De Villiers, JPCronje, JNicolls, FC2012-01-162012-01-162011-04De Villiers, JP, Cronje, J and Nicolls, FC. 2011. Improved neural network modeling of inverse lens distortion. Defense, Security, and Sensing (DSS11), Orlando World Center Marriott Resort & Convention Centre, Orlando, Florida, USA, 26-28 April 2011, 9 pphttp://hdl.handle.net/10204/5488Defense, Security, and Sensing (DSS11), Orlando World Center Marriott Resort & Convention Centre, Orlando, Florida, USA, 26-28 April 2011Inverse lens distortion modelling allows one to find the pixel in a distorted image which corresponds to a known point in object space, such as may be produced by a RADAR. This paper extends recent work using neural networks as a compromise between processing complexity, memory usage and accuracy. The already encouraging results are further enhanced by considering different neuron activation functions, architectures, scaling methodologies and training techniques. The errors are given in terms of microns on the detector to facilitate fair comparison between different resolutions and pixel sizes.enNeural networksLens distortionInverse distortion correctionsRadarConference PresentationDe Villiers, J., Cronje, J., & Nicolls, F. (2011). Improved neural network modeling of inverse lens distortion. http://hdl.handle.net/10204/5488De Villiers, JP, J Cronje, and FC Nicolls. "Improved neural network modeling of inverse lens distortion." (2011): http://hdl.handle.net/10204/5488De Villiers J, Cronje J, Nicolls F, Improved neural network modeling of inverse lens distortion; 2011. http://hdl.handle.net/10204/5488 .TY - Conference Presentation AU - De Villiers, JP AU - Cronje, J AU - Nicolls, FC AB - Inverse lens distortion modelling allows one to find the pixel in a distorted image which corresponds to a known point in object space, such as may be produced by a RADAR. This paper extends recent work using neural networks as a compromise between processing complexity, memory usage and accuracy. The already encouraging results are further enhanced by considering different neuron activation functions, architectures, scaling methodologies and training techniques. The errors are given in terms of microns on the detector to facilitate fair comparison between different resolutions and pixel sizes. DA - 2011-04 DB - ResearchSpace DP - CSIR KW - Neural networks KW - Lens distortion KW - Inverse distortion corrections KW - Radar LK - https://researchspace.csir.co.za PY - 2011 T1 - Improved neural network modeling of inverse lens distortion TI - Improved neural network modeling of inverse lens distortion UR - http://hdl.handle.net/10204/5488 ER -