MISSION
AFR Engine was founded in 2018 by members of the law enforcement community with decades of experience. With our first hand knowledge of the challenges that law enforcement face on a daily basis, our mission was to create an investigations software powered by facial recognition to help combat crime and keep communities safe.
HOW IT WORKS
AFR Engine serves as a centralized repository of booking photographs for known offenders, consisting of individuals who have been arrested and had their mugshots taken during the booking process. We manage a nationwide database of booking photos that is continuously updated on a daily basis. Importantly, AFR Engine does not collect images from social media or the general public.
When a user uploads an image, AFR Engine searches the database for potential candidates whose facial features closely resemble those in the uploaded image. The search results are then presented for manual human review, where the user compares each candidate's image side-by-side with the uploaded image. During this review, the user evaluates both similarities and dissimilarities to determine whether any of the candidates could be relevant to their investigation. If further investigation is warranted, AFR Engine can provide basic booking information associated with the photo, such as name, date of birth, age, or booking number.
It is essential to note that AFR Engine’s facial recognition technology does not establish positive identifications. Rather, AFR Engine generates investigative leads by identifying individuals within its database who share similar facial characteristics. Users are reminded that arrests shall not be made solely based on facial recognition results; probable cause must be established in accordance with both state and federal law.
TESTING AND VALIDATION
AFR Engine actively participates in the National Institute of Standards and Technology (NIST) Face Recognition Technology Evaluation (FRTE), a globally recognized benchmark for assessing the performance of facial recognition systems. NIST, a division of the U.S. Department of Commerce, was established by Congress in 1901 with the mission of providing rigorous, unbiased evaluations of industrial and technological performance capabilities.
In the NIST 1:N Face Recognition Technology Evaluation, published on September 8th, 2024, AFR Engine’s algorithm achieved an accuracy rate of 99.79% when identifying the correct individual from a database of 12 million images—surpassing the accuracy of human recognition.