https://ico.org.uk/about-the-ico/news-and-events/information-commissioner-s-opinion-addresses-privacy-concerns-on-the-use-of-live-facial-recognition-technology-in-public-places/

It is important that we arrive at a societal consensus through informed public debate, and this blog strives to do just that.

In her blog, Elizabeth states : “In the US, people did not trust the technology.”

Unfortunately and und ultimately, it’s not the technology people mistrust, but the authorities using it.
(Our views on this here.)

There are a myriad of uses of this powerful tech. Some abhorrent, some of great value. It’s no different from any technology. It can be applied for good and for bad.

We should be careful not to throw the baby out with the bath water. It is not sufficient to ask “should the technology be used in place xyz?”, but also:

1. By whom?
2. For what?
3. Under what control and governance?

However, page 65 of the referenced report specifies that Precision (defined as “… the percentage of positively-identified cases that are in fact positive.”) should be an important metric. Here, I somewhat disagree. If nobody in the watchlist walks past a camera, does that mean the precision of the system is 0?
(our views on this here.)

It is commonly misrepresented that if facial recognition has been deployed in a location, everybody in that location is being “tracked”. Not so. The people in the watchlist are being tracked. So, who’s in the watchlist? and Why?
(our views on this here.)

#privacy #trust #facialrecognition #cctv #ethics #gdpr #ICO

The post Comment on the UK Information Commissioner’s Opinion on Facial Recognition appeared first on Allevate.

Allevate qualifies as a supplier to the UK Crown Commercial Service’s Digital Outcomes and Specialists 5 (DOS5) framework, from 20th January 2021.

LONDON, UK 20th January 2021:  Allevate today announces it has qualified as a supplier to the UK Crown Commercial Service’s Digital Outcomes and Specialists 5 (DOS5) framework.

This framework helps the public sector buy, design, build and provide bespoke digital outcomes by finding appropriate specialists to provide services and outcomes.

DOS 5 replaces DOS 4 and is available as of January 20, 2021.

About Allevate Limited

Founded in London in 2007, Allevate works with law enforcement, intelligence and government agencies to enhance public safety by ensuring positive identification through the application of biometric and identification technology.

• Ensure Positive Identification
• Enhance Public Safety
• Reduce Operational Costs

Visit us at https://allevate.com , email us at contact@allevate.com, call us on +44 20 3239 6399 and follow us at @Allevate.

About the UK Crown Commercial Service and DOS

The UK Crown Commercial Service (CCS) works with both departments and organisations across the whole of the public sector to ensure maximum value is extracted from every commercial relationship and improve the quality of service delivery. The CCS goal is to become the “go-to” place for expert commercial and procurement services.

The DOS 5 framework is designed to help the public sector buy, design, build and deliver bespoke digital solutions and services. 94% of the 3,340 suppliers on the new agreement are SMEs.

The public sector has spent more than £2.5bn through Crown Commercial Service’s DOS agreements since 2016, with 34% of that spend – more than £800m – going directly to SMEs. 

.

The post Allevate Qualifies as a UK Crown Commercial Service DOS5 Supplier. appeared first on Allevate.

However, there are vocal groups focussed on privacy that are well intentioned in their vociferous opposition to this technology. They perform a vital role; the industry has learned from them greatly and continues to make significant adjustments and improvements because of their opposition.

But no matter how far industry goes to address these concerns, it may never be far enough for them. Some revel in the inaccuracies, biases and lack of standards, as it provides a focal point of opposition. And the more accurate, standardised and unbiased the technology becomes, the more it will be opposed, not less.

So why do some people mistrust facial recognition? My epiphany, which surprised me, was that it is not the technology that is mistrusted.

It is the police, governments and authorities that people mistrust; the users of the technology, not the technology itself.

Society has become increasingly polarised. Conspiracy theorists abound. Many simply don’t want this technology in the hands of the authorities, especially if its flaws are resolved.

Many of these concerns are well founded. There is a sufficient track record of misuse of tools and abuse of authority to warrant the imposition of and adherence to publicly agreed best practices and the holding of our authorities to account.

Whilst industry can and should make strides in addressing these issues and improving the technology, industry itself cannot resolve this. It is a societal conundrum that requires a societal debate, but I fear it is unlikely this mistrust will be assuaged anytime soon in a society that embraces division and polarisation.

In the interim, facial recognition has the potential to significantly enhance public safety and improve efficiency. It would be a shame to throw this opportunity away.

Almost any technology can be used for bad as well as good; facial recognition is not alone in this respect.

If we really don’t trust our authorities, what other tools should we take away from them?

Facial Recognition Accuracy: A Worked Example

The post Why do Some People Mistrust Facial Recognition? appeared first on Allevate.

Allevate announces the release of R3.0 of its Face-Searcher EDGE face detection capability, coupled with Tygart’s MXSERVER, and their availability on the UK CCS G-Cloud12 framework. Cloud Hosting via partners up to OFFICIAL SENSITIVE or TIER2. Provides powerful facial recognition enabled forensic analysis of media integrated with live CCTV surveillance watchlist alerting. Available to all UK government, intelligence and law-enforcement agencies with a flexible monthly commercial model.

LONDON, UK 25 September 2020:  Allevate today announces the release of R3.0 of it’s Face-Searcher EDGE face-detection utility and the continued availability of Tygart’s MXSERVER  on the UK Crown Commercial Service’s Digital Marketplace G-Cloud 12 Framework.

Allevate’s G-Cloud service can be optionally powered by Sungard Availability Services, who provide a Secure Managed Cloud using AWS (Amazon Web Services) up to OFFICIAL SENSITIVE.

For higher levels of assurance,  UKCloud’s UKCloudX division, which exclusively serves the community of Defence, Intelligence, Security and Law Enforcement agencies in the UK, can provision TIER2 hosting to power MXSERVER.

Forensic Analysis of Media Integrated with Live Surveillance Watchlist Alerting

This unique offering now enables any UK Public Sector sector organisation to utilise MXSERVER  and Allevate’s unique Face-Searcher architecture to benefit from “ubiquitous identification”.

• Forensically analyse masses of digital media.
• Maintain and manage biometrically enabled watchlists.
• Implement live facial recognition surveillance services on any number of geographically dispersed cameras from a single central infrastructure.

All using a centralised cloud-infrastructure coupled with Edge Detection over low-bandwidth networks.

Pricing uses an easy to calculate monthly service fee, based on either:

• Monthly Payment Based on Server Footprint
  Pay a monthly software license fee based on the size of the server footprint in the cloud or your datacentre. Suitable for forensic analysis of bulk media.

• Monthly Payment Per Live Surveillance Camera
  Pay a monthly software license fee per camera, based on the number of surveillance cameras submitting live searches. Includes unlimited server-side software to meet your response and workload requirements. Suitable for a deployment supporting live facial recognition alerting for any number of cameras.

About Allevate Limited

Founded in London in 2007, Allevate works with law enforcement, intelligence and government agencies to enhance public safety by ensuring positive identification through the application of biometric and identification technology.

• Ensure Positive Identification
• Enhance Public Safety
• Reduce Operational Costs

Visit us at https://allevate.com , email us at contact@allevate.com, call us on +44 20 3239 6399 and follow us at @Allevate.

About MXSERVER

MXSERVER, from UK Crown Commercial Service’s Digital Marketplace G-Cloud 12 Framework, processes vast amounts of textual, video and photo collections quickly – automatically discovering, grouping and extracting segments depicting people. Using face recognition technology, this solution searches media archives to find other assets which depict individuals of interest. It also indexes the digital media to enable it to be efficiently searched using a photograph of a face, previewed and analysed via an intuitive web-based user interface. Results become available in minutes rather than hours or days because the digital media files are processed in parallel over a distributed cloud-architecture.

MXSERVER automates the bulk-processing of media for forensic analysis and is already proven by US Federal agencies to provide an “Order of Magnitude” efficiency gain and significantly enhanced identification of suspects.

About the UK Crown Commercial Service and G-Cloud

The UK Crown Commercial Service (CCS) works with both departments and organisations across the whole of the public sector to ensure maximum value is extracted from every commercial relationship and improve the quality of service delivery. The CCS goal is to become the “go-to” place for expert commercial and procurement services.

The G-Cloud catalogue is open to all public sector clients and is designed to provide a simple streamlined process for buying ICT focused products and services that use cloud computing as a commodity.

The post “Ubiquitous Identification” using Facial Recognition Available on G-Cloud 12 on UK Sovereign OFFICIAL SENSITIVE or TIER 2 Platform appeared first on Allevate.

We were being lined up for a large, national and significant overseas government deployment of our Face-Searcher Facial Recognition Solution.

Though there were no legal obstacles or export restrictions to the government in question, our investigations made clear there were significant human rights considerations and evidence of civil rights abuses that were too great to ignore. This revenue would have been a major boon, but we realised we could not have undertaken this project and felt good about it.

Facial Recognition technology has the ability to significantly improve Public Safety, but we are all too aware it is also subject to abuse. We are determined to do our upmost to stay on the right side of the line and ensure our technology be a force for good and societal positive improvement, not oppression and suppression of rights.

#ethics #cctv #surveillance #facialrecognition #privacy #police #bordersecurity #nationalsecurity #doingtherightthing #humanrights #civilrights #publicsafety

The post Ethical Use of Facial Recognition: Doing the Right Thing appeared first on Allevate.

There is always room for significant improvement. Also, there needs to be open, comprehensive and inclusive debate on the appropriate use of the technology.

However, often those who claim the technology is wildly inaccurate, citing figures of 50-90% inaccuracy, do so on the basis of incorrectly assessing the statistics.

An Example

There are 1,000 people in front of you. 6 dangerous criminals loiter amongst them:

• You deploy an automated system to ASSIST.
• The system instantly inspects all 1,000 people and selects 10 for you to manually assess.
• 5 of the selected 10 people are from the 6 criminals. The other 5 are not criminals.
• YOU (NOT the system) assess these 10 people to make a determination.
• There are 994/1,000 non-criminals.  995 are deemed non-criminals. 1 slipped through the net.
• The system’s false reject rate was 1 / 1,000 (the missed criminal) = 0.1% *
• The system’s false accept rate was 5 / 1,000 (the non-criminals it selected for you to assess) = 0.5% *
• The accuracy of the system was NOT 50% (5 genuine matches out of 10 total matches.)
• You only needed to assess 10 people, not 1,000.
• You caught 5 of 6 criminals, which you otherwise would not have. 

If the system was 81% inaccurate, as reported, it would have selected over 800 people from the 1,000 for you to manually assess.

Consider if there did not happen to be any criminals in the 1,000. That does not mean the system is 100% inaccurate because there were no positive matches.

So, with respect to facial recognition accuracy, is this an accurate or an inaccurate system?

* The FAR and FRR would actually be lower as there would be multiple people in the watchlist. The total number of comparisons is 1,000 time the number of people in the watchlist.

Same Example, Different Words

Let’s use exactly the same example, but let’s talk about marbles instead of people.

There are 1,000 marbles in a bucket. Amongst them are 6 red marbles. The other 994 are blue. You must find as many red marbles as quickly as you can.

• You deploy an automated system to ASSIST.
• The system instantly inspects all 1,000 marbles and selects 10 for you to manually assess.
• 5 of the selected 10 marbles are red. The other 5 are blue.
• YOU (NOT the system) assess these 10 marbles to make a determination.
• In the bucket, 994/1,000 of the marbles were blue marbles.  995 are deemed to be blue. One red marble was missed.
• The system’s false reject rate was 1 / 1,000 (the missed red marble) = 0.1%.
• The system’s false accept rate was 5 / 1,000 (the blue marbles it selected for you to assess) = 0.5%.
• The accuracy of the system was NOT 50% (5 genuine matches out of 10 total matches.)
• You only needed to assess 10 marbles, not 1,000.
• You found 5 of 6 red marbles, which you otherwise would have taken you much longer.
• The 5 blue marbles caught in your sweep you place back in the bucket.
• There is no way of later distinguishing these blue marbles from the rest of the blue marbles. They are not tracked in any way. 

If the system was 81% inaccurate, it would have selected over 800 marbles from the 1,000 for you to manually assess, not saving you any time at all.

Consider if there did not happen to be any red balls in the bucket. That does not mean the system is 100% inaccurate because no red balls were selected.

Now imagine you have many, many of these buckets of 1,000 marbles, in which you need to find as many red marbles as possible in as little time as possible. Would you use this tool? Or would you inspect them all manually?

The main difference between these two examples is that, in the scenario where you are looking for red marbles, all red marbles are considered equally. With people, you need to actually differentiate between the different criminals. i.e. you have multiple shades of red, one for each distinct criminal you are looking for.

You can read further for a fuller analysis of facial recognition accuracy, within the context of live operational use by the South Wales Police Force in the United Kingdom.

The post Facial Recognition Accuracy: A Worked Example appeared first on Allevate.

You can find it here:

We applaud these initiatives being undertaken by NPCC and the Surveillance Camera Commissioner’s Office and are confident that the pillars of Education, Standardisation and Best Practice will revolutionise innovative and pro-active policing in the UK.

Below are a couple of comments on her blog post.

1. Society holistically has a responsibility to find the balance between privacy and security and adhere to it. I believe, however, that aspects of the narrative have been hijacked by certain groups who spin misinformation with a view of achieving their predetermined / desired outcome.  Equally as important as “Ensuring Best Practice” is “Educating the Public” and dissemination of informative fact. All the best practice, standardisation and returns on technology such as Facial Recognition will be meaningless if it ends up being banned because we haven’t pro-actively informed the public.

2. Much of the public discourse tends to focus on the “live” aspects of surveillance. However, significant benefit can be realised from the bulk forensic analysis of surveillance footage.  You touch upon a significant obstacle to achieving this with your reference to managing “… CCTV footage to ensure its standardised …”. CCTV footage as commonly sourced during investigations today exists in a myriad of proprietary formats.  Analytic technology providers are challenged with developing capabilities to ingest this landscape of ever-changing formats as this is a highly specialised domain of expertise in and of itself. If Police implemented an “enterprise video normalization capability” that accepted “any” video file format and outputted a normalized video adhering to an open video standard, each video analytic vendor could then focus its resources on solving other problems, rather than all of them attempting to solve the same global foundational problem.

The post Comment on UK Surveillance Camera Commissioner’s Office Latest Blog Post appeared first on Allevate.

MXSERVER Available on G-Cloud. Optionally utilise a UK Sovereign OFFICIAL SENSITIVE Platform. Facial Recognition Enabled Forensic Analysis of Media Integrated with Live CCTV Surveillance Watchlist Alerting. Available to all UK government, intelligence and law-enforcement agencies on the G-Cloud 11 framework.

LONDON, UK 2 July 2019:  Allevate today announces the continued availability of MXSERVER on the UK Crown Commercial Service’s Digital Marketplace G-Cloud 11 Framework.

Allevate’s G-Cloud SaaS service is optionally enabled by Sungard Availability Services.  Sungard AS provide a Secure Managed Cloud using AWS (Amazon Web Services). This fully UK Sovereign platform is AWS approved, Public Service Network (PSN)-connected and managed / supported by Security-cleared, UK-based personnel.

You may host your MXSERVER deployment on Sungard’s managed PaaS platform or on a platform of your choice.

You can view Allevate’s MXSERVER G-Cloud service descriptions here.

Forensic Analysis of Media Integrated with Live Surveillance Watchlist Alerting

With MXSERVER available on G-Cloud, any UK Public Sector sector organisation can utilise MXSERVER and Allevate’s unique Face-Searcher architecture to:

• Forensically analyse masses of digital media.
• Maintain and manage biometrically enabled watchlists.
• Implement live facial recognition surveillance services from any number of cameras.

… all on a fully managed UK Sovereign AWS Cloud-hosted platform for OFFICIAL and OFFICIAL SENSITIVE workloads.

Access to both the software and all hosting and storage services are available on Digital Services Marketplace G-Cloud  framework. Pricing uses an easy to calculate monthly service fee.

About Allevate Limited

Founded in London in 2007, Allevate works with law enforcement, intelligence and government agencies to enhance public safety by ensuring positive identification through the application of biometric and identification technology.

• Ensure Positive Identification
• Enhance Public Safety
• Reduce Operational Costs

Visit us at https://allevate.com , email us at contact@allevate.com, call us on +44 20 3239 6399 and follow us at @Allevate.

About MXSERVER

MXSERVER, from Tygart Technology, processes vast amounts of textual, video and photo collections quickly – automatically discovering, grouping and extracting segments depicting people. Using face recognition technology, this solution searches media archives to find other assets which depict individuals of interest. It also indexes the digital media to enable it to be efficiently searched using a photograph of a face, previewed and analysed via an intuitive web-based user interface. Results become available in minutes rather than hours or days because the digital media files are processed in parallel over a distributed cloud-architecture.

MXSERVER automates the bulk-processing of media for forensic analysis and is already proven by US Federal agencies to provide an “Order of Magnitude” efficiency gain and significantly enhanced identification of suspects.

About the UK Crown Commercial Service and G-Cloud

The UK Crown Commercial Service (CCS) works with both departments and organisations across the whole of the public sector to ensure maximum value is extracted from every commercial relationship and improve the quality of service delivery. The CCS goal is to become the “go-to” place for expert commercial and procurement services.

The G-Cloud catalogue is open to all public sector clients and is designed to provide a simple streamlined process for buying ICT focused products and services that use cloud computing as a commodity.

The post MXSERVER Available on G-Cloud : UK Sovereign OFFICIAL SENSITIVE Platform appeared first on Allevate.

Following 10 trials by the London Metropolitan Police, the London Policing Ethics Panel has established a framework for Police use of facial recognition technology.

Five Conditions for the Use of Facial Recognition

Following an extensive review of the trials, the Panel yesterday published a comprehensive report which recommends that live facial recognition software should only be deployed by police if the following five conditions can be met:

1. Its use offers more than marginal benefit to the public, sufficient to compensate for the potential distrust it may invoke.
2. It does not generate unacceptable gender and racial bias into policing operations.
3. Deployments are independently assessed to ensure it is necessary and proportionate for the purpose.
4. Human operators are trained to understand the potential risk of injustice by inappropriate responses to alerts and are accountable for their actions.
5. The police and the Mayor’s Office for Policing and Crime jointly develop strict guidelines to balance the benefits of of the technology with its potential intrusiveness.

The report is balanced and suggests the panel is in favour of both:

• the use of facial recognition.
• the development of appropriate controls, guidelines and oversight governing such use.

The panel also suggest that police should continue the investigation of the use of facial recognition but say “...We argue therefore that MPS should proceed with caution…“ The panel states “…we have come to the view that there are important ethical issues to be addressed but these do not amount to reasons not to use LFR at all” but balance this with “...deployment of LFR must observe principles of necessity and proportionality. Neither of these principles could be satisfied by unrestricted use of LFR.“

Public Largely in Favour of Proportionate Use of Facial Recognition Technology

The report also summarises the findings of a survey that they undertook. 1,092 Londoners responded to the survey between 23 May and 4 June 2018. They find from the survey that “A central objection is that surveillance has the potential to produce a chilling effect on democratic debate and protest, and more generally dissuade people from engaging in legitimate activities in public space ” but “…the counter-argument that surveillance can make public spaces safer, including for vulnerable groups“. Specifically, the survey highlights:

• “Overall 57% of respondents thought that in general terms, police use of LFR was acceptable”
• “..in only two cases did support fall below 50%. So, …. (Public) support was almost overwhelming.“
• “…degrees of acceptance depended on the specific purposes and setting in which the technology might be used. “
• “…there is a clear rank order, with support for use to identify terrorists and those wanted for serious violence significantly higher than support for use to identify those wanted for minor crime and nuisance behaviour.”
• “…the idea of using LFR to apprehend criminals or people who are wanted by the police was well received, and the vast majority deemed it to be appropriate as long as ‘people of interest’ is clearly defined.”
• “Participants were hopeful the technology would allow police to make identifications more quickly and accurately”
• “Over one-third of those surveyed had concerns about LFR relating to privacy, both in relation to themselves and others, along with the ethics of police collecting data from people who have not committed a crime.”
• “Most respondents stated use of LFR would not discourage them from going to public events … because they … had ‘nothing to hide’…” and “… could appreciate the potential safety benefits … such as a reduction in the threat of terrorist attacks.”

You can read the full report at :

http://www.policingethicspanel.london/uploads/4/4/0/7/44076193/lfr_final_report_-_may_2019.pdf

The post Framework for Police Use of Facial Recognition Established by Policing Ethics Panel appeared first on Allevate.

The below is in response to a query on facial recognition in arenas and stadiums.

The Query

Looking into the use of facial recognition and its uses in stadiums; the standard required requires the recognition of faces from pictures in crowds with differing light levels.

The Response

The major factors that impact accuracy of facial recognition are:

• Sufficient resolution across the face (optical not digital zoom)

• Sufficient illumination across the face

• Uniform illumination across the face

• Minimal angle on the face

… on both the reference and search images.

The challenge you describe (crowds, distant cameras, lighting, angles) sounds like one of cameras producing decent imagery rather than an FR problem. Complex environments will require expensive camera systems. Simple environments can use much simpler (cheaper) cameras.

You will always get the best result when and where you can control the sensor and the sensor’s environment (lighting, glare, minimal distance from and angle on subject etc).

This is why unidirectional (everyone facing the same) way pinch points (such as entrances, turnstiles, hallways) will always yield the best results. Your challenge is not insurmountable, but it will possibly be prone to False Rejects (missed matches) and Fales Accepts (mismatches) unless thought and investment is made into the camera systems.

Be wary of anybody that tells you that their FR solution can “easily” deliver highly accurate solutions in this environment.

Also of interest, may be:

• Facial Recognition False Positives: What they Really Mean.

• Face-Searcher: Facial Recognition as a Cloud Service

The post Facial Recognition in Stadiums and Arenas appeared first on Allevate.