Camera - how to mount and frame it

🎯 Camera Mounting Height

• Minimum installation height: 4 meters

• Recommended height:

  • 4–5 m for standard cross-sections and smaller intersections

  • 6–7 m for large intersections and roundabouts with multiple lanes

📌 Why is this important?

• Camera mounted too low → further lanes obscured by vehicles closer to the lens.

• Such occlusions are difficult for AI to resolve correctly.

• The higher mount reduces obstruction problems and increases the field of view.

Quick Reference Table

Scenario	Height	Pitch Angle	Notes
Cross-section	4 m	15°–25°	Camera parallel to the road
Road intersection	5–7 m	30°–45°	Camera right next to the intersection
Roundabout	5–7 m	30°–45°	Camera right next to the roundabout

---

📐 Pitch Angle and Field of View (FOV)

Road Cross-section

• Tilt (Pitch): 15°–25°

• Horizontal FOV: 60°–80°

• Alignment: Parallel to the road axis – optimal for License Plate Recognition (LPR) and capturing vehicle side profiles.

Intersections

• Tilt (Pitch): 30°–45° (can be increased to 45°–60° for large interchanges)

• Horizontal FOV: 90°–120°

• Note: The closer to vertical (Pitch → 90°, i.e., "bird's-eye view"), the better the visibility of traffic relations and the fewer occlusions.

Roundabouts

• Tilt (Pitch): 60°–90° (optimal: vertical, Pitch ≈ 90°)

• Horizontal FOV: 100°–130°

• Alignment: Central mounting above the roundabout (if possible) or multiple cameras at entries – each with its own specific angle.

---

Optimal Camera Framing for Traffic Flow Counting and ANPR

An example of a correct frame for a road intersection.

Example of a correct frame for a road cross-section.

---

🛣️ Mounting Location

One-way road / Cross-section

• Pole-mounted by the road

• Camera parallel to the road axis

• Digital cropping available in AITracker if the camera is positioned away from the road

Intersections

• Mounted on a pole directly at the intersection

• Large intersections: 1 camera per entry

Roundabouts

• Mounted on a pole directly at the roundabout

---

Correct Camera Frame Examples

Intersection or Roundabout Counting

Example of a correct frame for an intersection.

Sectional Speed Measurement

Correctly defined speed measurement section.

ALPR for Parking Time Measurement

A camera mounted inside a vehicle traveling on the road and counting traffic.

---

🌙 Night Measurement and Weather Conditions

• Use cameras with IR (Infrared), especially in unlit areas.

• Rain, fog, or dirt on the lens → although the AI is trained on such footage, severe weather conditions can still reduce analysis effectiveness.

---

⚠️ Common Mounting Errors

Error	Consequence
Mounting too low	Occlusions caused by vehicles in the foreground
Incomplete coverage of the intersection/roundabout	Failure to count entries that are outside the frame
Trees in the middle of a roundabout	Interrupted tracking trajectories
"Face-on" camera alignment instead of an angled view	Causes additional occlusions
Frame too narrow or insufficient resolution	Causes AI errors or failure to recognize objects

---

💡 How Does Night Lighting Work with the Cameras?

Scenario	AI Recognition
✅ A visible vehicle silhouette	Ensures AI recognition
✅ A partially visible vehicle silhouette	Sufficient for AI
❌ A vehicle as a ball of light	Will NOT be recognized
❌ Vehicles with only headlights visible	Insufficient for AI
⚠️ High speed and significant blur	Borderline recognizable

---

🧩 Example Problems and Solutions

Problem: "The camera cannot see the second lane; it is obstructed by a truck."

✅ Solution: Increase the mounting height or add a second camera.

Problem: "ALPR fails to recognize plates at dusk."

✅ Solution: Reduce the camera's shutter speed or increase the FPS (frames per second). Both solutions improve image sharpness.

Problem: "ALPR fails to recognize plates at night."

✅ Solution: The camera must be equipped with IR illumination, and the light's angle of incidence should be 45 degrees.