In coded-aperture imaging, a technique useful in nuclear medicine, a coded image of dimensionality one less than that of the object is recorded. It is possible to obtain a crude reconstruction of the object with the use of a form of correlation decoding of the coded image, but to date these methods have been unsatisfactory. We present an iterative reconstruction algorithm that makes use of a priori knowledge of object border and object postivity. Reconstructions obtained in simulations are consistent with the constraints and coded image and are surprisingly accurate in light of the extensive information loss involved in the image-formation process.
© 1984 Optical Society of America
Equations on this page are rendered with MathJax. Learn more.