Search

Abstract Deep learning methods, particularly Convolutional Neural Network (CNN), have been widely used in hyperspectral image (HSI) classification. CNN can achieve outstanding performance in the field of HSI classification due to its advantages of fully extracting local contextual features of HSI. However, CNN is not good at learning the long-distance dependency relation and dealing with the sequence properties of HSI. Thus, it is difficult to continuously improve the performance of CNN-based models because they cannot take full advantage of the rich and continuous spectral information of HSI. This paper proposes a new Double-Branch Feature Fusion Transformer model for HSI classification. We introduce Transformer into the process of HSI on account of HSI with sequence characteristics. The two branches of the model extract the global spectral features and global spatial features of HSI respectively, and fuse both spectral and spatial features through a feature fusion layer. Furthermore, we design two attention modules to adaptively adjust the importance of spectral bands and pixels for classification in HSI. Experiments and comparisons are carried out on four public datasets, and the results demonstrate that our model outperforms any compared CNN-Based models in terms of accuracy.

Most deep learning methods in hyperspectral image (HSI) classification use local learning methods, where overlapping areas between pixels can lead to spatial redundancy and higher computational cost. This paper proposes an efficient global learning (EGL) framework for HSI classification. The EGL framework was composed of universal global random stratification (UGSS) sampling strategy and a classification model BrsNet. The UGSS sampling strategy was used to solve the problem of insufficient gradient variance resulted from limited training samples. To fully extract and explore the most distinguishing feature representation, we used the modified linear bottleneck structure with spectral attention as a part of the BrsNet network to extract spectral spatial information. As a type of spectral attention, the shuffle spectral attention module screened important spectral features from the rich spectral information of HSI to improve the classification accuracy of the model. Meanwhile, we also designed a double branch structure in BrsNet that extracted more abundant spatial information from local and global perspectives to increase the performance of our classification framework. Experiments were conducted on three famous datasets, IP, PU, and SA. Compared with other classification methods, our proposed method produced competitive results in training time, while having a greater advantage in test time.

Objective The study aimed to determine if and how environmental factors correlated with asthma admission rates in geographically different parts of Guangxi province in China.Setting Guangxi, China.Participants This study was done among 7804 asthma patients.Primary and secondary outcome measures Spearman correlation coefficient was used to estimate correlation between environmental factors and asthma hospitalisation rates in multiple regions. Generalised additive model (GAM) with Poisson regression was used to estimate effects of environmental factors on asthma hospitalisation rates in 14 regions of Guangxi.Results The strongest effect of carbon monoxide (CO) was found on lag1 in Hechi, and every 10 µg/m3 increase of CO caused an increase of 25.6% in asthma hospitalisation rate (RR 1.26, 95% CI 1.02 to 1.55). According to the correlation analysis, asthma hospitalisations were related to the daily temperature, daily range of temperature, CO, nitrogen dioxide (NO2) and particulate matter (PM2.5) in multiple regions. According to the result of GAM, the adjusted R2 was high in Beihai and Nanning, with values of 0.29 and 0.21, which means that environmental factors are powerful in explaining changes of asthma hospitalisation rates in Beihai and Nanning.Conclusion Asthma hospitalisation rate was significantly and more strongly associated with CO than with NO2, SO2 or PM2.5 in Guangxi. The risk factors of asthma exacerbations were not consistent in different regions, indicating that targeted measures should differ between regions.

As a class of hard combinatorial optimization problems, the school bus routing problem has received considerable attention in the last decades. For a multi-school system, given the bus trips for each school, the school bus scheduling problem aims at optimizing bus schedules to serve all the trips within the school time windows. In this paper, we propose two approaches for solving the bi-objective school bus scheduling problem: an exact method of mixed integer programming (MIP) and a metaheuristic method which combines simulated annealing with local search. We develop MIP formulations for homogenous and heterogeneous fleet problems respectively and solve the models by MIP solver CPLEX. The bus type-based formulation for heterogeneous fleet problem reduces the model complexity in terms of the number of decision variables and constraints. The metaheuristic method is a two-stage framework for minimizing the number of buses to be used as well as the total travel distance of buses. We evaluate the proposed MIP and the metaheuristic method on two benchmark datasets, showing that on both instances, our metaheuristic method significantly outperforms the respective state-of-the-art methods.

In the application field of fresh products distribution, it is necessary to use multi-compartment vehicles for distribution because of their particular demands for temperature. This paper studied the multi-compartment vehicle routing problem with soft time windows for multiple fresh products distribution. Firstly, a mathematical model of the issued problem was built, which aims to minimize the total cost including vehicle cost, delivery cost, refrigeration cost, damage cost and penalty cost of delivery time. Then, we presented an improved particle swarm optimization algorithm to solve this problem. In the process of particle updating, the sequential crossover operator usually used in genetic algorithm was introduced to enhance the diversity of particles. Finally, the proposed algorithm was evaluated on some benchmark instances, and the experiment results demonstrate its effectiveness and good stability, when compared with genetic algorithm and simulated annealing algorithm. It can draw a conclusion that the proposed algorithm can provide a reliable and stable solution approach for the distribution of fresh products. [ABSTRACT FROM AUTHOR]

This paper deals with multi-objective school bus routing problem, which includes route balance, total number of school buses and total travel distance optimization objectives. An improved non-dominated sorting genetic algorithm (NSGAII) is proposed to solve this problem. First, the definition of measurement indicator that denotes the degree of route balance is given based on the analysis of the solved problem. And then, the multi-objective optimization function is provided. In the proposed algorithm, the individuals are obtained by using the tournament selection, sequential crossover and inverse mutation. The 2-opt neighborhood operator is adopted to improve the best individuals obtained in each iteration. At the same time, the route selection rule based on the degree of route balance is applied to select the final optimal solution set. The solution with better balance degree will be taken as the best solution. Finally, some benchmark instances are used to test the effectiveness of proposed algorithm. The results reveal that the proposed algorithm outperforms the standard NSGAII and Multi-objective Evolutionary Algorithm(MOEA). The experimental results also show that our algorithm has good stability. [ABSTRACT FROM AUTHOR]

School bus service is divided into many kinds of operation modes, such as single school, single load multiple schools and mixed load multiple schools and so on. When it comes to plan bus routes satisfying various constraints and different objectives, there are many applications of school bus routing problem (SBRP). It is generally recognized that middle and large sacle SBRP applications are resolved by heuristic algorithms, because SBRP is a NP-hard problem. This paper proposes a local search-based metaheuristic algorithm framework for SBRP on the basis of analysis of SBRP problem models. The framework, which is implemented by C#, is made up of basic data structures, operation functions, neighborhood operators, initial solution construction algorithm components and heuristic strategies, etc. The neighborhoodcentered metaheuristic algorithms, such as simulated annealing (SA), iterated local search (ILS), variable neighborhood search (VNS), can be designed based on the framework to solve a certain problem type of SBRP. Using this framework, we can solve three operation modes SBRP with homogeneous or heterogeneous bus fleet. Moreover, the proposed metaheuristic framework also provides the development of metaheuristic for the capacitated vehicle routing problem(CVRP) by reducing or modifying the constraints of SBRP. The experimental results of a set of instances show that the metaheuristic algorithm built based on the proposed framework, can be quickly realized and applied to different SBRP applications. Meanwhile, the designed metaheuristic framework is also very effective and extensible. [ABSTRACT FROM AUTHOR]

In the practice of multi-school bus route planning, the characteristics of the bus fleet and the operation modes of buses have produced a variety of bus planning scenarios. In recent years, many methods have been used to arrange the routes of school buses for the school bus routing problem (SBRP) with a specific planning scenario. However, it is still a challenging task to develop a general-purpose algorithm that can effectively apply to the bus route planning for a variety of bus planning scenarios. This paper tries to develop a hybrid iterated local search (ILS) metaheuristic algorithm for SBRP with multiple planning scenarios, which including homogenous or heterogeneous fleets, single load or mixed load operation modes. Within the framework of ILS, a variety of neighborhood structures are used to improve the initial solution as well as the routes generated in the process of local search are recorded. For heterogeneous SBRP, the fleet adjustment strategy based on route segments is also applied. In addition, the perturbation mechanism and the acceptance of worse solutions within a certain deviation range are adopted to enhance the diversity of solutions. Finally, the local best solution is further promoted by the set partitioning procedure (SP), which is modeled by the history routes in the local search process. The experiment results prove that the proposed algorithm is effective and it also outperforms the existing algorithms for multi-school SBRP. [ABSTRACT FROM AUTHOR]