Please refer to published version for the most recent bibliographic citation information. If a published version is known of, the repository item page linked to above, will contain details on accessing it.

Recent years have witnessed a big leap in automatic visual saliency detection attributed to advances in deep learning, especially Convolutional Neural Networks (CNNs). However, inferring the saliency of each image part separately, as was adopted by most CNNs methods, inevitably leads to an incomplete segmentation of the salient object. In this paper, we describe how to use the property of part-object relations endowed by the Capsule Network (CapsNet) to solve the problems that fundamentally hinge on relational inference for visual saliency detection. Concretely, we put in place a two-stream strategy, termed Two-Stream Part-Object RelaTional Network (TSPORTNet), to implement CapsNet, aiming to reduce both the network complexity and the possible redundancy during capsule routing. Additionally, taking into account the correlations of capsule types from the preceding training images, a correlation-aware capsule routing algorithm is developed for more accurate capsule assignments at the training stage, which also speeds up the training dramatically. By exploring part-object relationships, TSPORTNet produces a capsule wholeness map, which in turn aids multi-level features in generating the final saliency map. Experimental results on five widely-used benchmarks show that our framework consistently achieves state-of-the-art performance. The code can be found on https://github.com/liuyi1989/TSPORTNet.

We address the problem of large-scale annotation of web images. Our approach is based on the concept of visual synset, which is an organization of images which are visually-similar and semantically-related. Each visual synset represents a single prototypical visual concept, and has an associated set of weighted annotations. Linear SVM's are utilized to predict the visual synset membership for unseen image examples, and a weighted voting rule is used to construct a ranked list of predicted annotations from a set of visual synsets. We demonstrate that visual synsets lead to better performance than standard methods on a new annotation database containing more than 200 million images and 300 thousand annotations, which is the largest ever reported.

Vibration signal analysis is one of the most effective methods for mechanical fault diagnosis. Available part of the information is always concealed in component noise, which makes it much more difficult to detect the defection, especially at early stage of the development. This paper presents a new approach for mechanical fault diagnosis based on time domain analysis and adaptive fuzzyC-means clustering. By analyzing vibration signal collected, nine common time domain parameters are calculated. This lot of data constitutes data matrix as characteristic vectors to be detected. And using adaptive fuzzyC-means clustering, the optimal clustering number can be gotten then to recognize different fault types. Moreover, five parameters, including variance, RMS, kurtosis, skewness, and crest factor, of the nine are selected as the new eigenvector matrix to be clustered for more optimal clustering performance. The test results demonstrate that the proposed approach has a sensitive reflection towards fault identifications, including slight fault.

No abstract