A Step-by-Step Exact Measuring Angle Calibration Applicable for Multi-Detector Stitched Aerial Camera

Abstract

Nowadays in surveying and mapping applications, the requirement of aerial mapping camera with large image frame is increasing rapidly, for its advantages of great imaging efficiency. Because of difficulty in fabrication of single large detector, stitching of multiple detectors is an approach to constructing large format digital aerial cameras. Also, it is shown in practice that imagery created from multiple images acquired by multi-detector camera system can have similar metric quality, so this sort of camera system will stay in production. As a result, output from multi-center projected images acquired by multiple detectors to an equivalent single center projected virtual image is a core issue, where an accurate calibration is necessary. Exact measuring angle method is a common camera calibration method in the laboratory, and it is a very accurate method. It also performs well when dealing with calibration for long-focal length camera indoors, without limitation of imaging distance. However, it is not applicable to cameras of which the central area is not imaging. This chapter proposes a step-by-step calibration method. Step 1 is to calibrate the central area with traditional exact angle measurement. And step 2 is to establish the image point transformation according to the relative position of various detectors and hence accomplish the calibration of edge area using the result of central area. Through theoretical derivation and simulation, it proves that the proposed method is suitable for multi-detector stitching aerial camera, and it overcomes the shortcomings of impossible measurement for edge area. Also, measurements error caused by optical axis deviation and focal length inconsistency is considered, and its influence on focal length and principle point calculation is analyzed. By deriving the error propagation model of the proposed exact measuring angle method, the assembly accuracy of some key points in the camera system is estimated, which will be a key indicator of accuracy control during the process of camera system production.