Live Facial Recognition Technology Terms and Definitions
A human assessment of an alert generated by the Live Facial Recognition (LFR) application by an LFR Engagement Officer (supported, as needed by the LFR Operator) to decide whether to engage further with the individual matched to a watchlist image. In undertaking the Adjudication process, regard is to be paid to subject, system and environmental factors.
A specially trained person who has access rights to the LFR application in order to optimise and maintain its operational capability.
An alert is generated by the Live Facial Recognition application when a facial image from the video stream is being compared against the watchlist and returns a comparison (similarity) score above the threshold.
A true alert is determined when the probe image is the same as the candidate Image in the watchlist.
Confirmed True Alert
Following engagement, a confirmed true alert is determined when the engaged individual is the same as the person in the candidate image in the watchlist.
True Recognition Rate
It is the total number of times an individual(s) on a watchlist known to have passed through the Zone of Recognition, correctly generating an alert, as a proportion of the total number of times those individuals pass through the Zone of Recognition. It is irrelevant whether an alert is generated by the LFR application or not.
This is also referred to as the True Positive Identification Rate.
When it is determined by the operator that the probe image is not the same as the candidate image in the watchlist, based on adjudication without any engagement.
(The false alert rate is one of the two measures relevant to determining application accuracy).
Confirmed False Alert
Following engagement, confirmed false alert is determined when the engaged individual is not the same as the person in the candidate image in the watchlist.
False Alert Rate
The number of individuals that are not on the watchlist who generate a false alert or confirmed false alert, as a proportion of the total number of people who pass through the zone of recognition. This is also referred to as false positive identification rate.
Application accuracy can be considered to consist of the combined LFR technology accuracy and the human in the loop decision-making process. Accuracy is determined by measuring two metrics, the True Recognition Rate and the False Alert Rate. This is further explained below. The example given has been simplified to demonstrate the concept, but note that the metrics have been calculated in accordance with the agreed scientific method as set out by the International Organisation for Standardisation:
Authorising Officer (AO)
An authorising officer (usually holds the rank of Superintendent or above) provides the authority for LFR to be used.
A digital representation of the features of the face that have been extracted from the facial image.
It is these templates (and not the images themselves) that are used for searching and which constitute biometric personal data. Note that templates are proprietary to each facial recognition algorithm and new templates will need to be generated from the original images if the LFR application’s algorithm is changed.
A watchlist comprises known persons that can be used to test system performance, for example, police officers / staff may be placed on a blue watchlist and `seeded’ into the crowd who walk through the zone of recognition during a deployment.
Image of a person from the watchlist returned as a result of an alert.
Use of an LFR application as authorised by an authorising officer to locate those on an LFR watchlist.
An amalgam of the LFR application, the written authority document and the LFR cancellation report. This sets out the details of a proposed deployment including – but not limited to:
b. dates and times
c. deployment and watchlist rationale
d. legal basis
h. watchlist composition
i. authorising officer
k. relevant statistics
m. summary of any issues
An officer communicating with a member of the public as a result of an alert.
They are external element that affect LFR application performance such as dim lighting, glare, rain, mist etc.
Faces per frame
A configurable setting that determines the number of faces that can be analysed by the LFR application in each video frame.
Facial Recognition Technology (FRT)
This technology works by analysing key facial features, generating a mathematical representation of these features, and then comparing them against the mathematical representation of known faces in a database and generates possible matches. This is based on digital images (either still or from live camera feeds).
Where a person on the watchlist passes through the zone of recognition but no alert is generated. There are a number of reasons false negatives occur; these include application, subject and environmental factors, and how high the threshold is set.
Is the officer who assumes overall command and has ultimate responsibility and accountability for the Deployment. (They are responsible and accountable for the policing operation/event and determine the strategic objectives).
Live Facial Recognition (LFR)
LFR is a real-time deployment of facial recognition technology, which compares a live camera feed(s) of faces against a predetermined watchlist in order to locate persons of interest by generating an alert when a possible match is found.
LFR Engagement Officer
An officer whose role is to undertake the adjudication process following an alert, which may or may not result in that officer undertaking an engagement. These officers will also assist the public by answering questions and helping them to understand the purpose and nature of the LFR deployment.
An officer or staff member whose primary role is operating the LFR system. They will consider alerts and, via the adjudication process, will assist LFR engagement officers in deciding whether an alert should be actioned.
LFR System Engineer
A person who SWP deems to have suitable technical qualifications and experience to optimise and maintain the operational capability of SWP LFR system.
Person(s) of Interest
This term comprises persons on a watchlist
A facial image which is searched against a watchlist.
Retrospective Facial Recognition (RFR)
A post-event use of facial recognition technology, which compares still images of faces of unknown subjects against a reference image database in order to identify them.
The officer who commands and coordinates the overall tactical implementation of the LFR Deployment in compliance with the strategy set by the Gold Commander. (The silver commander develops, commands and coordinates the overall tactical response of an operation, in accordance with the strategic objectives set by the gold commander).
Is a numerical value indicating the extent of similarity between the probe and candidate image, with a higher score indicating greater points of similarity.
A factor linked to the individual, for example, demographic factors or physical features or behaviours for example, the individual is wearing a head covering, is smoking, eating, or looking down at the time of passing the camera.
A factor relating to the LFR application such as the algorithm.
The configurable point at which two images being compared will result in an alert. The threshold needs to be set with care to maximise the probability of returning true alerts whilst keeping the false alert rate to an acceptable level.
In the context of authorising an LFR deployment, a deployment that is related to an: Imminent threat-to-life or serious harm situation; and/or intelligence / investigative opportunities with limited time to act, where the seriousness and potential benefits support the urgency of action.
A set of known reference images against which a probe image is searched. The watchlist is normally a subset of a much larger collection of images (from the reference image database) and will have been created specifically for the LFR deployment.
Zone of Recognition
Three-dimensional space within the field of view of the camera and in which the imaging conditions for robust face recognition are met. In general, the zone of recognition is smaller than the field of view of the camera, e.g. not all faces in the field of view may be in focus and not every face in the field of view is imaged with the necessary resolution for face recognition.