Calibraton of a digital single lens reflex (SLR) camera using artificial neural networks

Abstract

The use of non-metric digital cameras in close-range photogrammetric applications and machine vision has become a popular research agenda. Being an essential component of photogrammetric evaluation, camera calibration is a crucial stage for non-metric cameras. Therefore, accurate camera calibration and orientation procedures have become prerequisites for the extraction of precise and reliable 3D metric information from images. The lack of accurate inner orientation parameters can lead to unreliable results in the photogrammetric process. A camera can be well defined with its principal distance, principal point offset and lens distortion parameters. Different camera models have been formulated and used in close-range photogrammetry, but generally sensor orientation and calibration is performed with a perspective geometrical model by means of the bundle adjustment. In this study, a feed-forward network structure, learning the characteristics of the training data through the backpropagation learning algorithm, is employed to model the distortions measured for the Olympus E-510 SLR camera system that are later used in the geometric calibration process. It is intended to introduce an alternative process to be used at photogrammetric calibration stage. Experimental results for SLR camera with two focal length setting (14 and 42 mm) were estimated using standard calibration and neural network techniques. The modeling process with ANNs is described and the results are quantitatively analyzed. Results show the robustness of the ANN approach in this particular modeling problem and confirm its value as an alternative to conventional techniques. © 2018 Elsevier B.V., All rights reserved.