Camera View Verification for Katy, TX 77492 Businesses

Start from the incident, not from the camera

Most commercial systems were designed around the building — corners, entrances, whatever the conduit could reach. A system that works is designed backwards, starting from the specific incidents a business of your type actually experiences, because each one demands a different distance, a different angle and a different amount of detail.

Run that exercise against your own camera list and the gaps tend to announce themselves. The most common finding across Katy businesses is not too few cameras — it is six cameras all answering the same low-value question and none answering the expensive one.

The export test: can you hand an officer a file that plays?

This is the part of the verification that changes the most minds, because it is the part a business has never done. We ask a member of your staff — not us — to export a five-minute clip from a specified camera at a specified time. What we are testing is not the camera at all.

• Can anyone on site do it? On a striking number of businesses, the only person who knows the recorder password is the installer, who may no longer be in business. At six in the morning with an officer waiting, that is a total system failure regardless of picture quality.
• Does the exported file play anywhere else? Many recorders default to a proprietary container that needs the manufacturer’s player. Detectives and adjusters want something ordinary that opens on any computer. If your system cannot produce that, it is worth knowing now rather than then.
• Does the export carry its own timestamp? A clip with no visible time and no accompanying metadata invites an argument about when it was recorded.
• Is it the full-quality stream? If the recorder was set to archive the low-resolution sub stream, the export is a small, soft file and no setting applied afterwards will recover the detail.
• Is the incident still on the drive? Internal theft is typically discovered at inventory, weeks later. A system that holds a fortnight cannot answer a question about last month.

We finish this step by writing a one-page export procedure and leaving it with the recorder, so the person who needs it at 6 a.m. is reading instructions rather than guessing.

Why plate capture is a separate camera, not a setting

“Can we read plates?” is the most common request we get from Katy business owners, and the honest answer is that a plate is a different photographic problem from a person. It has four constraints, and a general-purpose overview camera violates all of them at once.

• Angle. Characters compress as the viewing angle opens up. Keep the camera within roughly thirty degrees of the plate’s axis horizontally and about the same vertically. A camera on a building corner looking across a lot is often at sixty degrees or worse, where the characters simply are not resolvable.
• Character size. A US plate is a foot wide, and reading it reliably wants somewhere around a hundred-plus pixels across that foot. That means a tight lens on a narrow slice of lane, not a wide view of the whole entrance.
• Shutter speed. A moving vehicle at night needs a very fast shutter to freeze the characters — far faster than a scene camera would ever use. That setting makes the rest of the frame nearly black, which is exactly why a dedicated plate camera looks broken to anyone who does not know what they are looking at.
• Speed of approach. Vehicles rolling to a gate are an easy problem. Vehicles crossing a driveway at speed are a hard one, and the answer is usually to move the capture point to where traffic has to slow.

High bays, dock doors and fenced yards along the corridor

A large share of Katy-area commercial space is flex, light-industrial and contractor yard along and off the I-10 corridor, and those buildings break the assumptions a retail camera layout is built on.

Clear height works against you

In a bay with twenty-four to thirty feet of clear height, the obvious place to mount is the structure — and from up there every person is a hard hat and a pair of shoulders. Usable detail in a warehouse comes from cameras brought down to column-mounted positions at eight or nine feet, aimed along aisles rather than down at them. The high mount still has a role for overall movement; it just cannot identify anybody.

Racking is not permanent

A camera plan drawn against a racking layout has a shelf life, because racking gets reconfigured. Any system in a warehouse should be re-verified after a layout change, which is one of the few times we will suggest a scheduled re-check rather than a one-off.

Dock doors are two scenes

An open dock door at midday is one of the highest-contrast scenes in commercial video: full Texas sun in the opening, deep shade in the building. Closed, the same frame is dark. One camera must handle both, so genuine multi-exposure wide dynamic range is not a luxury here, and the camera should be positioned so the opening sits to one side of the frame rather than filling the centre.

Distance and dust

Yards run past what copper can carry — an Ethernet channel tops out around 328 feet, and a fence line or a back gate on an acre-plus site passes that quickly, so the honest answer is a mid-point switch in an enclosure, a powered extender, or fibre to a remote cabinet. Meanwhile forklift traffic and yard dust film a dome housing in weeks rather than years, which is why a cleaning interval belongs in the report alongside everything else.

Counter, entry and back-of-house on a retail floor

Retail is the opposite problem: short distances, plenty of light, and cameras that still fail to produce a usable face. Three positions do most of the work.

• Over the counter, facing out. Mounted behind the register looking at the customer. The common error is the reverse — a camera behind the customer that records your own employee’s back through every transaction and nothing else.
• Inside the entry, facing the door. People look up as they leave, and a camera at head height just inside the exit collects clean, front-on faces at close range. It is consistently the single most productive camera in a shop, and it is the one most layouts do not have.
• Back-of-house and the stockroom door. Shrink concentrates where goods leave the sales floor. A camera on that doorway is worth more than two more on the floor.

Two things to watch on a sales floor. Bright menu boards and display screens in frame cause the auto-exposure to hunt, darkening faces every time the display changes; and seasonal resets move gondolas and end caps, which redraws aisle sightlines two or three times a year. Both are cheap to correct once somebody looks.

Retention: the arithmetic behind “thirty days”

Retention is not a feature you enable. It is division, and it is worth doing on the back of an envelope before you rely on a number somebody quoted you.

Take sixteen cameras recording continuously at four megabits per second each. That is sixty-four megabits per second, which is eight megabytes per second, which is about 691 gigabytes every twenty-four hours. On a four-terabyte drive that is under six days — not thirty.

The levers that actually move it, in order of how much they buy you:

• Recording mode. Motion or event-based recording can multiply retention several times over on a site that is quiet at night. It also risks missing the seconds before an event, so the pre-record buffer has to be set deliberately.
• Codec and bitrate. Modern compression roughly halves the storage for equivalent quality, but only if the cameras are configured to use it rather than left on a legacy profile.
• Per-camera tuning. A plate camera on a narrow lane and an overview camera on a sales floor do not need the same bitrate. Blanket settings waste storage on cameras that do not need it and starve the ones that do.
• Drive capacity. The blunt lever, and the last one to reach for after the free ones.

Two cautions we include in every commercial report. Surveillance recording is a continuous write workload, so the drives should be rated for it, and they wear out — we read the drive’s own health data during the visit. And a mirrored pair protects against a drive dying, not against somebody deleting footage or walking off with the recorder, so if the recorder is reachable from the sales floor, that is a finding in itself.

How EVOTECH runs a commercial verification

1. Before the visit we ask what has actually happened to you — past claims, losses, disputes — because that list determines what the cameras need to prove.
2. A walk with your manager, camera by camera, recording model, lens, height, channel and cable path into a document that in most cases does not exist yet.
3. Measurement and a staged subject in every frame at the distances that matter, watched on the recorded stream rather than the live wall.
4. Condition testing — dock doors open and closed, lot cameras against headlights, the sales floor with the displays running, then a pass after dark.
5. The recorder — stream configuration, frame rate, motion and privacy zones, clock accuracy, drive health, retention arithmetic, network path, switch PoE budget and UPS runtime.
6. The export test, performed by your staff, followed by a written export procedure left on site.
7. A coverage map and per-camera report, then an itemised quote for whatever you choose to correct, staged so it can be budgeted.

What drives the cost, and the five failures we find most

Commercial pricing is set on site and itemised — never over the phone, because the variables are physical. What moves it: total camera count and number of recorders; ceiling height and whether a lift is required; how many buildings or yard positions are involved; whether cabling has to be traced because documentation was never produced; whether the work has to happen outside trading hours; and whether staff training on export and playback is included.

The five findings that repeat

1. The archive is the sub stream. Live video at full resolution, recorded video at a fraction of it.
2. Nobody on site can export. No password, no procedure, no practice.
3. The clock is wrong. Sometimes minutes, occasionally hours, always an argument later.
4. Retention is shorter than the discovery window. The system overwrites the evidence before the business learns it needs it.
5. The cameras were aimed at the building. Handsome coverage of walls, doors and parking rows; nothing framed on the place an incident actually occurs.

None of the five is expensive to correct. All five are expensive to discover late.

Related services

Frequently asked questions