IEEE 2017-2018 Image Processing & Multimedia Projects in DotNet

Abstract:

In the study of Location-Based Social Network (LBSN) sign-in data as the recommended point of interest for groups, there are some problems such as poor recommendation accuracy and high bias in recommendation results because of the unbalanced number and diversity of individual sign-in and the different degree of group user association. In this paper, a new group recommendation model is proposed. Firstly, the existing individual recommendation model is combined with the text retrieval idea and the threshold function to improve the user rating strategy. Secondly, the recommendation strategy is used to aggregate the individual recommendation list. Considering users' friends relationship, similarity and frequency of sign-in, lead into the user gregariousness weight and activity weight, and form a new group user preference model to make recommendation. The experimental results show that the improved scoring strategy can improve the accuracy of recommendation, and the new group weighting model which recommend the points of interest for the groups can improve the recommendation quality by reducing the recommended deviation.

Abstract:

Low-rank restoration has recently attracted a lot of attention in the research of computer vision. Empirical studies show that exploring the low-rank property of the patch groups can lead to superior restoration performance, however, there is limited achievement on the global low-rank restoration because the rank minimization at image level is too strong for the natural images which seldom match the low-rank condition. In this paper, we describe a flexible global low-rank restoration model which introduces the local statistical properties into the rank minimization. The proposed model can effectively recover the latent global low-rank structure via nuclear norm, as well as the fine details via Gaussian mixture model. An alternating scheme is developed to estimate the Gaussian parameters and the restored image, and it shows excellent convergence and stability. Besides, experiments on image and video sequence datasets show the effectiveness of the proposed method in image inpainting problems.

Abstract:

Image sharing is one of the most attractive features facilitated by different social media sites such as Facebook, Flickr, Pinterest, and Instagram. People frequently use these social media sites to express various aspects of their life with peers they are connected through these sites. The service providers of these sites sometimes use the image features for social discovery such as friend recommendation, group or community recommendation, etc. As images are rich in content and more expressive, it also reveals much sensitive information about a user and impedes their privacy. Due to storage constraints, many popular social media sites prefer to outsource their data to the cloud server. However, if the cloud server gets compromised, then an adversary can use these sensitive images for malicious purposes. In this paper, we propose a privacy-preserving image-centric social discovery framework using the neural network and efficient anonymization scheme based on optimum feature selection. Experimental results show that our proposed approach provides better accuracy than existing method as well as is scalable for big datasets.

Abstract:

Docker containers run from Docker images, which can be distributed through so-called Docker registries. The currently available support for searching images in registries is however limited. Available registries (e.g., Docker Hub) only permit searching images "by name", i.e. by specifying a term occurring in the image name, in the image description or in the name of the user that created such image. In this paper we try to enhance the support for discovering Docker images by introducing DockerFinder, a microservice-based prototype that permits searching for images based on multiple attributes, e.g., image name, image size, or supported software distributions. DockerFinder crawls images from a remote Docker registry, it automatically analyses such images to produce multi-attribute descriptions to be stored in a local repository, and it permits searching for images by querying the local repository.

Abstract:

Social media sharing Websites allow users to annotate images with free tags, which significantly contribute to the development of the web image retrieval. Tag-based image search is an important method to find images shared by users in social networks. However, how to make the top ranked result relevant and with diversity is challenging. In this paper, we propose a topic diverse ranking approach for tag-based image retrieval with the consideration of promoting the topic coverage performance. First, we construct a tag graph based on the similarity between each tag. Then, the community detection method is conducted to mine the topic community of each tag. After that, inter-community and intra-community ranking are introduced to obtain the final retrieved results. In the inter-community ranking process, an adaptive random walk model is employed to rank the community based on the multi-information of each topic community. Besides, we build an inverted index structure for images to accelerate the searching process. Experimental results on Flickr data set and NUS-Wide data sets show the effectiveness of the proposed approach.

Abstract:

Online social networks (OSNs) gradually integrate financial capabilities by enabling the usage of real and virtual currency. They serve as new platforms to host a variety of business activities, such as online promotion events, where users can possibly get virtual currency as rewards by participating in such events. Both OSNs and business partners are significantly concerned when attackers instrument a set of accounts to collect virtual currency from these events, which make these events ineffective and result in significant financial loss. It becomes of great importance to proactively detecting these malicious accounts before the online promotion activities and subsequently decreases their priority to be rewarded. In this paper, we propose a novel system, namely ProGuard, to accomplish this objective by systematically integrating features that characterize accounts from three perspectives including their general behaviors, their recharging patterns, and the usage of their currency. We have performed extensive experiments based on data collected from the Tencent QQ, a global leading OSN with built-in financial management activities. Experimental results have demonstrated that our system can accomplish a high detection rate of 96.67% at a very low false positive rate of 0.3%

Abstract:

Recently, advances in intelligent mobile device and positioning techniques have fundamentally enhanced social networks, which allows users to share their experiences, reviews, ratings, photos, check-ins, etc. The geographical information located by smart phone bridges the gap between physical and digital worlds. Location data functions as the connection between user's physical behaviors and virtual social networks structured by the smart phone or web services. We refer to these social networks involving geographical information as location-based social networks (LBSNs). Such information brings opportunities and challenges for recommender systems to solve the cold start, sparsity problem of datasets and rating prediction. In this paper, we make full use of the mobile users’ location sensitive characteristics to carry out rating prediction. We mine: 1) the relevance between user's ratings and user-item geographical location distances, called as user-item geographical connection, 2) the relevance between users’ rating differences and user-user geographical location distances, called as user-user geographical connection. It is discovered that humans’ rating behaviors are affected by geographical location significantly. Moreover, three factors: user-item geographical connection, user-user geographical connection, and interpersonal interest similarity, are fused into a unified rating prediction model. We conduct a series of experiments on a real social rating network dataset Yelp. Experimental results demonstrate that the proposed approach outperforms existing models.

Abstract:

The past decade has witnessed the emergence and progress of multimedia social networks (MSNs), which have explosively and tremendously increased to penetrate every corner of our lives, leisure and work. Moreover, mobile Internet and mobile terminals enable users to access to MSNs at anytime, anywhere, on behalf of any identity, including role and group. Therefore, the interaction behaviors between users and MSNs are becoming more comprehensive and complicated. This paper primarily extended and enriched the situation analytics framework for the specific social domain, named as SocialSitu, and further proposed a novel algorithm for users’ intention serialization analysis based on classic Generalized Sequential Pattern (GSP). We leveraged the huge volume of user behaviors records to explore the frequent sequence mode that is necessary to predict user intention. Our experiment selected two general kinds of intentions: playing and sharing of multimedia, which are the most common in MSNs, based on the intention serialization algorithm under different minimum support threshold (Min_Support). By using the users’ microscopic behaviors analysis on intentions, we found that the optimal behavior patterns of each user under the Min_Support, and a user’s behavior patterns are different due to his/her identity variations in a large volume of sessions data.