To detect wildfire ignition as soon as possible using visual AI detection, image quality is critical. To capture high-quality images, we recommend video surveillance cameras with rectilinear lenses, IP66/IP68 enclosures with wipers, and a resolution of 1920×1080.
Rectilinear Lens
A rectilinear lens is a type of camera lens that produces images with minimal distortion, keeping straight lines straight. In visual AI detection, a rectilinear lens offers significant advantages over a fisheye lens. By maintaining accurate and undistorted representations of objects and scenes, the rectilinear lens ensures precise measurements, size estimations, and spatial relationships crucial for object recognition and tracking. This reliability enhances the overall performance and accuracy of visual AI detection systems.
Why Rectilinear lens is expensive and heavy? What to know more? Below is a “Fisheye vs Rectilinear lenses” video by Dave McKeegan.
Video 1: Fisheye vs Rectilinear Lenses by Dave McKeegan
IP66/IP68 Enclosure with Wiper
Cameras deployed in forest environments are subject to harsh weather and working conditions. We recommend camera systems with IP66 and IP68 rating enclosures. They are reliable systems with minimal downtime and require less maintenance and support costs.
The IP (Ingress Protection) rating is an international standard that determines the degree of protection or sealing efficacy in enclosures against the infiltration of objects, water, dust, and contact. It is also compliant with European Standard EN 60529.
Products are noted by the initials IP and followed by two digits. The two digits denote the level of intrusion protection versus solids and liquids. Devices with an IP68 rating offer protection against immersion in liquid up to 1.5 meters deep, while the IP66 rating provides shielding from powerful water jets.
Dirt, raindrops, and moisture on the camera windows refract light, reduce image quality, and cause false positive AI detections. A camera housing with a wiper removes raindrops and dust from the windows, keeping the window surfaces clear of moisture.
(image source: AlertWildfire)
One More Thing: Detection Range
(image source: Omni Calculator)
Earth curvature refers to the natural curvature of the Earth’s surface due to its spherical shape. As one moves away from a given point on the Earth’s surface, the ground gradually drops away, causing distant objects to become hidden from view. This phenomenon limits visual distance, making objects beyond a certain distance appear obscured by the curvature of the Earth. The extent of this limitation depends on the height of the observer’s viewpoint and the height of the observed object. This curvature plays a crucial role in various applications, such as long-distance photography, surveying, and telecommunications, as it affects the visibility and accuracy of observations over vast distances.
The Distance to the Horizon is the furthest distance that an observer can see the ground level. Below is its formula:
(image source: Omni Calculator)
Some of our customers install their cameras on telecom towers. The cameras are about 75 meters in height from the ground level. The Distance to the Horizon of the camera is about 30km (18.6 miles). Based on our experience, an early-stage ground fire at 30km is too small to detect. The optimal detection distance is 15km, and within this range, our AI wildfire detection can usually find the fire within the first 15 minutes after ignition.
Our favourite camera? AXIS Communications Q6225-LE. Its image quality is crystal clear in both daytime and night-time. It is a heavy-duty PTZ camera that moves fast and works very reliably. It looks cool!
Please contact us at info@roboticscats.com for any questions.
