Accurate detection of pests and diseases in eggplant under real-field conditions remains challenging due to large variations in target scales, complex background clutter, and the frequent presence of small and occluded objects. To address these issues, this paper proposes Eggplant-DETR, an improved detection model based on the RT-DETR architecture. The model introduces a collaborative mechanism that integrates multi-scale feature enhancement, semantic fusion, and frequency perception. The key components of this mechanism include: (1) CPSE module is designed in the shallow layers of the backbone to enhance detailed features of small pest targets, providing more discriminative low-level features for subsequent processing and mitigating detail loss of small objects; (2) The CHSFPN module effectively integrates detail-enhanced features from CPSE and global semantic information from AIFI, generating multi-scale and semantically rich feature maps for the subsequent WTMANet module; (3) at the end of the encoder integrating the WTMANet module refines the multi-scale features from CHSFPN and significantly improves perception capability for irregularly shaped targets (e.g., FruitRot) and small objects (e.g., MelonThrips). Extensive experiments on a public eggplant pest and disease dataset demonstrate that the proposed method achieves mAP50 of 77.8%, parameters reduces to 15.77M. The overall performance surpasses that of the other 11 mainstream comparison models. The framework provides an effective solution for smart agricultural pest and disease detection, while also offering valuable technical insights into the application of frequency-domain feature representation in this field.

Pest management is essential for agricultural production and food security, as pests can cause significant crop losses and economic impact. Early pest detection is key to timely intervention. While object detection models perform well on various datasets, they assume i.i.d. data, which is often not the case in diverse real-world environments, leading to decreased accuracy. To solve the problem, we propose the CrossDomain-PestDetect (CDPD) method, which is based on the YOLOv9 model and incorporates a test-time adaptation (TTA) framework. CDPD includes Dynamic Data Augmentation (DynamicDA), a Dynamic Adaptive Gate (DAG), and a Multi-Task Dynamic Adaptation Model (MT-DAM). Our DynamicDA enhances images for each batch by combining strong and weak augmentations. The MT-DAM integrates an object detection model with an image segmentation model, exchanging information through feature fusion at the feature extraction layer. During testing, test-time adaptation updates both models, continuing feature fusion during forward propagation. DAG adaptively controls the degree of feature fusion to improve detection capabilities. Self-supervised learning enables the model to adapt during testing to changing environments. Experiments show that without test-time adaptation, our method achieved a 7.6% increase in mAP50 over the baseline in the original environment and a 16.1% increase in the target environment. Finally, with test-time adaptation, the mAP50 score in the unseen target environment reaches 73.8%, which is a significant improvement over the baseline.

In tomato cultivation, various diseases significantly impact tomato quality and yield. The substantial scale differences among diseased leaf targets pose precise detection and identification challenges. Moreover, early detection of disease infection in small leaves during the initial growth stages is crucial for implementing timely intervention and prevention strategies. To address these challenges, we propose a novel tomato disease detection method called TomatoLeafDet, which integrates multi-scale feature processing techniques and small object detection technologies.Initially, we designed a Cross Stage Partial -Serial Multi-kernel Feature Aggregation (CSP-SMKFA) module to extract feature information from targets at different scales, enhancing the model's perception of multi-scale objects. Next, we introduced a Symmetrical Re-calibration Aggregation (SRCA) module, incorporating a bidirectional fusion mechanism between highresolution and low-resolution features. This approach facilitates more comprehensive information transmission between features, further improving the efficacy of multi-scale feature fusion. Finally, we proposed a Re-Calibration Feature Pyramid Network with a small object detection head to consolidate the multi-scale features extracted by the backbone network. This network provides richer multi-scale feature information input for detection heads at various scales. Results indicate that our method outperforms YOLOv9 and YOLOv10 on two datasets. Notably, on the CCMT tomato dataset, the proposed model achieved improvements in mean Average Precision (mAP50) of 4.4%, 1.9%, and 2.3% compared to the baseline model, YOLOv9s, and YOLOv10n, respectively, exhibiting significant efficacy.

Tomato leaf diseases significantly impact the yield and quality of tomatoes during cultivation, the main of which are septoria leaf spot, leaf curl virus, verticillium wilt, and early blight. These diseases necessitate prompt detection and management strategies to mitigate their deleterious effects on crop productivity. Due to the considerable scale variations in diseased tomato leaves, accurate and rapid detection and diagnosis remain challenging. To address the detection of tomato leaf diseases at different scales, we propose a real-time detection model incorporating a Multi-kernel Inception Aggregation Diffusion Network. In this paper, (1) We present a Multi-kernel Inception Aggregation Diffusion Network (MIADN) for the feature processing stage, which facilitates the aggregation and diffusion of multi-scale features across hierarchical levels, benefiting the detection of targets at various scales. (2) We present the Multi-kernel Inception Module (MKIM), designed to extract multi-scale object features using diverse convolution kernels, thereby enhancing the model's feature fusion and representation capabilities. (3) We incorporate the efficient FasterNet network at the feature extraction stage to preserve feature diversity and improve the model's ability to extract complex target features. (4) Extensive comparative and ablation experiments demonstrate that our method achieves the mean average precision (mAP50) of 96.6%, surpassing the baseline model by 4.1% and the advanced YOLOv9s model by 2.0%. This method provides an effective solution for high-quality tomato cultivation.

Under certain conditions, when crude oil is moved by external forces, the property of internal friction generated between crude oil molecules is called crude oil viscosity. The viscosity of crude oil reflects its complex seepage state in porous media. Underground crude oil with high viscosity, always has great flow resistance in porous media, thus the flowing becomes more difficult. Oil viscosity is an indispensable key parameter in the process of dynamic analysis, reservoir engineering calculation and reservoir numerical simulation, which has critical influence on the field of well production or crude oil storage and transportation. Due to different oil viscosity, recovery approach of oil reservoirs, technical measures for storage and transportation, and the quality of oil products will be affected. The composition of crude oil is complicated, but it is mainly composed of carbon and hydrogen elements. The composition has a crucial effect on oil viscosity. Therefore, according to composition data of the actual oil sample, the determination dataset of oil viscosity is constructed together with other key parameters that affect the viscosity of crude oil within the reservoirs. Based on various machine learning algorithms, like extremely randomized trees and XGBoost, determination methods of oil viscosity based on component data and machine learning algorithms are established. In the construction process of computational model of oil viscosity, whole dataset is parted to the training dataset and the testing dataset in the ratio of 8:2. The training dataset is mainly used to determine the best hyper-parameter combination of machine learning algorithm, while the testing dataset is used to determine the accuracy and adaptability of the corresponding method. Compared with methods such as experimental method and empirical formula method, the determination method of oil viscosity based on component data and machine learning algorithm does not require extra experimental costs and has a considerable degree of accuracy. Once the relevant input parameters are determined, the viscosity determination of multiple groups of oil samples could be completed quickly and accurately.

At the present stage, the field of detecting vegetable pests and diseases is in dire need of the integration of computer vision technologies. However, the deployment of efficient and lightweight object-detection models on edge devices in vegetable cultivation environments is a key issue. To address the limitations of current target-detection models, we propose a novel lightweight object-detection model based on YOLOv8n while maintaining high accuracy. In this paper, (1) we propose a new neck structure, Focus Multi-scale Feature Diffusion Interaction (FMDI), and inject it into the YOLOv8n architecture, which performs multi-scale fusion across hierarchical features and improves the accuracy of pest target detection. (2) We propose a new efficient Multi-core Focused Network (MFN) for extracting features of different scales and capturing local contextual information, which optimizes the processing power of feature information. (3) We incorporate the novel and efficient Universal Inverted Bottleneck (UIB) block to replace the original bottleneck block, which effectively simplifies the structure of the block and achieves the lightweight model. Finally, the performance of YOLO-FMDI is evaluated through a large number of ablation and comparison experiments. Notably, compared with the original YOLOv8n, our model reduces the parameters, GFLOPs, and model size by 18.2%, 6.1%, and 15.9%, respectively, improving the mean average precision (mAP50) by 1.2%. These findings emphasize the excellent performance of our proposed model for tomato pest and disease detection, which provides a lightweight and high-precision solution for vegetable cultivation applications.

In recent years, unmanned aerial vehicle (UAV) remote sensing has developed rapidly in the field of farmland information monitoring. Real-time and accurate access to farmland information and crop growth dynamics is a prerequisite for the implementation of precision agriculture. Machine learning identifies existing knowledge to acquire new knowledge, promotes the development of Artificial Intelligence, and brings a large number of data training sets for machine learning. This article aims to ensure the safe operation of agricultural information systems and guarantee the data security of intelligent agriculture. The machine learning method explores the wireless network deployment of the UAV system. The geographical location deployment of agricultural information security can effectively carry out rapid security detection of agricultural information security. First, the UAV-assisted information acquisition system was studied. Besides, a double deep $Q$ -network (DDQN) algorithm for location deployment based on geography position information (GPI) was proposed to quickly optimize the deployment location of UAVs. GPI can avoid the complicated calculation process of channel state information. The DDQN algorithm was introduced to obtain the functional relationship between the GPI and the optimal UAV deployment position, forming a new GPI-Learning strategy. In addition, the convolutional neural network (CNN) and long short-term memory (LSTM) are integrated as the CNN–LSTM algorithm to build the intrusion detection system for Agricultural Internet of Things (AIoT) for agriculture. In the integrated network structure of the system, LSTM is responsible for data transmission, and CNN is capable of network model building. Combined with the influence of various parameters on the performance of the UAV deployment location algorithm, the simulation experiment set the population size as 36, the discovery probability as 0.25, the step scaling factor as 0.8, and the Levy flight index as 1.25. The network throughput performance of the GPI-Learning algorithm combined with cuckoo search was better than other algorithms under different numbers of UAVs. On the KDD-CUP99 data set, the accuracy and detection rate of the AIoT intrusion detection system based on the CNN+LSTM algorithm reached 93.5% and 94.4%, respectively. In general, the AIoT intrusion detection system reported here has crucial practical reference value for the safe operation of agricultural information systems.

The suspension system of the drum washing machine is the object of study. To reduce the vibration of drum washing machines during high-speed dehydration, a shape memory alloy is proposed to be used to make suspension spring. The spring stiffness can be controlled according to the characteristic that the elastic modulus of SMA varies with temperature to achieve damping effect. The mathematical model and multi-body dynamics simulation model are established for the suspension system of the drum washing machine, the vibration generation mechanism of the drum washing machine and the SMA active vibration control principle are analyzed, and the vibration control effect of the SMA suspension spring is studied for different speed simulations. The results show that the SMA variable stiffness suspension spring has a significant vibration control effect compared with the spring of general material.

The ignition temperature of a diesel oxidation catalyst (DOC) and the internal temperature-field distribution of the diesel particulate filter (DPF) during active regeneration are investigated during an engine bench test in this study. Based on the dropped to idle (DTI) test, a test method is developed to determine the safe regeneration temperature of the DPF. The results show that when the inlet temperature of the DOC is more than 240 °C, the DOC begins ignition and reaches the target temperature of 600 °C set for active regeneration of DPF; when the inlet exhaust temperature of the DOC is between 240 and 280 °C, a higher injection rate is required to reduce the secondary pollution of HC and thus make the DPF reach the set target temperature as soon as possible. The active regeneration process of the DPF is divided into three stages. During ignition, the temperature of the DPF inlet and outlet increases rapidly and successively. The internal and outlet temperatures of DPF during regeneration are approximately 50 °C higher than the inlet temperature. At the end of regeneration, the DPF inlet to outlet temperature drops rapidly. A feed-forward design and feedback algorithm are used to verify the change in the target regeneration temperature. The overshoot of the DPF control strategy was less than 3%, and the steady-state temperature control error was less than 20 °C. The results of this study provide a basis for the safe control of DPFs’ active regeneration temperatures.

There is increasing demand for the on-board diagnosis of lubricating oils. In this research, we consider various sensor principles for on-board diagnosis of the thermal aging of engine oils. One of the parameters investigated is the viscosity of the lubricating oil, which can be efficiently measured using a microacoustic sensor. Compared with conventional viscometers, these sensors probe a different rheological domain, which needs to be considered in the interpretation of measurement results. This specific behavior is examined by systematically investigating engine oils, with and without additive packages, that were subjected to a defined artificial aging process. This paper presents design strategies for the algorithm developed and applied for direct on-board diagnosis of engine oil conditions with a fluid property sensor; this enables prediction of remaining oil life and optimization of oil change intervals, thereby minimizing the likelihood of dramatic engine failure and reducing maintenance costs. After a general description of the principles of sensor measurement, different engine oil contaminants, aging phenomena, and associated sensor detection and measurement capabilities are discussed.