As a generalization of the fuzzy set and intuitionistic fuzzy set, the neutrosophic set (NS) have been developed to represent uncertain, imprecise, incomplete and inconsistent information existing in the real world. Now the interval neutrosophic set (INS) which is an expansion of the neutrosophic set have been proposed exactly to address issues with a set of numbers in the real unit interval, not just one specific number. After definition of concepts and operations, INS is applied to image segmentation. Images are converted to the INS domain, which is described using three membership interval sets: T, I and F. Then, in order to increase the contrast between membership and evaluate the indeterminacy, a fuzzy intensification for each element in the interval set is made and a score function in the INS is defined. Finally, the proposed method is employed to perform image segmentation using the traditional kmeans clustering. The experimental results on a variety of images demonstrate that the proposed approach can segment different sorts of images. Especially, it can segment "clean" images and images with various levels of noise.

The team RY presents a solution for Sussex-Huawei Locomotion-Transportation (SHL) recognition challenge, which aims at differentiating eight transportation modes with mobile phone signal sensor data in this paper. This study first extracted a set of reasonable and discriminative features after data-preprocessing. Then, decision tree bagging, random forest, lightGBM are trained separately as basic models, whose predictions are integrated and afterward smoothed. The method gets 0.65 accuracy score on validation dataset. CCS CONCEPTS• Computing methodologies → Artificial intelligence; Simulation environments; • General and reference → Cross-computing tools and techniques.

The segmentation of high-grade gliomas (HGG) using magnetic resonance imaging (MRI) data is clinically meaningful in neurosurgical practice, but a challenging task. Currently, most segmentation methods are supervised learning with labeled training sets. Although these methods work well in most cases, they typically require time-consuming manual labeling and pre-trained models. In this work, we propose an automatically unsupervised segmentation toolbox based on the clustering algorithm and morphological processing, named AUCseg. With our toolbox, the whole tumor was first extracted by clustering on T2-FLAIR images. Then, based on the mask acquired with whole tumor segmentation, the enhancing tumor was segmented on the post-contrast T1-weighted images (T1-CE) using clustering methods. Finally, the necrotic regions were segmented by morphological processing or clustering on T2-weighted images. Compared with K-means, Mini-batch K-means, and Fuzzy C Means (FCM), the Gaussian Mixture Model (GMM) clustering performs the best in our toolbox. We did a multi-sided evaluation of our toolbox in the BraTS2018 dataset and demonstrated that the whole tumor, tumor core, and enhancing tumor can be automatically segmented using default hyper-parameters with Dice score 0.8209, 0.7087, and 0.7254, respectively. The computing time of our toolbox for each case is around 22 seconds, which is at least 3 times faster than other state-of-the-art unsupervised methods. In addition, our toolbox has an option to perform semi-automatic segmentation via manually setup hyper-parameters, which could improve the segmentation performance. Our toolbox, AUCseg, is publicly available on Github. (https://github.com/Haifengtao/AUCseg).

In this paper, a JPEG steganalysis algorithm based on locality preserving projection (LPP) dimensionality reduction method is proposed for detecting the unseen stego algorithms. The co-occurrence features are extracted from DCT-DWT domain and dimension is reduced by using the LPP method. For improving the generalization capability of the algorithm, SVDD is used as the classifier. Experimental results reveal the fact that our scheme has better generalization capability and is more effective than others.