Geometric model for high-resolution SAR-GEC images

Estimation and correction of geometric distortion of stroke-based symbology for avionics display system using curve fitting

“…The least-square curve-fitting method is used to minimise the square of the error between the original data and the values predicted by the equation, whereas the non-linear curve fitting takes a user-defined equation to fit a set of plotted data. The least-square method is relatively simple and easy to implement, but not statistically robust method of fitting a function to a data-set, whereas the non-linear curve fitting is the most powerful curve fit to fit a set of plotted data by using user-defined equations (Farebrother 1999, Ching-Liang 2000, Yun-He et al 2003, Toutin 2004, Bi et al 2006, Zhang et al 2013.…”

“…If the image acquired by the camera is distorted or data coordinates are not matched with the template coordinates, the correction is necessarily needed. The correction function for the geometric distortion is computed using the polynomial curve-fitting expressions (Weiss et al 1989, Launay et al 1995, Ching-Liang 2000, Toutin 2004, Bi et al 2006, Zhang et al 2013 a process to generate a mathematical function to find the 'best fit' to a series of data points. It is also known as regression analysis and used to produce a mathematical equation that can be used to find points anywhere along the curve or to smooth the data and improve the appearance of the plot.…”

“…The correction function for the geometric distortion is computed using the polynomial curve-fitting expressions, and finally, the output section presents the correction function for the distorted image to the DCS. The proposed methodology is a swift and accurate approach for image processing (Blacknell et al 1989, Launay et al 1995, Roman et al 1998, Toutin 2004, Zhang et al 2013.…”

“…To know the correction functions of individual test image, curve-fitting toolbox of MATLAB is used. The complete DCS system is developed using MATLAB7 (Blacknell et al 1989, Launay et al 1995, Roman et al 1998, Zhang et al 2013.…”

“…Also, the correction of geometric distortion associated with the synthetic aperture radar images and remote sensing data is generally performed using the mathematical algorithm based on the polynomial coefficients. These algorithms have been tested to be robust, computationally efficient and easy to realise, but it is widely applied to the offline digital images and post-processing of the remote sensing data (Toutin 2004, 2011, Li and Zhu 2010, Zhang et al 2013, Mumtaza et al 2014.…”