Inertial Sensors and Their Applications - Signal Processing Systems - page 72

Cameras use inertial sensors for image stabilization to reduced blurring associ-
ated with the motion of a camera during exposure. It compensates for angular yaw
and pitch movement of the camera. There are two ways for images stabilization
in cameras: (1) make adjustments to the image sensor or the lenses to ensure
that the image remains as motionless as possible, (2) digital image stabilization in
which the physical image is allowed to track the scene on the sensor by software
to produce a stable image. The digital technique requires the pixel count to be
increased to allow the image to move on the sensor while keeping reference points
within the boundaries of the capture chip. Different companies have different
names for their image stabilization technology: Image Stabilizer (Canon), Vibration
Reduction (Nikon), Optical SteadyShot (Sony Cyber-Shot), Super SteadyShot
(Sony), MEGA Optical Image Stabilizer (Panasonic and Leica), Optical Stabilizer
(Sigma), Vibration Compensation (Tamron) and Shake Reduction (Pentax).
Fall detection is an important safety feature to protect hard disk drives in laptops
and some other portable, “always on” devices like MP3 players. Many of these
devices feature an accelerometer which is used to detect drops. If a drop is detected,
the heads of the hard disk are parked to avoid data loss and possible head or disk
damage caused by the shock.

1.3.5 Sport

Pedometer is an example of application that can encourage a healthy lifestyle. A
pedometer in its simplest form is a step counter and can be implemented using
low-cost accelerometers. Other examples of inertial sensors in sport include motion
analysis such as golf swing, figure skating jumps, and trajectory analysis in ski
jumping and javelin. Xsens MVN Motion Capture [ 24 , 33 ] is an interesting example
of how inertial sensors can be used to record human movement. The motion capture
suit includes 17 inertial trackers strapped to the different parts of the body. The data
can be used in medical and sports applications to analyze human movement and
gait. It can be also used to animate digital characters in movies, games, and virtual
environments.

2 Performance of Inertial Sensors

Selection of the most suitable inertial sensors for a particular application is a
difficult task. Among the parameters that have to be considered are resolution,
dynamic range, accuracy, cost, power consumption, reliability, weight, volume,
thermal stability, and immunity to external disturbances. Usually when sensors are
examined for compliance, accuracy is the first parameter to start with.
All accelerometers and gyros are subject to errors which limit their accuracy
in the measurement of the applied acceleration or angular rate. The measurement
error is defined as the difference between the measured and the true value of the