Advancements in deep learning and computer vision have led to the discovery of numerous effective solutions to challenging problems in the field of agricultural automation. With the aim to improve the detection precision in the autonomous harvesting process of green asparagus, in this article, we proposed the DA-Mask RCNN model, which utilizes the depth information in the region proposal network. Firstly, the deep residual network and feature pyramid network were combined to form the backbone network. Secondly, the DA-Mask RCNN model added a depth filter to aid the softmax function in anchor classification. Afterwards, the region proposals were further processed by the detection head unit. The training and test images were mainly acquired from different regions in the basin of the Yangtze River. During the capturing process, various weather and illumination conditions were taken into account, including sunny weather, sunny but overshadowed conditions, cloudy weather, and daytime greenhouse conditions as well as nighttime greenhouse conditions. Performance experiments, comparison experiments, and ablation experiments were carried out using the five constructed datasets to verify the effectiveness of the proposed model. Precision, recall, and F1-score values were applied to evaluate the performances of different approaches. The overall experimental results demonstrate that the balance of the precision and speed of the proposed DA-Mask RCNN model outperform those of existing algorithms.

Under a student-centered educational paradigm, project-based learning (PBL) assessment requires accurate identification of classroom behaviors to facilitate effective teaching evaluations and the implementation of personalized learning strategies. The increasing use of visual and multi-modal sensors in smart classrooms has made it possible to continuously capture rich behavioral data. However, challenges such as lighting variations, occlusions, and diverse behaviors complicate sensor-based behavior analysis. To address these issues, we introduce CSSA-YOLO, a novel detection network that incorporates cross-scale feature optimization. First, we establish a C2fs module that captures spatiotemporal dependencies in small-scale actions such as hand-raising through hierarchical window attention. Second, a Shuffle Attention mechanism is then integrated into the neck to suppress interference from complex backgrounds, thereby enhancing the model’s ability to focus on relevant features. Finally, to further enhance the network’s ability to detect small targets and complex boundary behaviors, we utilize the WIoU loss function, which dynamically weights gradients to optimize the localization accuracy of occluded targets. Experiments involving the SCB03-S dataset showed that CSSA-YOLO outperforms traditional methods, achieving an mAP₅₀ of 76.0%, surpassing YOLOv8m by 1.2%, particularly in complex background and occlusion scenarios. Furthermore, it reaches 78.31 FPS, meeting the requirements for real-time application. This study offers a reliable solution for precise behavior recognition in classroom settings, supporting the development of intelligent education systems.

Sandstone is widely used a construction and building material. However, its uniaxial tensile strength (UTS) is not adequately understood. To characterize the uniaxial tensile strength of natural sandstone, three groups of specimens were fabricated for four-point bending, uniaxial compressive, and tensile tests. To characterize the evolution of the stress–strain profiles obtained via these tests, representative expressions were developed in terms of normalized strain and strength. The magnitude of the uniaxial tensile strength exceeded that of the four-point bending strength, indicating that the uniaxial tensile strength cannot be represented by the four-point bending strength. The experimental ratio of uniaxial tensile and compression strength (33–41) was underestimated by the empirical expressions reported in the literature. The suggested correction coefficient for the FBS is 0.25. The compressive modulus (E_c) was generally identical to the experimental results published in the literature, whereas the tensile modulus (E_t) was overestimated. The experimental modular ratio, E_t/E_c, ranged from 0.12 to 0.14; it was not sensitive to Poisson’s ratio, but it increased slightly with the compressive modulus. This work can serve as a reference for computing the load-bearing capacity of sandstone components under tension.

With the intensification of armed conflicts driven by regional incentives, global geopolitical conflicts are becoming increasingly intense. In addition, the possibility of another financial crisis is approaching, and global inflation is rapidly rising. As a result, Europe and the United States must restart coal-fired power generation to cope with energy shortages and delay carbon neutrality goals. However, the current political and public opinion about the environment has led to a one-sided exaggeration and political criticism of China’s carbon emissions, resulting in China’s contribution to carbon neutrality being intentionally or unintentionally ignored. The uncertainty surrounding future low-carbon policies has made climate observers increasingly concerned about the threat of environmental degradation to fragile intergovernmental decarbonization efforts. This article aims to clarify the one-sided view of China’s carbon emissions internationally, clarify China’s measurement indicators for carbon emissions, analyze China’s advantages in responding to the global warming crisis under complex historical and political conditions, summarize China’s efforts to achieve its dual carbon goals in the current situation, and thus summarize China’s unique and absolute advantages in cooperation in clean energy, energy storage, and ultra-high voltage transmission networks that are beneficial to global climate change. This will clarify the truth regarding China’s carbon emissions in the global context and boost global confidence in addressing climate change.

Enhancing vehicle emergency braking performance is crucial for vehicular safety and reliability. We have observed that the traditional vehicle dynamics based model predictive control (MPC) algorithm which is used to produce emergency braking strategy fails to achieve the minimization of emergency braking distance. To address this issue, this article employed a simulation based approach to improve emergency braking performance via machine learning based algorithms. We design a data optimization model which optimizes the longitudinal forces and slip ratios of the four wheels based on back-propagation neural network (BPNN) and constraint optimization algorithm to reduce emergency braking distance. In addition, a Balltree nearest neighbor search based generator is proposed to produce superior emergency braking strategies for real-time purpose. The experimental results demonstrate that the emergency braking strategy which we optimized achieves an average enhancement of 13% in emergency braking performance compared to the traditional MPC algorithm. Unlike other machine learning based algorithms that have a delay of over 0.01 s, the proposed generator has an execution time of 0.0008 s, meeting the delay requirements under emergency braking conditions.