Kalman filter
IPA: / ˈkælmən ˈfɪltər /
Plural: Kalman filters
An adaptive filter used to estimate the state of a system from measurements that contain random errors. This recursive adaptive filter determines the correct parameters of a process model. Each new measurement allows the parameters of a model to be predicted and adjusted, thus providing an estimate of error at each update. The Kalman filters computational structure and its ability to incorporate the effects of noise (from both measurement and modeling) recommends itself for use in computer vision tracking applications. See also active vision, distortion, modeling, noise.
Words nearby
kallac · kallidin · kallikrein · kallikreinogen · Kallmann's syndrome · Kalman filter · Kalmar · kalmia · Kalmuck · Kalmuk · Kalmyk