Perceptual grouping along well-established Gestalt laws provides one set of traditional methods that provide a tiny set of meaningful parameters to be adjusted for each application field. More complex and challenging tasks require a hierarchical setting, where the results aggregated by a first grouping process are later subject to further processing on a larger scale and with more abstract objects. This can be several steps deep. An example from the domain of forestry provides insight into the search for suitable parameter settings providing sufficient performance for the machine-vision module to be of practical use within a larger robotic control setting in this application domain. This sets a stark contrast in comparison to the state-of-the-art deep-learning neural nets, where many millions of obscure parameters must be adjusted properly before the performance suffices. It is the opinion of the author that the huge freedom for possible settings in such a high-dimensional inscrutable parameter space poses an unnecessary risk. Moreover, few-parameter learning is getting along with less training material. Whereas the state-of-the-art networks require millions of images with expert labels, a single image can already provide good insight into the nature of the parameter domain of the Gestalt laws, and a domain expert labeling just a handful of salient contours in said image yields already a proper goal function, so that a well working sweet spot in the parameter domain can be found in a few steps. As compared to the state-of-the-art neural nets, a reduction of six orders of magnitude in the number of parameters results. Almost parameter-free statistical test methods can reduce the number of parameters to be trained further by one order of magnitude, but they are less flexible and currently lack the advantages of hierarchical feature processing.
Citation: Eckart Michaelsen. Few-parameter learning for a hierarchical perceptual grouping system[J]. Mathematical Biosciences and Engineering, 2023, 20(5): 9364-9384. doi: 10.3934/mbe.2023411
Perceptual grouping along well-established Gestalt laws provides one set of traditional methods that provide a tiny set of meaningful parameters to be adjusted for each application field. More complex and challenging tasks require a hierarchical setting, where the results aggregated by a first grouping process are later subject to further processing on a larger scale and with more abstract objects. This can be several steps deep. An example from the domain of forestry provides insight into the search for suitable parameter settings providing sufficient performance for the machine-vision module to be of practical use within a larger robotic control setting in this application domain. This sets a stark contrast in comparison to the state-of-the-art deep-learning neural nets, where many millions of obscure parameters must be adjusted properly before the performance suffices. It is the opinion of the author that the huge freedom for possible settings in such a high-dimensional inscrutable parameter space poses an unnecessary risk. Moreover, few-parameter learning is getting along with less training material. Whereas the state-of-the-art networks require millions of images with expert labels, a single image can already provide good insight into the nature of the parameter domain of the Gestalt laws, and a domain expert labeling just a handful of salient contours in said image yields already a proper goal function, so that a well working sweet spot in the parameter domain can be found in a few steps. As compared to the state-of-the-art neural nets, a reduction of six orders of magnitude in the number of parameters results. Almost parameter-free statistical test methods can reduce the number of parameters to be trained further by one order of magnitude, but they are less flexible and currently lack the advantages of hierarchical feature processing.
| [1] |
Y. LeCun, B. Boser, J. S. Denker, D. Henderson, R. E. Howard, W. Hubbard, et al., Backpropagation applied to handwritten zip code recognition, Neural Comput., 1 (1989), 541–551. https://doi.org/10.1162/neco.1989.1.4.541 doi: 10.1162/neco.1989.1.4.541
|
| [2] |
A. Krizhevsky, I. Sutskever, G. E. Hinton, ImageNet classification with deep convolutional neural networks, Commun. ACM, 60 (2017), 84–90. https://doi.org/10.1145/3065386 doi: 10.1145/3065386
|
| [3] | K. Simonyan, A. Zisserman, Very deep convolutional networks for large-scale image recognition, preprint, arXiv: 1409.1556. |
| [4] | A. Bochkovskiy, C. Y. Wang, H. Y. M. Liao, YOLOv4: Optimal speed and accuracy of object detection, preprint, arXiv: 2004.1093. |
| [5] | A. Gruen, O. Kuebler, P. Agouris, Automatic Extraction of Man-made Objects from Aerial and Space Images, Birkhä user Verlag, Basel, 1995. |
| [6] |
J. Dai, R. Ma, L. Gong, Z. Shen, J. Wu, A model-driven-to-sample-driven method for rural road extraction, Remote Sens., 13 (2021), 1417. https://doi.org/10.3390/rs13081417 doi: 10.3390/rs13081417
|
| [7] | P. Viola, M. Jone, Rapid object detection using a boosted cascade of simple features, in Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2001, 1 (2001). https://doi.org/10.1109/CVPR.2001.990517 |
| [8] |
O. Ludwig, U. Nunes, B. Ribeiro, C. Premebida, Improving the generalization capacity of cascade classifiers, IEEE Trans. Cybern., 46 (2013), 2135–2146. https://doi.org/10.1109/TCYB.2013.2240678 doi: 10.1109/TCYB.2013.2240678
|
| [9] | T. Matsuyama, V. S. S. Hwang, SIGMA A Knowledge-based Aerial Image Understanding System, Plenum Press, 1990. |
| [10] |
M. Wertheimer, Untersuchungen zur Lehre der Gestalt. Ⅱ, Psychologische Forschung, 4 (1923), 301–250. https://doi.org/10.1007/BF00410640 doi: 10.1007/BF00410640
|
| [11] | D. Marr, Vision, Freeman & Co., San Francisco, 1982. |
| [12] | D. G. Lowe, Perceptual Organization and Visual Recognition, Kluwer, Boston, 1986. |
| [13] | Z. Pizlo, Y. Li, T. Sawada, R. M. Steinman, Making a Machine that Sees like Us, Oxford University Press, USA, 2014. |
| [14] | A. Desolneux, L. Moisan, J. M. Morel, From Gestalt Theory to Image Analysis: A Probabilistic Approach, Springer, Berlin, 2008. |
| [15] | E. Michaelsen, J. Meidow, Hierarchical Perceptual Grouping for Object Recognition, Springer, Berlin, 2019. |
| [16] |
E. Michaelsen, Self-organizing maps and Gestalt organization as components of an advanced system for remotely sensed data: An example with thermal hyper-spectra, Pattern Recogn. Lett., 83 (2016), 169–177. https://doi.org/10.1016/j.patrec.2016.06.004 doi: 10.1016/j.patrec.2016.06.004
|
| [17] | C. Funk, S. Lee, M. R. Oswald, S. Tsogkas, W. Shen, A. Cohen, et al., 2017 ICCV challenge: detecting symmetry in the wild, in Proceedings of the IEEE International Conference on Computer Vision Workshops, (2017), 1692–1701. |
| [18] | E. Michaelsen, A. Schunert, U. Soergel, Utilizing phase for the grouping of PS in urban high-resolution in-SAR-images, in 2011 Joint Urban Remote Sensing Event, IEEE, (2011), 189–192. https://doi.org/10.1109/JURSE.2011.5764752 |
| [19] |
E. Michaelsen, J. Meidow, Stochastic reasoning for structural pattern recognition: an example from image-based UAV navigation, Pattern Recogn., 47 (2014), 2732–2744. https://doi.org/10.1016/j.patcog.2014.02.009 doi: 10.1016/j.patcog.2014.02.009
|
| [20] | Amarjot, Canny Edge Detector, Available from: https://www.mathworks.com/matlabcentral/fileexchange/40737-canny-edge-detector. |
| [21] | V. N. Vapnik, A. Ya, Chervonenkis: Theory of Pattern Recognition, Nauka, 1974. |
| [22] | L. Gao, Latin Squares in Experimental Design, Michigan State University, 2005. |
Figures(9) / Tables(1)
Eckart Michaelsen. Few-parameter learning for a hierarchical perceptual grouping system[J]. Mathematical Biosciences and Engineering, 2023, 20(5): 9364-9384. doi: 10.3934/mbe.2023411
DownLoad: