This is the code used to produce the results and visualizations published in
Schmittwilken, L. & Maertens, M. (2022). Fixational eye movements enable robust edge detection. Journal of Vision, 22(5). doi:10.1167/jov.22.8.5
The repository contains the following:
-
The data from the psychophysical experiment of Betz et al. (2015) and the Contour Image Database by Grigorescu et al. (2003) that is used as test cases of the model: databases
-
Two Jupyter-Notebooks with a step-by-step guide through the proposed active edge detection model active_edge-model.ipynb and a demonstration of how spatial edge models work spatial_edge-models.ipynb
-
Code to create the results shown in the paper: simulations. To reproduce the results of test case 1 (edge detection in narrowband noise), run main_case1.py. To reproduce the results of test case 2 (contour detection in natural images), run main_case2.py
-
Code to create the visualizations from the manuscript: visualize_results. In order to re-create the visualizations, first run the simulations to produce the respective results.
Code written by Lynn Schmittwilken (l.schmittwilken@tu-berlin.de)
Betz, T., Shapley, R., Wichmann, F. A., & Maertens, M. (2015a). Noise masking of White's illusion exposes the weakness of current spatial filtering models of lightness perception. Journal of Vision, 15(14), 1, doi:10.1167/15.14.1
Grigorescu, C., Petkov, N., & Westenberg, M. A. (2003). Contour detection based on nonclassical receptive field inhibition. IEEE Transactions on Image Processing, 12(7), 729–739, doi:10.1109/TIP.2003.814250