Large-scale applications of Internet of things (IoT), which require considerable computing tasks and storage resources, are increasingly deployed in cloud environments. Compared with the traditional computing model, characteristics of the cloud such as pay-as-you-go, unlimited expansion, and dynamic acquisition represent different conveniences for these applications using the IoT architecture. One of the major challenges is to satisfy the quality of service requirements while assigning resources to tasks. In this paper, we propose a deadline and cost-aware scheduling algorithm that minimizes the execution cost of a workflow under deadline constraints in the infrastructure as a service (IaaS) model. Considering the virtual machine (VM) performance variation and acquisition delay, we first divide tasks into different levels according to the topological structure so that no dependency exists between tasks at the same level. Three strings are used to code the genes in the proposed algorithm to better reflect the heterogeneous and resilient characteristics of cloud environments. Then, HEFT is used to generate individuals with the minimum completion time and cost. Novel schemes are developed for crossover and mutation to increase the diversity of the solutions. Based on this process, a task scheduling method that considers cost and deadlines is proposed. Experiments on workflows that simulate the structured tasks of the IoT demonstrate that our algorithm achieves a high success rate and performs well compared to state-of-the-art algorithms.

No abstract

Weakly supervised semantic segmentation under image-level annotations is effectiveness for real-world applications. The small and sparse discriminative regions obtained from an image classification network that are typically used as the important initial location of semantic segmentation also form the bottleneck. Although deep convolutional neural networks (DCNNs) have exhibited promising performances for single-label image classification tasks, images of the real-world usually contain multiple categories, which is still an open problem. So, the problem of obtaining high-confidence discriminative regions from multi-label classification networks remains unsolved. To solve this problem, this article proposes an innovative three-step framework within the perspective of multi-object proposal generation. First, an image is divided into candidate boxes using the object proposal method. The candidate boxes are sent to a single-classification network to obtain the discriminative regions. Second, the discriminative regions are aggregated to obtain a high-confidence seed map. Third, the seed cues grow on the feature maps of high-level semantics produced by a backbone segmentation network. Experiments are carried out on the PASCAL VOC 2012 dataset to verify the effectiveness of our approach, which is shown to outperform other baseline image segmentation methods.

No abstract