Multispectral (MS) pansharpening is crucial to improve the spatial resolution of MS images. MS pansharpening has the potential to provide images with high spatial and spectral resolutions. Pansharpening technique based on deep learning is a topical issue to deal with the distortion of spatio-spectral information. To improve the preservation of spatio-spectral information, we propose a novel three-stage detail injection pansharpening network (TDPNet) for remote sensing images. First, we put forward a dual-branch multiscale feature extraction block, which extracts four scale details of panchromatic (PAN) images and the difference between duplicated PAN and MS images. Next, cascade cross-scale fusion (CCSF) employs fine-scale fusion information as prior knowledge for the coarse-scale fusion to compensate for the lost information during downsampling and retain high-frequency details. CCSF combines the fine-scale and coarse-scale fusion based on residual learning and prior information of four scales. Last, we design a multiscale detail compensation mechanism and a multiscale skip connection block to reconstruct injecting details, which strengthen spatial details and reduce parameters. Abundant experiments implemented on three satellite data sets at degraded and full resolutions confirm that TDPNet trades off the spectral information and spatial details and improves the fidelity of sharper MS images. Both the quantitative and subjective evaluation results indicate that TDPNet outperforms the compared state-of-the-art approaches in generating MS images with high spatial resolution.
In this paper, we develop three supply chain game models, i.e., the basic model, the single trade credit model, and the trade credit and revenue sharing collaboration model. Conditional value-at-risk (CVaR) criterion is used as the measure of risk assessment in these models. We analyze the optimal decisions in the centralized and decentralized situations, respectively, and verify that single trade credit cannot coordinate the supply chain. However, the collaboration contract can coordinate the supply chain. Furthermore, this paper explores the influence of risk-aversion factor, trade credit period, revenue sharing coefficient, and other parameters on the optimal decisions and studies the feasible range of Pareto improvement in the collaborative model. In numerical experiments, the results show that the decisions and profits of both the manufacturer and the retailer reply on the degree of the risk aversion, the trade credit period, and the revenue sharing coefficient. The collaborative contract effectively improves supply chain performance and achieves a ‘win-win’ situation for the supply chain members. In addition, we also consider two extensions for our research. One extension shows that the collaborative contract of trade credit and buyback can also coordinate the supply chain in a certain range. The other extension considers the optimal decision of a risk-averse manufacturer with CVaR.
Naive-Bayes algorithm acts as a key baseline of massive text classification, which is widely used in fields of detecting spam, online marketing and so on. Multicore processor is a suitable platform to implement Naive-Bayes because of its flexibility, high performance, and energy-efficiency. This paper proposes a new hopscotch hash scheme to improve the performance of data storing and indexing of Naive-Bayes algorithm, and presents a software implementation of Naive-Bayes text classification mapped in Topo-MapReduce model on a multicore processor with circuit switching and packet switching. Experimental results show that the improved hopscotch hash speeds up by 33% at maximum compared to the original hash, and the proposed Topo-MapReduce speeds up the Naive-Bayes algorithm by 29% at maximum compared to the original MapReduce.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.