames, clear dehazed surveillance images are obtained. The performance evaluation of the proposed algorithm on six datasets yields the following metrics: the highest Peak Signal-to-Noise Ratio (PSNR) is 8.507, the Structural Similarity Index (SSIM) is 0.826, the Universal Image Quality Index (UQI) is 0.732, and the information entropy is 7.663. These quantified performance indicators clearly illustrate the remarkable performance of the proposed algorithm in dehazing tasks, providing solid support for its practical application in the field of surveillance image enhancement.