As a key global food reserve, rice disease detection technology plays an important role in promoting food production, protecting ecological balance and supporting sustainable agricultural development. However, existing rice disease identification techniques face many challenges, such as low training efficiency, insufficient model accuracy, incompatibility with mobile devices, and the need for a large number of training datasets. This study aims to develop a rice disease detection model that is highly accurate, resource efficient, and suitable for mobile deployment to address the limitations of existing technologies. We propose the Transfer Layer iRMB-YOLOv8 (TLI-YOLO) model, which modifies some components of the YOLOv8 network structure based on transfer learning. The innovation of this method is mainly reflected in four key components. First, transfer learning is used to import the pretrained model weights into the TLI-YOLO model, which significantly reduces the dataset requirements and accelerates model convergence. Secondly, it innovatively integrates a new small object detection layer into the feature fusion layer, which enhances the detection ability by combining shallow and deep feature maps so as to learn small object features more effectively. Third, this study is the first to introduce the iRMB attention mechanism, which effectively integrates Inverted Residual Blocks and Transformers, and introduces deep separable convolution to maintain the spatial integrity of features, thus improving the efficiency of computational resources on mobile platforms. Finally, this study adopted the WIoUv3 loss function and added a dynamic non-monotonic aggregation mechanism to the standard IoU calculation to more accurately evaluate and penalize the difference between the predicted and actual bounding boxes, thus improving the robustness and generalization ability of the model. The final test shows that the TLI-YOLO model achieved 93.1% precision, 88% recall, 95% mAP, and a 90.48% F1 score on the custom dataset, with only 12.60 GFLOPS of computation. Compared with YOLOv8n, the precision improved by 7.8%, the recall rate improved by 7.2%, and mAP@.5 improved by 7.6%. In addition, the model demonstrated real-time detection capability on an Android device and achieved efficiency of 30 FPS, which meets the needs of on-site diagnosis. This approach provides important support for rice disease monitoring.

Convolutional Neural Networks (CNNs) have been widely applied in various edge computing devices based on intelligent sensors. However, due to the high computational demands of CNN tasks, the limited computing resources of edge intelligent terminal devices, and significant architectural differences among these devices, it is challenging for edge devices to independently execute inference tasks locally. Collaborative inference among edge terminal devices can effectively utilize idle computing and storage resources and optimize latency characteristics, thus significantly addressing the challenges posed by the computational intensity of CNNs. This paper targets efficient collaborative execution of CNN inference tasks among heterogeneous and resource-constrained edge terminal devices. We propose a pre-partitioning deployment method for CNNs based on critical operator layers, and optimize the system bottleneck latency during pipeline parallelism using data compression, queuing, and “micro-shifting” techniques. Experimental results demonstrate that our method achieves significant acceleration in CNN inference within heterogeneous environments, improving performance by 71.6% compared to existing popular frameworks.

With the explosive growth of terminal devices, scheduling massive parallel task streams has become a core challenge for distributed platforms. For computing resource providers, enhancing reliability, shortening response times, and reducing costs are significant challenges, particularly in achieving energy efficiency through scheduling to realize green computing. This paper investigates the heterogeneous parallel task flow scheduling problem to minimize system energy consumption under response time constraints. First, for a set of independent tasks capable of parallel computation on heterogeneous terminals, the task scheduling is performed according to the computational resource capabilities of each terminal. The problem is modeled as a mixed-integer nonlinear programming problem using a Directed Acyclic Graph as the input model. Then, a dynamic scheduling method based on heuristic and reinforcement learning algorithms is proposed to schedule the task flows. Furthermore, dynamic redundancy is applied to certain tasks based on reliability analysis to enhance system fault tolerance and improve service quality. Experimental results show that our method can achieve significant improvements, reducing energy consumption by 14.3% compared to existing approaches on two practical workflow instances.

In recent years, the surge in network traffic has led to a substantial increase in energy consumption, making the construction of green and energy-efficient networks a critical challenge in the field of communications. Software-Defined Networking (SDN), with its centralized control characteristic, provides a new paradigm for the collaborative scheduling of actuators. However, traditional distributed network architectures lack global regulation capabilities, resulting in low resource utilization. Moreover, existing SDN traffic management methods mostly adopt fixed-weight reward functions, which are difficult to adapt to the dynamic fluctuation of network traffic and device heterogeneity, failing to meet the real-time and stability requirements of actuators in control scenarios. To address these issues, this study proposes a Dynamic Weight Generation Deep Q-Network (DWG-DQN) framework. By integrating a Long Short-Term Memory (LSTM) network with the SDN actuator scheduling mechanism, the system dynamically generates adaptive weight vectors, enabling real-time collaborative optimization of energy consumption, load balancing, and bandwidth utilization. Experimental results demonstrate that in fat-tree topology experiments, the proposed method achieves a 12.23% increase in average reward, a 33.93% reduction in energy consumption, a 31.12% improvement in load balancing, and a 24.03% enhancement in bandwidth utilization. Compared with fixed-weight method, it consistently outperforms in key performance indicators. The dynamic weight generation mechanism effectively solves the multi-objective optimization problem of actuators in dynamic network environments, offering a viable solution for the intelligent scheduling of actuators in SDN-based green traffic management.

With the rapid advancement of cloud computing, edge computing, and the Internet of Things, traditional routing protocols such as OSPF—which rely solely on network topology and link state while neglecting computing power resource status—struggle to meet the network-computing synergy demands of the Computing Power Network (CPN). Existing reinforcement learning-based routing approaches, despite incorporating deep strategies, still suffer from issues such as resource imbalance. To address this, this study proposes a reinforcement learning-based computing-aware routing path selection method—the Computing-Aware Routing-Reinforcement Learning (CAR-RL) algorithm. This achieves coordination between network and computing power resources through multi-factor joint computation of “computational power + network”. The algorithm constructs a multi-factor weighted Markov Decision Process (MDP) to select the optimal computing-aware routing path by real-time perception of network traffic and computing power status. Experiments conducted on the GN4–3N network topology using Mininet and K8S simulations demonstrate that compared to algorithms such as Q-Learning, DDPG, and CEDRL, the CAR-RL algorithm achieves performance improvements of 24.7%, 35.6%, and 23.1%, respectively, in average packet loss rate, average latency, and average throughput. This research not only provides a reference technical implementation path for computing-aware routing selection and optimisation in computing power networks but also advances the efficient integration of network and computing power resources.

Automatic classroom student behavior recognition faces three major challenges: the coexistence of small targets and complex backgrounds, the conflict between details and semantics, and the interference from low-quality samples. To address these issues, this paper proposes a scene-driven integrated optimization framework, termed Hyper-YOLO-G. In the backbone network, a Squeeze-and-Excitation (SE) attention module is embedded to purify features and thereby enhance the channels of key behaviors. In the neck, a bidirectional weighted feature pyramid network (BiFPN) is introduced to perform scale-balanced multi-scale fusion. Moreover, the Wise-IoU v3 (WIoU v3) loss function is adopted to calibrate gradients and reduce the negative impact of low-quality samples. Experimental results on a public classroom behavior dataset show that Hyper-YOLO-G achieves 73.7% mAP@50 (mean average precision at IoU threshold 0.5), which is 4.9% higher than the baseline Hyper-YOLO. Ablation studies and generalization experiments on an independent multi-class dataset further demonstrate the combined effectiveness of each module and the generalization ability of the framework. This study provides a reliable technical path for high-precision, lightweight behavior recognition in complex classroom environments.