Motion-blurred image restoration using modified inverse all-pole filters

摘要

The restoration of vibration-blurred images using measured motion function is considered. Since the blurring filter is of the finite impulse response type, the inverse one is of the all-pole infinite impulse response type. Direct application of the inverse filter to restore images blurred by vibration is attractive because of reduced computation requirements. Space domain filtering allows high par-allelization of the process and reduction of required processor speed. However, a pure inverse filter provides excessive noise amplification and is possibly unstable. The proposed technique is to construct a modified inverse filter with preserved all-pole structure and optimized noise and stability properties. A mathematical concept was developed using z-transform properties. An experiment testing the proposed technique was setup and the results indicate restoration is 60%-80% of that possible with ideal Wiener filtering. However, the reduced computation and high parallelization can facilitate real-time restoration.