With the increasing demand for greater security, video surveillance technologies have recently received a lot of attention. As video surveillance cameras become ubiquitous, there are growing concerns over the cost of monitoring these systems and the possible invasion of privacy. In this paper, we discuss a system architecture of privacy-preserving video surveillance for a community that achieves a good balance between security and privacy. We call the designed system 'PriSurv'. A subset of PriSurv is implemented, and privacy protected image processing modules are installed. The privacy preserving image processing is called visual abstraction. We conducted an experiment using five abstracted images and a questionnaire to evaluate PriSurv. Analyzing the results of the experiment using factor analysis, we obtained seven factors. With a hierarchical cluster analysis, we divided the subjects into three clusters. A significant difference in the choice of images among the clusters was found using the Pearson χ-squared test. From this analysis, we show that the relationships between subjects and monitors affect the subjects' sense of privacy; furthermore, the subjects' sense of privacy depends on the individual person. The results of analysis supports PriSurv's design principle, which can adapt the personal sense of privacy.
We propose an advanced visual hull technique to compensate for outliers using the reliabilities of the silhouettes. The proposed method consists of a foreground extraction technique based on the Generalized Gaussian Family model and a compensated shape-from-silhouette algorithm. They are connected by the intra-/inter-silhouette reliabilities to compensate for carving errors from defective segmentation or partial occlusion which may occur in a real environment. The 3D reconstruction process is implemented on a graphics processing unit (GPU) to accelerate the processing speed by using the huge computational power of modern graphics hardware. Experimental results show that the proposed method provides reliable silhouette information and an accurate visual hull in real environments at a very high speed on a common PC.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.