Mosaicking with MOPEX

We present MOPEX a software package for image mosaicking and point source extraction. MOPEX has been developed for the Spitzer Space Telescope. This paper concentrates on the mosaicking aspects of the package. MOPEX features the use of several interpolation techniques, coaddition schemes, and robust and flexible outlier detection based on spatial and temporal filtering. A number of original algorithms have been designed and implemented in MOPEX. Among them is direct plane-to-plane coordinate transformation, which allows at least an order of magnitude speed up in performing coordinate transformation by bypassing the sky coordinates. The dual outlier detection makes possible outlier detection in the areas of even minimal redundancy. Image segmentation based on adaptive thresholding is used for object detection, which is part of outlier detection. Efficient use of computer memory allows mosaicking of data sets of very deep coverage of thousands of images per pointing, as well as areas of sky covering many square degrees. Although designed for Spitzer data, MOPEX does not require any Spitzer-specific fits header keywords to run, and can be applied to other data, that have standard header information on the image geometry and pointing. The package is available for distribution at http://ssc.spitzer.caltech.edu/postbcd/.