An Efficient Density-based clustering algorithm for face groping

A portable clustering algorithm based on compact neighbors for face tagging

We focus on the following problem: Given a collection of unlabeled facial images, group them into the individual identities where the number of subjects is not known. To this end, a Portable clustering algorithm based on Compact Neighbors called PCN is proposed. (1) Benefiting from the compact neighbor, the local density of each sample can be determined automatically and only one user-specified parameter, the number of nearest neighbors k, is involved in our model. (2) More importantly, the performance of PCN is not sensitive to the number of nearest neighbors. Therefore this parameter is relatively easy to determine in practical applications. (3) The computational overhead of PCN is O(nk(k²+log(nk))) that is nearly linear with respect to the number of samples, which means it is easily scalable to large-scale problems. In order to verify the effectiveness of PCN on the face clustering problem, extensive experiments based on a two-stage framework (extracting features using a deep model and performing clustering in the feature space) have been conducted on 16 middle- and 5 large-scale benchmark datasets. The experimental results have shown the efficiency and effectiveness of the proposed algorithm, compared with state-of-the-art methods. [code].

Robot location privacy protection based on Q-learning particle swarm optimization algorithm in mobile crowdsensing

In the recent years, with the rapid development of science and technology, robot location-based service (RLBS) has become the main application service on mobile intelligent devices. When people use location services, it generates a large amount of location data with real location information. If a malicious third party gets this location information, it will cause the risk of location-related privacy disclosure for users. The wide application of crowdsensing service has brought about the leakage of personal privacy. However, the existing privacy protection strategies cannot adapt to the crowdsensing environment. In this paper, we propose a novel location privacy protection based on the Q-learning particle swarm optimization algorithm in mobile crowdsensing. By generalizing tasks, this new algorithm makes the attacker unable to distinguish the specific tasks completed by users, cuts off the association between users and tasks, and protects users' location privacy. The strategy uses Q-learning to continuously combine different confounding tasks and train a confounding task scheme that can output the lowest rejection rate. The Q-learning method is improved by particle swarm optimization algorithm, which improves the optimization ability of the method. Experimental results show that this scheme has good performance in privacy budget error, availability, and cloud timeliness and greatly improves the security of user location data. In terms of inhibition ratio, the value is close to the optimal value.