Target Recognition and Classification

The distance that you will be able to detect and recognize what a hot target is depends on many different camera design and atmospheric factors. Most manufacturers use a mathematical model to establish their range performance claims. As the leaders in maritime imaging, we provide actual test results and validate the range performance in real life conditions.

There are three industry-standard benchmark measurements normally used to present the range performance of a thermal imaging system: Detection, Recognition, and Identification.

Detecting a target simply means you can see a hot spot on the monitor, even if it’s just one pixel. Recognizing a target means you can tell if it’s a boat, as opposed to an island, and this could be as few as five or six pixels. For maritime market, instead of using the term Identification, FLIR uses “Classification of Human Activity,” since you can’t really determine the identity of a human using thermal cameras.

Classifying human activity is a more meaningful benchmark, given that in most first responder missions, officers are primarily interested- in confirming that people are onboard, determining how many there are, and understanding what the people are up to.

Detection Image

Detection means you can see something on the monitor. It might only represent one or two pixels on the camera, but it shows up just the same.

Recognition Image

Recognition means you can tell it’s a boat, as opposed to a small island or a marker. At FLIR, we use a 30’ outboard boat as the benchmark.

Human Activity Image

Instead of recognition, we use “Classification of human activity” as the model. In most maritime first responder and commercial applications, we want to see if people are moving around on a boat and gain some sense of what they are up to.