Search

With the fast development of artificial intelligence (AI) and industrial Internet of Things (IIoT) technologies, it is challenging to deal with the problems of data privacy protection and secure computing. In recent years, federated learning (FL) is proposed to attack the challenges of learning shared models collaboratively while protecting security based on the data from cross-domain clients. However, data in the real environment is usually not independent and identically distributed (Non-IID) due to the differences in business, working environments, and data acquisition, and thus classic federated methods suffer from significant performance degradation. In this paper, a novel federated framework is proposed for secure textile fiber identification (FedTFI) via cross-domain texture representation based on high-definition fabric images. In addition to sharing the gradient of FedTFI, the local patch of feature maps between cross-domain clients is randomly exchanged to build a richer image texture feature distribution while protecting data security simultaneously for secure computing. Furthermore, a texture embedding layer is designed to provide a joint representation through similarity measure between triplet samples in low-dimensional space. To verify the effectiveness of the proposed framework, two textile image datasets, i.e., one public and the other we collected, are utilized to construct four Non-IID scenarios, including label skew, feature skew, and two combined skew scenarios. The experimental results confirm the effectiveness of our model to obtain better detection accuracies than benchmarks in four Non-IID scenarios by keeping data privacy for secure computing in fabric IIoT. [ABSTRACT FROM AUTHOR]

Similarity join on time series databases is an essential operation for data analysis applications. Due to the curse of dimensionality, it is not suitable to use traditional index techniques, such as R-tree and kd-tree. In the paper, a dynamic segment index (i.e., DSTree) is utilized to reduce the huge comparison cost on the similarity join on time series databases. However, the DSTree is designed for similarity search and only supports bound estimations between a time series and a batch of time series in a DSTree node. To make the DSTree suitable for the similarity join on time series databases, it is necessary to have tight bounds for the nodes to achieve a better pruning power, where the biggest challenge is that the DSTree nodes may have different segmentations. To solve the problem aforementioned, a segmentation alignment and synopsis evaluation method is proposed to support the estimation of DSTree nodes to significantly reduce the time cost by pruning unnecessary comparisons. Moreover, to make our approach I/O efficient, a caching strategy is proposed by taking advantage of both graph partitioning and the locality of the DSTree index. The efficiency and effectiveness of the proposed approaches are verified by experiments on real-life datasets. [ABSTRACT FROM AUTHOR]