Camera basics
How cameras trigger: PIR vs image detection
Most wildlife cameras fire on a passive infrared (PIR) sensor: moving heat. A few use image-based motion detection instead. The choice shapes what gets captured, how much power it costs, and how many empty frames you collect.
Before any AI is involved, something has to decide when to take a picture. Usually that is a trigger — a sensor firing on activity — and cameras use one of two trigger methods, which behave quite differently in the field. A camera can also skip triggering altogether and shoot on a fixed schedule; we cover that below too.
PIR: moving heat
A passive infrared sensor watches for a moving warm body against a cooler background. When a mammal crosses the beam, the temperature difference trips the shutter. Its big advantage is power: the sensor sits in a low-draw standby and only wakes the camera on a heat event, which is how a PIR camera can run for months on a set of batteries. Its limits: it can miss cold-blooded animals, struggles on hot days when an animal barely stands out from a warm background, reacts to heat shimmer and sun-warmed branches (false triggers), and only covers a limited range and angle.
Image (pixel) detection
An image-based camera samples the scene and fires when enough pixels change between frames. It catches anything that moves — including cold-blooded animals and subjects farther out than a PIR beam would reach — and it is the first step toward doing analysis on the camera itself. The cost is power: the sensor has to stay awake to keep comparing frames, which drains batteries far faster, and changing light or wind produces more false triggers to sort through.
Time-lapse: capture on a schedule
Not every camera waits to be triggered. In time-lapse mode the camera takes a frame at a fixed interval — every minute, every few minutes — whether or not anything moved. Because it never relies on heat or motion, it never misses a slow or cold subject, which makes it the natural choice for things that change gradually: a nesting site, a carcass, a water level, plant phenology. The trade-off is volume: time-lapse produces far more frames, most of them empty, so there is much more to filter afterward. In practice, motion and time-lapse are complementary — each catches animals the other misses — which is why many cameras support a hybrid setup, for example time-lapse through the day and PIR at night.
At the data level this is the real dividing line. Monitoring standards such as Camtrap DP record a camera's capture method as either activity detection (PIR or image-based) or time-lapse — the two categories that describe how a record came to exist.
What it means for your monitoring
Most trail cameras are PIR because battery life usually wins in the field. That is a fine default — as long as you know its blind spots: slow, cold, or distant subjects, and hot afternoons. If those matter for your target species, an image-based, time-lapse, or hybrid setup helps, at a power cost you plan for. Either way, place the camera to reduce empty triggers at the source; see preventing false triggers.
Whatever trips the shutter, the cloud AI sorts it. However your camera triggers, uploads run through European-hosted wildlife AI: empty and false triggers drop out, and real captures get a species label with a confidence score. See Wildlife AI.
