Arresting the storage challenges of body cameras
President Obama has requested $75 million over three years to fund the purchase of body cameras for police officers in his recent budget – with an objective to make the use of body cameras a core part of day-to-day law enforcement operations.
While the legislative outcome will no doubt remain a matter of debate, body cameras are gaining favor due to studies indicating that wearing cameras are associated with dramatic reductions in use-of-force and complaints against officers.
The University of Cambridge’s Institute of Criminology conducted a 2014 study, for example, that found body-worn cameras reduce the use of force by roughly 50 percent and that complaints against police dropped 90 percent year-over-year.
In order for body cameras to deliver on their full potential, however, local law enforcement agencies must grapple with a substantial increase in the amount of video footage that needs to be processed, stored and accessed for both internal and external use. The task is even more difficult because of the unstructured nature of this type of raw video. The challenge comes down to data storage: Who will be responsible for downloading the video? How will it be stored and for how long? What about its use in criminal proceedings – will it be accepted in a court trial? Will it be considered physical evidence and require chain-of-custody handling or other special types of controls? Every law enforcement agency needs to carefully consider how they will use this footage and how they will store and protect it.
Create a chain of evidence
Once an officer equipped with a body camera returns to the station, there needs to be a clear system in place for how that video is handled.
Tampering: Law enforcement agencies will need a system that includes technological safeguards that eliminate the possibility of tampering with the footage.
Downloading: Clear criteria should be established as to who is responsible for physically taking possession of the camera and downloading the video to the storage system.
Auditing: A key component of any storage system is a record of who accesses the video data, when they access it and why. This trail of access is especially vital as it relates to usability in court proceedings. An audit function must be part of any defined storage system.
Backup: Whether backup takes place manually – through copying files downloaded from the camera onto a disk that is stored on premise or using a body camera system with a backup automatically built into the system – having such protection is essential.
Indexing: Once the data has been downloaded, it must be indexed to create an easily retrievable file for police who need to present video in court or in response to public records requests.
Metadata: Video metadata boils down to two types: automatically collected or manually written. Automatically collected video metadata includes routine data collected from the body camera itself – date, time, location – if GPS is enabled.
Manually written metadata allows law enforcement agencies to assign text tags to footage to aid in its storage and retrieval. Agencies need a system that makes full of use of the manually written metadata and catalogs each clip with regard to assigned values such as dialog, officers involved, radio call number and more. Proper use of metadata from the beginning will alleviate retrieval headaches in the future.
Cloud vs on-premises storage
Once the video files have been retrieved from the camera and downloaded into the storage system, the agency must consider how to store all of the additional video data.
Most daily footage will be non-evidentiary videos that will play no part in an investigation or prosecution. Videos from these encounters, such as aiding a stranded motorist, will not require long- term storage, making it a good candidate for cloud storage. Many agencies may also assign a shorter storage retention time for such videos – the most common retention time for non-evidentiary video is between 60 to 90 days – also significantly reducing data storage costs.
Evidentiary footage will need to be stored for longer, and it will often have to adhere to more stringent regulations affecting access, making on-premises storage a more viable option. These evidentiary videos should be separated according to the seriousness of the crime. The videos surrounding minor infractions, such as routine traffic tickets, would not have to be retained as long as the video from more serious incidents, such as a homicide. This variation in on-premises retention times equates to significant cost differentials.
Agencies moving toward body cameras need to take all of these considerations into account before they implement their new video requirements. If done properly, they will be saving time and money as well as a large number of headaches.
Raj Rana is senior manager, systems engineering, for NetApp.