Abstract Marine litter detection plays this important role in low-carbon environments. Although deep-learning methods have proposed more effective solutions for task, unclear marine photography, severe occlusion of seafloor litter, and poor real-time pose great challenges to task. Based on the above problems, paper proposes add Shuffle Attention local modeling YOLOv8 achieve a complete feature fusion mechanism by effectively reusing multiscale features. Next, Explicit Visual Center (EVC)...

According to the problem of large amount carbon emissions during cold chain distribution process, a route optimization method for fresh agricultural products under tax mechanism was proposed. Firstly, with goal minimizing emission cost and comprehensive cost, quantitative analysis is introduced, considering demand quantity, time unloading constraints, mathematical model established. In addition, an improved quantum bacterial foraging algorithm put forward, which uses information update...

Abstract Cold chain logistics distribution orders have increased due to the impact of COVID-19. In view increasing difficulty route optimization and increase carbon emissions in process cold distribution, a mathematical model for vehicles with minimum comprehensive cost is established by considering emission intensity comprehensively this paper. The main contributions paper are as follows: 1) An improved hybrid ant colony algorithm proposed, which combined simulated annealing get rid local...

The direct air-cooling steam turbines have been operated more and in the north of China. backpressure a turbine is affected easily with weather varies very often short time. variation larger range from about 10 to 60 kPa causes many problems design operation turbine. To study properties wet flow low pressure turbine, an optical—pneumatic probe was developed based on multi-wavelength light extinction four-hole wedge probe. Measurements this 300 MW were carried out. measured local wetness,...

Abstract Vehicle detection based on deep learning achieves excellent results in normal environments, but it is still challenging to detect objects low-quality picture obtained hazy weather. Existing methods tend ignore favorable latent information and difficult balance speed accuracy, etc. Therefore, the existing neural network studied, YOLOv3 algorithm improved ResNet. Aiming at problem of low utilization shallow features, DensNet added feature extraction stage reduce loss increase...

For the purpose of solving flexible job-shop scheduling problem (FJSP), an improved quantum genetic algorithm based on earliness/tardiness penalty coefficient is proposed in this paper. minimizing completion time and cost, a simulation model was established firstly. Next, according to characteristics due production, double designed chains coding method proposed. At last, effectiveness verified through being applied Kacem example compared with some existing algorithms.

Folk dance is an important intangible cultural heritage in China. In the environment where movement recognition technology widely used, there still no research field on protection and inheritance of folk culture. order to better protect inherit minority dance, screening typical movements 5 types through video frame processing, obtain key 19 class sequence, build national action data set, put forward a 3D CNN fusion Transformer network model (FCTNet), rate 96.7% experiment. The results show...

Abstract To address the challenges posed by diverse pattern-background elements, intricate details, and complex textures in semantic segmentation of ethnic clothing patterns, this research introduces a novel network model called MST-Unet (Mixed Swin Transformer U-net). The proposed combines U-shaped structure with multiple attention mechanisms. upper layers employ classical convolutional operations, focusing on local relationships initial containing high-resolution details. In deeper layers,...

Abstract In order to further reduce the carbon emission of manufacturing process in flexible job shop, a multi-objective integrated optimization model job-shop scheduling (FJSP) is proposed. A mathematics built this paper minimize makespan, total workload machines and emissions optimize method each machine characteristic, sequence allocation. Considering many parameters are interactional be optimized proposed model, quantum bacterial foraging designed code related parameters. On basis Kacem...

In order to solve the multi-objective flexible job-shop scheduling problem, a novel cloud bacterial foraging optimization algorithm is proposed. this paper, simulation model established maximize makespan and workload of machines. The optimal individuals can preserve c uring position additional turning common ones swim direction them absorb location information. crossover operator mutation are designed avoid shortcoming premature convergence. proposed method verified be more effective than...

Focus on the distribution vehicle scheduling problem under travel time delay, a series of rescue modes generalized from practical planner' experiences were presented. The measurement method value function based prospect theory and disruption management strategy customer's psych ological perception are put forward in this paper. A multi-objective optimization model for logistics is established through programming at first. Next, idea gradual target to obtain terminal adjustment scheme with...

Abstract In order to reduce the carbon emission and improve de-carbonization in distribution of vehicle routing problem with simultaneous delivery pickup (VRPSDP), a two-stage disruption management method based on changes customer demand is proposed this paper. Firstly, mathematics model VRPSDP established optimize parameters cost time window deviation. Next, several equations are inducted designed. Considering strong global searching ability quantum particle swarm optimization (QPSO)...

Abstract To address the limitations of computer vision-assisted table tennis ball detection, which heavily relies on vision acquisition equipment and exhibits slow processing speed, we propose a real-time calculation method for determining landing point balls. This novel approach is based spatial domain information reduces dependency equipment. incorporates several steps: employing dynamic color thresholding to determine centroid coordinates all objects in video frames, utilizing target area...

Steam condensation in steam turbines is a complex problem and its process still not clear. To investigate the turbine, measurements were carried out with an optical/pneumatic probe 300 MW direct air-cooling turbine by changing backpressure from 42 to 24 kPa. The thermodynamic states of exhaust changed superheat wet too. existence heterogeneous was demonstrated actual measurement results. found be condensation, heterogeneous/homogeneous then homogeneous condensation. mean diameter droplets...

The shortest path problem is one of the most important operation planning in traffic navigation system. To view optimization, proposed hybrid genetic algorithm combining general and pattern search method.Use simulation examples to verify feasibility effectiveness new method, experimental results show that standard deviation smaller, pr-ecision more accurate stable using method.