Automated Item of Interest Detection

Overview

Development stage

DeployedSource: c) Deployed – The use case is being actively authorized or utilized to support the functions or mission of an agency.

High-impact designation

Presumed high-impact, determined notSource: b) Presumed high-impact, but determined not high impact

Impact justification

The AI outputs do not serve as a principal basis for a decision or action concerning a specific individual or entity that has a legal, material, binding, or similarly significant effect on that individual’s or entity’s civil rights, civil liberties or privacy, equal opportunities, or access to or the ability to apply for critical government resources or services. The AI runs on video and images captured from lawfully deployed technologies used to support the U.S. Border Patrol mission between Ports of Entry. The AI provides alerts when it detects the presence of an IoI, such as persons, vehicles, animals in the image frame. With regard to persons, this computer vision application is trained to determine if the object in the image frame is a person with a certain of level of confidence and not another object that may be shaped similarly to a person. After the alert of a detection, a trained agent or user, reviews the image to identify and classify the activity taking place. The AI merely alerts to the presence of an item it was trained to detect.

Start date

2020-01-31 (day precision)Source: 2020-01-31T00:00:00

Withheld from public reporting

Not reported

Mission

Topic area

Law Enforcement

Operational functions

No function tag matched Derived by keyword rules; see methodology

Problem the AI is intended to solve

The software is designed to analyze photographs and video feeds captured by field imaging equipment for review by U.S. Border Patrol (USBP) agents and personnel. Using proprietary software, the system processes and annotates images to identify whether they contain human subjects, animals, or vehicles. The system is designed to incorporate future enhancements that expand its detection capabilities and to improve accuracy based on user feedback.

Expected benefits

The software analyzes images and video that are taken by operationally deployed equipment, which are then fed into CBP systems for review by USBP agents, Office of Field Operations (OFO) officers, and other CBP users. It provides quick identification of people either crossing into the U.S. at a time and place other than designated for entry, circumventing security at a port of entry, or those already inside the U.S. trying to elude capture, as well as the ability for human operators to quickly determine if subjects in an image are, in fact, human.

System outputs

The system creates a layer which is overlaid over the image to produce a box around items of interest it has determined to be likely human beings.

Technology

AI classification

Classical / predictive MLSource: Classical/Predictive Machine Learning: Models trained on data to make predictions or classifications based on identified patterns or relatio…

System name(s)

I4 Viewer

Custom-developed code

Yes

Public source code

Not reported

Sourcing

How it was built

Contract and in-houseSource: c) Developed with both contracting and in-house resources

Vendor (as reported)

Matroid

Vendors (standardized)

None on the standard list

A vendor is the supplier named by the agency. It does not identify the underlying model or AI technology.

Data and privacy

Involves PII

No

Privacy Impact Assessment

Not reported

Authorization to Operate

Yes

Demographic features

Not reported

Training and evaluation data

All of the image data fed to the models are owned by USBP.

Federal Data Catalog

Not reported

Governance

0 of 8 minimum-practice questions answered. A blank answer means the agency reported nothing; it does not mean the practice is absent.

Potential impacts and how they were identified

Not reported