October 1, 2015
By Jeff Gould, SafeGov.
From police chiefs and police unions to civil society groups and political leaders in the UK and elsewhere, a consensus is emerging that we need to deploy police body-worn cameras fast as possible to make policing safer and fairer for all.
But there are more than 100,000 police officers across the UK who might ultimately wear such cameras, including as part of the Metropolitan Police’s recent commitment to issue 20,000 devices by early 2016. Deploying these devices on a vast scale will be extremely challenging for both technical and policy reasons.
Only the cloud is big enough to store all the video police cameras will generate
The most obvious problem is where to put all the video the devices will produce. A single officer equipped with a camera may generate a terabyte of data per year. Even more data will come from dash cams, interrogation room cameras and fixed surveillance devices.
The only realistic place to put all this video is in the cloud. Many police forces using body cameras today operate storage systems on their own premises. But it’s clear that this approach can’t scale. The leading camera vendors like Taser and Vievu already work with cloud providers like Amazon and Microsoft to bundle “all you can eat” cloud storage services with their devices. Costs for a barebones configuration presently run $50 or more per month per user in the U.S., including both the device and cloud storage.
Police forces will struggle to manage huge video data volumes – automation will be essential
Ensuring that police and other law enforcement agencies procure cloud services that adhere to strict security and privacy requirements is imperative. In the US, the Federal Bureau of Investigation requires compliance with Criminal Justice Information Services (CJIS) standards. To aid the procurement of cloud services by police, the International Association of Chiefs of Police (IACP) recently published their “Guiding Principles on Cloud Computing in Law Enforcement”. In the UK there are similarly strict access and security rules, including for the recently created National Police Database, however the impact on rules for third party cloud storage systems appears less clear. Territorial police forces that retain their own procurement units could benefit from more uniform guidance. Further, moving towards harmonized international standards could be one way to increase the effectiveness of cross-border law enforcement efforts while maintaining a high level of security requirements.
But beyond standards for procurement of storage solutions, a much bigger problem looms on the horizon. The purpose of body-worn cameras is not to fill petabytes and exabytes of disk space in football-field-size data centres. The goal is to improve interactions between the police and the public they serve. To justify its cost, police forces must be able to filter through footage quickly and effectively. They will need to review it for investigative, training and disciplinary purposes. They will need to share it with fellow criminal justice agencies, such as the Crown Prosecution Service, as well as investigative bodies and legal professionals. Last but not least, they will need disclose it – at least selectively – to the public and the media. All of this will have to happen while guaranteeing chains of custody, ensuring that only authorized users have access, and protecting the privacy of citizens and officers.
The fundamental problem that police forces gathering large amounts of video face is that the daily tasks they need to perform with this video are very labour-intensive. Searching through thousands of hours of video, transcribing and indexing what is said in them, blurring the faces and other identifying information of citizens or officers to protect their privacy – these tasks are simply impossible to perform at scale without assistance from powerful automation tools.
The only way to manage such immense quantities of data is with machine learning, an advanced form of software that can perform tasks that until now only humans could do. Examples include understanding speech and recognizing human faces or other complex shapes. Researchers are already working to bring machine learning tools to law enforcement bodies to help manage their body-worn camera footage. While many firms are working in this area, here we consider two examples under development by Microsoft.
Machine learning will automate the management of police video at scale
A first example of what machine learning can do is the automated transcription and indexing of the spoken words captured in police videos. Police forces that accumulate thousands of hours of video archives need practical ways of searching this video or it will be of little use. Linking a time-indexed transcript of spoken words with the video stream makes search fast and accurate. In the next few years scene analysis algorithms will go a step further and automatically generate simple textual descriptions of the objects, people and events recorded by videos.
Another example is automated redaction of sensitive information in images prior to their public release. The consensus is that body-cameras and other forms of police video should not be broadcast to the world without safeguards. But what will the rules be and who will make them? Civil society advocates such as Big Brother Watch rightly insist that the privacy of citizens captured on video must be protected. Police leadership and even UK Surveillance Camera Commissioner Tony Porter have also raised valid questions of their own about the privacy of both citizens and officers.
Machine learning offers a powerful and radical alternative to labour-intensive video redaction tools. Users simply tell the software what they want to redact – such as human faces or license plates – and let the technology do the work automatically. The savings in time and labour resulting from this approach are dramatic.
Automatic redaction of faces in police video
Software by itself cannot solve complex social problems. But it can help ensure that the coming mass deployment of body-worn cameras makes policing more effective and safer for all concerned. As elected officials, police forces and communities across the UK consider the policy challenges of police video, they should take the time to educate themselves about the new possibilities that innovations like machine learning bring.
Jeff Gould has 20 years of experience in technology publishing and IT market research. Jeff currently serves as the president of SafeGov Inc.