Applications of Evolutionary Computation in Image Processing and Pattern Recognition by Erik Cuevas Daniel Zaldívar & Marco Perez-Cisneros

Author:Erik Cuevas, Daniel Zaldívar & Marco Perez-Cisneros
Language: eng
Format: epub
Publisher: Springer International Publishing, Cham

6.6.3 Homography Estimation with Synthetic Data

This section reports on the experimental results corresponding to the estimation of homography matrix considering synthetic data. In the experiment, for the first view, 48 inliers have been generated through a rectangular pattern of 8 × 6 elements within a 2-dimensional space of [−300, 300]. Such points are transformed by a random homography H and contaminated by normally distributed noise for constructing their correspondences in the second view. A set of outliers was added by selecting data points randomly within the space limits. In the test, the fraction of outliers varies from 0 to 100 %.

In the experiment, each algorithm’s execution requires 1000 iterations. In order to illustrate the characteristics of each execution, Fig. 6.11 shows a test where the CSA-RANSAC has been applied to estimate a random H considering only the 75 % of additional outliers. Figure 6.11 presents the performance results for each algorithm. Such results represent the averaged outcome after 50 different executions.

Fig. 6.11A test example where the CSA-RANSAC has been applied to estimate a random transformation H considering only the 75 % of additional outliers. a the first view and b the second view, with black squares representing the detected inliers

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.