Using aerial imagery, the JPL FineCalibration (JPL FineCal) software automatically tunes a set of existing CAHVOR camera models for an array of cameras. The software finds matching features in the overlap region between images from adjacent cameras, and uses these features to refine the camera models. It is not necessary to take special imagery of a known target and no surveying is required.

JPL FineCal was developed for use with an aerial, persistent surveillance platform. Synchronized images from an array of cameras are captured and stitched together into a single, very high-resolution image that is projected onto an elevation map of the ground. A GUI (graphical user interface) tool allows the user to play a movie of any part of the imaged surface from any perspective.

JPL FineCal requires, as input, a set of CAHVOR camera models for the camera array. These models are typically developed on the ground using a calibration procedure requiring a known target at a short distance. JPL FineCal corrects the inaccuracy of the camera model extrinsic parameters resulting from the short target distance by using imagery, taken during flight, at an effective distance of infinity. It also makes small improvements to the intrinsic parameters.

JPL FineCal is an automated process that does not require the use of any special targets, and which may be applied during normal flight operations. Thus, it makes it simple to retune the camera models to correct for small misalignments that occur due to changes in aperture settings, vibration, or thermal changes.

This work was done by Daniel Clouse, Curtis Padgett, Adnan Ansar, and Yang Cheng of Caltech for NASA's Jet Propulsion Laboratory.

The software used in this innovation is available for commercial licensing. Please contact Karina Edmonds of the California Institute of Technology at (626) 395-2322. Refer to NPO-45715.



This Brief includes a Technical Support Package (TSP).
Document cover
Automated Camera Array Fine Calibration

(reference NPO-45715) is currently available for download from the TSP library.

Don't have an account? Sign up here.