Fusion of Static and Temporal Information for Threat Evaluation in Sensor Networks

Abstract

Abstract. In many CCTV and sensor network based intelligent surveillance systems, a number of attributes or criteria are used to individually evaluate the degree of potential threat of a suspect. The outcomes for these attributes are in general from analytical algorithms where data are often pervaded with uncertainty and incompleteness. As a result, such individual threat evaluations are often inconsistent, and individual evaluations can change as time elapses. Therefore, integrating heterogeneous threat evaluations with temporal influence to obtain a better overall evaluation is a challenging issue. So far, this issue has rarely be considered by existing event reasoning frameworks under uncertainty in sensor network based surveillance. In this paper, we first propose a weighted aggregation operator based on a set of principles that constraints the fusion of individual threat evaluations. Then, we propose a method to integrate the temporal influence on threat evaluation changes. Finally, we demonstrate the usefulness of our system with a decision support event modeling framework using an airport security surveillance scenario.

A Dempster‐Shafer theory and uninorm‐based framework of reasoning and multiattribute decision‐making for surveillance system

“…This preference ordering means that the potential threat of is higher than that of if and only if the overall threat assessment of is greater than that of . Hence, together with the equivalence relation (ie, if and ), we show that the preference ordering in Definition is a total order that satisfies the properties of completeness and strict transitivity …”

“…After obtaining the point‐valued DPTs of the watched objects regarding each attribute, to rank the potential threats of all the suspects to take the appropriate action directly, in this section we will discuss how to use a weighted aggregation operator to calculate the overall degree of a potential threat for each watched object after considering all relevant attributes. And Table lists notations used in Ma et al's paper, where DPT involves a given attribute for watched object .…”

“…The reason why we suggest to employ the weighted uninorm that it satisfies all the following basic principles that a threat assessment aggregation needs to follow, as argued by Ma, Liu, and Hong: Conclusion modification: Consider one more threat assessment for a given watched object regarding a new attribute can increase, or decrease, or remain the current assessment. Assessment consistency: The overall potential threat assessment for a given watched object should increase when the point‐valued DPT of a watched object regarding each relevant attribute increases. Assessment commensurability: A system needs to provide an overall assessment for each watched object after the aggregation process, and the overall assessments for different watched objects are comparable (on a commensurable scale). Irrelevance of evidence ordering: The result of an aggregation should not be affected by the ordering of aggregation. Importance dependency: The overall assessment depends on the importance (reflected as a weight) of each attribute.…”