As a promising new paradigm, Vehicular Edge Computing (VEC) can improve the QoS of vehicular applications by computation offloading. However, with more and computation-intensive applications, VEC servers face challenges limited resources. In this paper, we study how to effectively economically utilize idle resources in volunteer vehicles handle overloaded tasks servers. First, present model assisted edge computing, which cost utility functions are defined for requesting servers, encouraged...

Rapid and accurate detection of pathogen spores is an important step to achieve early diagnosis diseases in precision agriculture. Traditional methods are time-consuming, laborious, subjective, image processing mainly rely on manually designed features that difficult cope with spore complex scenes. Therefore, MG-YOLO algorithm (Multi-head self-attention Ghost-optimized YOLO) proposed detect gray mold rapidly. Firstly, Multi-head introduced the backbone capture global information spores....

Tomato is one of the most important vegetables in world. Timely and accurate identification tomato disease a critical way to ensure quality yield production. The convolutional neural network crucial means identification. However, this method requires manual annotation large amount image data, which wastes human cost scientific research.To simplify process labeling improve accuracy recognition balance various effects, BC-YOLOv5 proposed identify healthy growth nine types diseased leaves. In...

Tomato is one of the most important vegetables in world. Timely and accurate identification tomato diseases their types a critical way to ensure quality yield production. To simplify disease labeling process, increase that can recognize, obtain more balanced recognition effect for various diseases, method based on improved YOLOv5 proposed this paper. The approach uses YOLOv5-6.1 as base model, adds CBAM (Convolutional Block Attention Module) part CONV modules its network structure,...

Image-based 3D reconstruction is one of the most important tasks in Computer Vision with many solutions proposed over last few decades. The objective to extract metric information i.e. geometry scene objects directly from images. These can then be used a wide range applications such as film, games, virtual reality, etc. Recently, deep learning techniques have been tackle this problem. They rely on training vast amounts data learn associate features between images through convolutional neural...

Sports can cause the consumption of energy materials in body. The rational use nutritional supplements maintain homeostasis organism, which plays a very important role improving competitive performance sports athletes. purpose this study is to explore effect on basketball fatigue. method as follows: first all, 15 players our city were selected experimental objects, and they randomly divided into group control group. members took nutrients. After training, 6 days week, 3 hours morning...

Fungal diseases are the main factors affecting quality and production of vegetables. Rapid accurate detection pathogenic spores is great practical significance for early prediction prevention diseases. However, there some problems with microscopic images collected in natural environment, such as complex backgrounds, more disturbing materials, small size spores, various forms. Therefore, this study proposed an improved method GCS-YOLOv8 (Global context CARFAE Small detector-optimized YOLOv8),...

Energy has become an important resource for computer system. In multicore systems, shared can lead to contention between tasks running on different cores, which also will give rise poor power efficiency. The system with dynamic voltage and frequency scaling(DVFS) capability provide support energy management make a better However, the coming of virtualization technology in management.

At this stage, network information technology has been fully popularized and widely used in various fields achieved remarkable results. In the development of physical education courses colleges universities, effective use multimedia network-assisted teaching can greatly improve quality college students, stimulate students’ interest courses, promote all-round development. Therefore, after is over, comparative analysis to compare effects computer educational pattern previous determine...

Online teaching is becoming familiar in recent days for bridging the educational gap and long-distance coverage. The online modes provide better visualization representation of different subjects, including sports physical education classes. Data-to-representation provides a model without interrupting knowledge-based performance. For leveraging this feature, article introduces Conveyance-dependent Teaching Mode (CTM) paradigm. proposed paradigm employs neural learning to improve performance...

We introduce a software package called HIFIR for preconditioning sparse, unsymmetric, ill-conditioned, and potentially singular systems. computes hybrid incomplete factorization, which combines multilevel LU factorization with truncated, rank-revealing QR on the final Schur complement. This novel hybridization is based new theory of approximate generalized inverse $\epsilon$-accuracy. It enables near-optimal preconditioners consistent systems flexible GMRES to solve inconsistent when coupled...