It’s one hilarious fragment in the youth program The Test Kitchen. Presenter Willem wants to film while running, but that gives such a jerky image. That’s why he attaches his mini camera… to the head of a chicken, which he holds. A chicken naturally uses image stabilization: no matter how its body moves – up/down, left/right, and even when rotating, the head remains upright and in the same place. Biologists call that the stabilization reflex. This is caused by a complex interplay between eyes, brain, nerves and neck muscles. Modern optical equipment also provides image stabilization. But how?
Image stabilization builds on the principle of ‘gimbal suspension‘, which dates back to antiquity. In this case, an object is suspended in such a way that it can rotate freely around three axes that are all perpendicular to each other, often in the middle of three interlocking rings. This way, the object remains upright, even if the environment to which it is attached – for example a ship, a drone or a camera – moves. Ships in the sixteenth century already had a gimbal compass.
Modern equipment does not necessarily use that principle for mounting the lens or the light sensor itself. Nowadays there are some steps in between. It is the sensor that measures the movements of the device that uses this principle. That sensor is called a gyroscope – basically an ancient invention. A gyroscope measures rotations. A classic gyroscope contains a rapidly rotating element that wants to maintain the orientation of its axis of rotation. That is a law of nature: the ‘conservation of angular momentum‘. You also see it, for example, in a spinning top, which remains upright as long as it spins quickly, but falls over as soon as it comes to a stop. Modern electronic variants use vibrating microstructures for this.
Gyroscope as a balance organ
You can measure deviations in the direction of rotation or vibration very precisely. Devices then use that information to counteract and thus cancel movements. In other words: the gyroscope is, as it were, the ‘balancing organ’ of image stabilization. The gyroscope ‘detects’ that a device is rotating or vibrating slightly. Then another part intervenes to compensate for that movement.
This compensation is done mechanically, optically, electronically or software – or with a combination of these. With a camera it works usually like that: a gyroscope measures a small movement of your hand > a processor calculates how the image would shift > a lens group or the image sensor immediately moves in exactly the other direction > the image can therefore fall silently on the sensor > you get a sharp photo.
This happens with stabilized binoculars and telescopes something similarbut it is not the image sensor that is corrected – it is not in the device, but in your eye – but the light path itself. This is done with movable prisms. Microprocessors continuously adjust these prisms to compensate for the vibrations. Some systems even use tiny electric motors and gimbals for this purpose.
Smartphones contain miniature gyroscopes, based on MEMS: microelectromechanical systems. These are tiny mechanical structures on a chip that can measure how the device rotates or moves via micrometer-scale vibrations. In addition, smartphones also use digital correction through software. Video cameras use stabilization in which algorithms shift small portions of the image to smooth out shakes. Finally, drones not only stabilize the image inside the camera, but… also the entire camera itself with a motorized gimbal suspension.
