Fog computing provides users with data storage, computing, and other services by using fog layer devices close to edge devices. Tasks and resource scheduling in fog computing has become a research hotspot. For the multi-objective task-scheduling problem in fog computing, an adaptive multi-objective optimization task scheduling method for fog computing (FOG-AMOSM) is proposed in this paper. In this method, the total execution time and the task resource cost in the fog network are taken as the optimization target of resource allocation, and a multi-objective task scheduling model is designed. Since the objective model is a Pareto optimal solution problem, the global optimal solution can be obtained by using multi-objective optimization theory and the improved multi-objective evolutionary heuristic algorithm. Moreover, to obtain a better distribution of the current task scheduling group, the neighborhood is adaptively changed according to the current situation of the task scheduling group in fog computing, which avoids the problem that the neighborhood value caused by the neighborhood policy in the multi-objective algorithm affects the distribution of the task scheduling population. This algorithm is used to solve the non-inferior solution set of the utility function index of fog computing task scheduling to try to solve the multi-objective cooperative optimization problem in fog computing task scheduling. The results show that the proposed method has better performance than other methods in terms of total task execution time, resource cost and load dimensions. INDEX TERMS Cloud computing, fog computing, task scheduling, multi-objective optimization algorithm, cyber-physical-social service.

Considering the shortcomings of IDW interpolation, this study improved the IDW algorithm and proposed a new spatial interpolation method that is called Adjusted Inverse Distance Weighted (AIDW). The AIDW is capable of taking into account the comprehensive influence of relative distance and position of sample points on the interpolated point, by adding a coefficient (K) to IDW formula. The coefficient (K) is used to adjust the distance weight of sample point according to its shielded effect in sample point positions. Theoretical analysis and case study indicates that the AIDW algorithm could diminish the IDW interpolation error of nonuniform distribution of sample points, consequently the AIDW interpolating is more reasonable, compared with the IDW interpolating. On the other hand, the contour plotting of the AIDW interpolation can effectively avoid the implausible isolated circles and concentric circles that originated from the defect of the IDW interpolation. The contour derived from the AIDW interpolated surface is more similar to the professional manual identification than that from the IDW interpolated surface.

No abstract

Local features could learn semantic information for pedestrian images and they are very important for person re-identification (Re-ID) in harsh environments. However, most approaches only optimize one kind of local feature, which results in incomplete local features. In this paper, we propose Local Heterogeneous Features (LHF) to extract discriminative local features from three aspects. To this end, we utilize three kinds of losses to learn three kinds of local features, i.e., local discriminative features, local relative features, local compact features. As for local discriminative features, we split the attention maps into three horizontal sub-regions and perform the classification operation. Then, we divide the attention maps into two horizontal sub-regions, and we synchronously apply the triplet loss and center loss to learn local relative features and local compact features. Finally, we utilize local discriminative features to represent pedestrian. We evaluate LHF on public person Re-ID datasets and prove LHF is meaningful for local feature learning.INDEX TERMS Person re-identification, local heterogeneous features, harsh environments.

In the context of the strategic target of carbon emission peaking and carbon neutrality, industrial green technology innovation (GTI) has become the focus of discussion in academia these days. Based on the panel data of 30 provinces in China from 2011 to 2019, we construct Spatial Durbin Models to explore the spatial effects of capital enrichment (CE) on GTI by using the geographical distance matrix, the economic distance matrix and the adjacency matrix. The results reveal that: (1) The regional differences in the development of GTI are prominent, showing a higher level in the east and lower in the west. (2) GTI exhibits the spatial characteristic of polarization. Its spatiotemporal evolutionary pattern reveals a phased feature of first strengthened and then weakened. (3) The CE has a significant inhibitory effect on GTI, which may be caused by the “rebound effect”, dominated by short-term economic interests and the ineffective capital allocation. This effect is more prominent in regions with unbalanced economies. (4) The spatial spillover effect of CE is significantly negative, indicating a “siphon effect”. Based on these findings, the suggestions for promoting GTI are put forward.