d-anon: Automated Trespass Management

Rylan R.
rylan@turing.lol
www.d-anon.org

Abstract

d-anon is an automated trespass management platform for businesses. Using facial recognition, d-anon identifies every patron upon entry, syncs trespass records across all business locations in real time, and alerts your personnel the moment a trespassed individual attempts to re-enter. Upon an incident occurring d-anon makes it simple to bar the patron and provide the applicable evidence to report to law enforcement.

1. Introduction

Businesses suffer when items are locked up at retail. Spending drops as community centers become overrun with civil disorder. The only solution is strict, swift, and just enforcement from the private sector. Businesses need the ability to see problematic individuals coming their way and bar them or keep them under surveillance in order to keep services high for all other patrons.

The days of neighborhood stores knowing every regular are long gone in most parts of America. Expecting high-turnover staff to remember troublesome patrons is a tall order. That being said, disorder and theft are still rampant problems that need to be solved. Systems like d-anon already exist, but they are in the hands of government bodies or are industry-specific , such as Vegas's notorious blackbook . There needs to be an open platform that any business can connect to.

2. SCOR (Shared Criminal Offense Registry) and Indexes

In addition to the internal list of trespassed patrons. Any partner organization can submit to SCOR at any time. Submissions require video evidence, which is AI-verified for accuracy before syncing to all subscribed organizations. Business owners can subscribe to shared indexes from the d-anon database, covering categories like "Theft," "Misconduct," and "Violent Offenses".  These are sourced from SCOR entries and public court records. Activated indexes automatically flag matching individuals upon entry and trigger configurable response protocols (e.g., silent staff alerts, automatic law enforcement contact [1]).

Detections fall into three categories:

• Green: No match, cleared to enter.
• Yellow: History of incidents at other locations.
• Red: Currently banned / trespassing.

3. Detection

Cameras are installed at every entry and exit point of the establishment. This will give a clear picture of all individuals currently on the premises. When a person of interest is detected, the previously configured response protocols will activate. From there the businesses choose how to proceed. Whether that means a formal trespass notice issued by security staff or notifying local law enforcement of the individual's presence. [1] 

4. Post-Incident

After an incident has occurred on premises, select the person of interest from a timestamped list of entrants and enter all known details about the incident. Then describe and label the incident, e.g. disorderly conduct or theft. After this, a copy will be sent to local law enforcement and be added to the SCOR database.

5. Implementation

d-anon will go to market with a camera called the sentry-1. This is the camera that will be installed in new installations. Partners can choose to use their existing camera infrastructure if it meets the qualifications for deployment.

6. Privacy

The private data of entrants to an establishment is not compromised. Entrants are not added to the system, and there are no logs kept of them; their facial embed will only be used to make a match against existing persons in the SCOR database. The exception being information about an entrant will be saved upon their involvement in an incident. In applicable states, the correct precautions will be taken, such as optional opt out or affirmative consent.

7. Conclusion

Any establishment that opens itself up to the public would improve operations with this proposal. The proposed system allows all visible forms of misconduct to be tracked, and a historical view of a person's actions to be drawn. If there is an understanding within the general public that offenses will be documented and taken seriously, it will, in turn, lead to better and more orderly conduct being displayed.

[1] In the FTC’s Rite Aid enforcement action , the FTC alleged that Rite Aid’s use of facial recognition for retail security was unfair because the company failed to implement reasonable safeguards against false-positive matches and resulting consumer harm. In the FTC complaint , the FTC alleged that match alerts led to actions such as surveillance, removal from stores, public accusations, searches, blocked purchases, and police calls. The FTC order applies to Rite Aid specifically and does not categorically bar other companies from using facial-recognition security systems, but it signals that such systems should include safeguards such as accuracy testing, false-positive monitoring, staff training, complaint handling, retention/deletion controls, clear notice, and human review before adverse action.