What is Metadata and what does it do?
Understanding the Importance of Metadata in Physical Security, Surveillance and Retail Intelligence Applications.
Though its history goes back some 15 years further, Miriam-Webster first recorded the word “Metadata” in 1983. Defined as “data that provides information about other data,” the term was coined by Philip Bagley in 1968 in his work, Extension of Programming Language Concepts. In this series of posts, we will take a close look at metadata, its rationale and its implications for physical security, surveillance and business intelligence applications.
At its core, metadata gives identity and context to electronic content in order to provide a means to organize, discover and retrieve information from digital files. Metadata defines the content, giving additional structure or, in some cases providing structure where none exists. It was initially popularized in the field of library sciences as a way to manage the digitization of card catalogues in the 1970s and 80s, and brought to the mainstream through the adoption of the internet in the 1990s. Metadata is structured through sets of schema (designed data elements), semantics (the definition of those elements) and rules (such as those for formatting, representation, encoding and allowable values).
Metadata can be classified in two categories: that which is intended solely to describe electronic resources, and that which is intended to be machine-understandable. Both types of structured data – in particular machine-understandable data – are useful when applied to digital video content. Video content, being inherently unstructured, is not searchable in its own right. Machines cannot understand video on its own. A computer does not “watch” a video and must be told what the video contains – they need metadata in order to do this. Identifiers like time, location and other application-specific attributes can help a computer handle video content. Thus making it searchable and allowing the content to be analyzed for trends.
From a surveillance perspective, this can help streamline efficiencies and enable powerful analytics which are beneficial for both physical security and retail intelligence. Metadata allows surveillance video to be searched, correlated and distinguished based on any number of criteria. It also facilitates the management of physical security. The ability to conduct this deep analysis and track variables such as people counting, license as well as the ability to receive notifications and integrate surveillance with mobile technologies are made possible only via the use and extrapolation of metadata.
As network video is poised to surpass analog video this year (and eventually replace it altogether) understanding the use of metadata and how it applies to the surveillance market is ever more important for professionals in the industry as well as buyers.
Sincerely,
Justin Schorn
VP Product Management
|