UV theft detection powder is a mark without a clock. That is why it loses in court.

UV theft detection powder is a non-toxic fluorescent powder (or paste, when mixed with petroleum jelly) that is invisible to the naked eye and glows brightly under a 365 to 395 nm UV flashlight. Applied to items at risk of theft (mailroom shelves, tool boxes, copper spools, cash drawers, fire alarm pulls, keyboards), it transfers to the hands and clothing of anyone who touches the treated surface and typically stays detectable on skin for hours to several days depending on hand-washing behaviour. Commercial products are sold by GLO Effex, ASR Federal, ForensiKit by Crime Scene, TriTech Forensics, Shomer-Tec, and SpySite among others. The powder itself is cheap, fast to apply, and the technology is old and well understood. The unsolved part is not the chemistry, it is the evidentiary chain.

Two reasons, both well documented by defense attorneys. First, secondary transfer: the suspect's hands may have picked up the powder from a surface other than the target (a door handle someone else touched after a legitimate user, for example), so glowing hands prove contact with powder, not contact with the stolen item at the time of the theft. Second, similar fluorescence: many ordinary substances fluoresce under UV, including laundry detergent residue, toothpaste, tonic water, petroleum jelly, olive oil, and some bodily fluids. A defense lawyer only has to establish reasonable doubt. To rebut both challenges, a prosecutor needs an independent, timestamped record that places the accused in contact with the treated item at the specific moment the theft occurred. That record is what UV powder alone cannot produce.

A zone entry event is a verified record that a specific person was in a drawn polygon around the treated item, for a measured number of seconds, at a specific ISO-8601 timestamp. When Cyrano's edge unit detects this, it emits a payload containing the tile thumbnail, the zone name, the dwell seconds, the timestamp to the second, and delivers it to the property's WhatsApp thread in under 60 seconds. That record, stored and timestamped independently of the UV mark, is what corroborates powder transfer as occurring at the theft and not at some unrelated earlier moment. The evidentiary weight of UV powder goes up significantly when the prosecutor can pair the glow on the suspect's hands with a thumbnail of that same suspect standing over the treated target at a specific second.

No. Cyrano's capture point is the DVR or NVR's HDMI multiview output, the same signal that drives the guard monitor. That composite already has every camera on the recorder mosaiced into tiles. One HDMI tap gives inference access to every feed at once. Works on analog cameras, IP cameras, any DVR brand including Hikvision, Dahua, Lorex, Amcrest, Reolink, Uniview, Swann, Night Owl, Q-See, and their rebrands. No camera firmware involvement, no ONVIF negotiation, no credentials to coordinate. Physical install on a running DVR is under 2 minutes: HDMI in, HDMI out to the monitor, network, power. Up to 25 tiles per unit.

Useful: mailrooms and package lockers (small number of high-traffic treated items, recurring thefts, recurring actors), construction sites and laydown yards (copper spools, tool boxes, power tool chargers), cash drawers and till trays (substitution and skim cases), transformer pads and HVAC condenser cages (copper line sets). Overkill: parking lots, open perimeters, retail sales floors with high legitimate touch volume. The common pattern where powder earns its place is a defined treated surface that legitimate users rarely touch, paired with a specific list of suspects (employees, residents, contractors) where cross-checking a UV glow against a short list narrows the investigation quickly.

They work better together. A zone entry event alone is strong evidence of presence and dwell, which is enough for live intervention (talkdown, dispatch, arrest on scene) and for civil trespass. UV powder alone is weak evidence of contact at an unspecified time. Combined, they create a two-source corroboration: the person was at the treated surface for N seconds at 21:07:42 (camera), and that person has powder transfer consistent with having touched the treated surface (UV inspection). Each source closes the evidentiary gap the other leaves open. Most properties we see using powder were using it without camera correlation, and the cases that actually closed were ones where they had corroborating footage by luck, not by design.

Every zone entry event includes event.class (pre_action_zone_entry), property identifier, tile.label (camera name), zone name, dwell_seconds (integer), timestamp (ISO 8601 with timezone offset, to the second), a tile thumbnail URL (480x270 crop centered on the event), overlay_mask identifier (which DVR overlays were masked before inference), delivery channel (WhatsApp group or SMS fallback), and latency_ms (capture to delivery). The payload is deliberately minimal and deliberately machine-readable. If the property escalates to a police report, that payload is the record that goes on the evidence log next to the UV inspection result.

Three rules. First, keep application logs: date, time, applicator initials, exact surfaces treated, batch number of powder, and re-application cadence. Second, treat only surfaces where legitimate contact is narrow and documented (mailroom inner shelf, specific tool cribs, specific copper spool rack), not general handles and high-touch surfaces. Third, draw the camera zone around the treated surface with a dwell threshold tuned to how long someone actually has to stand there to steal. For a mailroom shelf, 10 to 20 seconds of dwell is enough to pair meaningfully with a UV contact. For a copper spool rack on a jobsite, 30 seconds or longer filters out workers walking past. Application logs plus event logs are the two documents a prosecutor will ask for.