Overhead-mounted fisheye or panoramic cameras are often used here because they cover a wide area (120° or more). However, these lenses can cause image distortion, especially around the edges, which can affect how well people are detected. For better results, people should be in the center of the image, and the camera should be mounted directly above the queue, typically 3–5 meters high and angled straight down. Flattening the image through software can also help reduce distortion.

In many cases, standard dome cameras may work better in these areas, as they offer more consistent image quality across the frame and may give more reliable results for retail queue monitoring.

Whichever camera is used, it should meet DORI’s Detection or Observation levels (62.5–125 px/m) to ensure accurate tracking. Combining the camera with retail video analytics tools — like heatmaps and queue monitoring — adds value by improving operations as well as security.

3. Sales Floor: Visibility + Intelligence

4. Stockrooms and Storage Areas: Narrow, Deep, and Dim

Back-of-house areas such as storage rooms, product cages, or loading spaces often have narrow aisles, high shelving, and limited lighting — making it important to choose a low light surveillance camera suited for these conditions.

Key factors in such spaces include good low-light performance, adjustable lenses, and adequate resolution to cover long and narrow views. Cameras with larger sensors (e.g., 1/1.8" or 1/2.8") can capture more light, improving image clarity in dim conditions. Likewise, lenses with wide apertures (e.g., F1.4–F1.6) enhance low-light sensitivity and help maintain detail at night.

Vari-focal IR cameras — often in the 2–4MP range — are a strong fit here. Their adjustable focal lengths (e.g., 2.7–13.5 mm or even 5–50 mm) allow precise coverage of deep aisles or high shelves. Reliable infrared illumination is also essential to maintain visibility in unlit zones.

For narrow spaces, enabling corridor mode (also called 9:16 vertical view) can significantly improve coverage. It rotates the video frame to match the shape of the aisle, maximizing the visible area and reducing wasted pixels on walls or ceilings.

Integrating these cameras with systems like warehouse management (WMS) or access control (ACS) further improves operational oversight — helping detect stock issues or unauthorized access quickly.

In loading zones and delivery gates, cameras face one of the most complex visual tasks: capturing license plates. These areas involve fast-moving vehicles, changing light, headlights, and awkward angles — demanding high precision.

While resolution matters, the key to success is image clarity and full control over lighting. The most critical requirement is strong and consistent lighting at the entry point — with external lamps aimed at vehicles to ensure plates are clearly visible in all conditions. This lighting works alongside the camera's built-in IR to reduce shadows and glare from headlights.

Bullet cameras with narrow-angle varifocal lenses (e.g., 5–50 mm) are a strong choice. They are typically mounted 1.5–2.5 meters high at a 25–35° angle to the plate. These cameras should support True WDR (at least 120 dB) and have IR illumination with at least 30 meters of range to handle low-light and night-time captures.

In TRASSIR’s approach, license plate recognition (LPR) is done server-side via AutoTRASSIR. This allows the use of standard cameras, while still achieving high recognition accuracy — provided the lighting and positioning are properly set up.

• 250 px/m is required for reliable identification at checkout or for LPR
  
• 125 px/m is sufficient for observing queues or general movement
  
• Detection-level coverage works well for large areas like warehouses
  

• Are your cameras meeting DORI pixel density for each task?
  
• Are any zones over- or under-covered?
  
• Are your retail video analytics tools fully leveraged?