Abstract
Morphological image operators are a class of non-linear image mappings studied in Mathematical Morphology. Many significant theoretical results regarding the characterization of families of image operators, their properties, and representations are derived from lattice theory, the underlying foundation of Mathematical Morphology. A fundamental representation result is a pair of canonical decompositions for any translation-invariant operator as a union of sup-generating or an intersection of inf-generating operators, which in turn can be written in terms of two basic operators, erosions and dilations. Thus, in practice, a toolbox with functional operators can be built by composing erosions and dilations, and then operators of the toolbox can be further combined to solve image processing problems. However, designing image operators by hand may become a daunting task for complex image processing tasks, and this motivated the development of machine learning based approaches. This paper reviews the main contributions around this theme made by the authors and their collaborators over the years. The review covers the relevant theoretical elements, particularly the canonical decomposition theorem, a formulation of the learning problem, some methods to solve it, and algorithms for finding computationally efficient representations. More recent contributions included in this review are related to families of operators (hypothesis spaces) organized as lattice structures where a suitable subfamily of operators is searched through the minimization of U-curve cost functions. A brief account of the connections between morphological image operator learning and deep learning is also included.
Similar content being viewed by others
Notes
Note that the target images \(T_i\) are in general manually generated from observed images \(S_i\), usually with the aid of an image editor.
References
Abu-Mostafa, Y., Lin, H., Magdon-Ismail, M.: Learning From Data. AMLBook (2012)
Atashpaz-Gargari, E., Reis, M., Braga-Neto, U., Barrera, J., Dougherty, E.: A fast branch-and-bound algorithm for U-curve feature selection. Pattern Recognit. (2018). https://doi.org/10.1016/j.patcog.2017.08.013
Banon, G., Barrera, J.: Minimal representations for translation-invariant set mappings by Mathematical Morphology. SIAM J. Appl. Math. (1991). https://doi.org/10.1137/0151091
Banon, G., Barrera, J.: Decomposition of mappings between complete lattices by Mathematical Morphology. Part I. General lattices. Signal Process. (1993). https://doi.org/10.1016/0165-1684(93)90015-3
Banon, G., Barrera, J.: Bases da Morfologia Matemática para a Análise de Imagens Binárias. IX Escola de Computação, Recife (1994) (In Portuguese)
Barrera, J., Salas, G.: Set operations on closed intervals and their applications to the automatic programming of morphological machines. J. Electron. Imaging (1996). https://doi.org/10.1117/12.240717
Barrera, J., Banon, G., Lotufo, R.: Mathematical morphology toolbox for the khoros system. In: Proceedings of the SPIE 2300, Image Algebra and Morphological Image Processing V, 30 June 1994. SPIE (1993). https://doi.org/10.1117/12.179195
Barrera, J., Dougherty, E., Tomita, N.: Automatic programming of binary morphological machines by design of statistically optimal operators in the context of computational learning theory. J. Electron. Imag. (1997). https://doi.org/10.1117/12.260010
Barrera, J., Banon, G., Lotufo, R., Hirata, R., Jr.: MMach: a mathematical morphology toolbox for the KHOROS system. J. Electron. Imaging (1998). https://doi.org/10.1117/1.482637
Barrera, J., Terada, R., Lotufo, R.A., Hirata, N.S.T., Hirata, R., Jr., Zampirolli, F.A.: An OCR based on Mathematical Morphology. In: Nonlinear Image Processing IX. Proceedings of SPIE, vol. 3304, pp. 197–208 (1998). https://doi.org/10.1117/12.304600
Barrera, J., Terada, R., Hirata, R., Jr., Hirata, N.: Automatic programming of morphological machines by PAC learning. Fundam. Inform. (2000). https://doi.org/10.3233/FI-2000-411208
Beucher, S.: Watersheds of functions and picture segmentation. In: Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 82), pp. 1928–1931 (1982). https://doi.org/10.1109/ICASSP.1982.1171424
Brun, M., Dougherty, E., Hirata, R., Barrera, J.: Design of optimal binary filters under joint multiresolution-envelope constraint. Pattern Recognit. Lett. (2003). https://doi.org/10.1016/S0167-8655(02)00217-9
Brun, M., Hirata, R., Jr., Barrera, J., Dougherty, E.: Nonlinear filter design using envelopes. J. Math. Imaging Vis. (2004). https://doi.org/10.1023/B:JMIV.0000026558.10581.e6
Castro, J.: Model selection for learning boolean hypothesis. PhD thesis, Instituto de Matemática e Estatística, Universidade de São Paulo, Brazil (2018). https://doi.org/10.11606/T.45.2019.tde-02042019-231050
Dellamonica, D., Jr., Silva, P., Humes, C., Jr., Hirata, N., Barrera, J.: An exact algorithm for optimal MAE stack filter design. IEEE Trans. Image Process. (2007). https://doi.org/10.1109/TIP.2006.888358
Dornelles, M., Hirata, N.: A genetic algorithm based approach for combining binary image operators. In: Proceedings of 21st International Conference on Pattern Recognition, pp. 3184–3187 (2012)
Dornelles, M., Hirata, N.: Selection of windows for W-operator combination from entropy based ranking. In: 28th SIBGRAPI Conference on Graphics, Patterns and Images, pp. 64–71 (2015)
Dougherty, E.: Optimal mean-square N-observation digital morphological filters I. Optimal binary filters. CVGIP Image Underst. (1992). https://doi.org/10.1016/1049-9660(92)90006-O
Dougherty, E.R., Giardina, C.R.: Error bounds for morphologically derived measurements. SIAM J. Appl. Math. (1987). https://doi.org/10.1137/0147028
Dougherty, E.R., Haralick, M.: Unification of nonlinear filtering in the context of binary logical calculus, part I: binary filters. J. Math. Imaging Vis. 2(2), 173–183 (1992). https://doi.org/10.1007/bf00118588
Dougherty, E., Lotufo, R.: Hands-on Morphological Image Processing, vol. 59. SPIE Press, Bellingham (2003)
Dougherty, E.R., Barrera, J., Mozelle, G., Seungchan, K., Brun, M.: Multiresolution analysis for optimal binary filters. J. Math. Imaging Vis. (2001). https://doi.org/10.1023/A:1008311431244
Estrela, G., Gubitoso, M., Ferreira, C., Barrera, J., Reis, M.: An efficient, parallelized algorithm for optimal conditional entropy-based feature selection. Entropy (2020). https://doi.org/10.3390/e22040492
Franchi, G., Fehri, A., Yao, A.: Deep morphological networks. Pattern Recognit. (2020). https://doi.org/10.1016/j.patcog.2020.107246
Freeman, H.: Computer processing of line-drawing images. ACM Comput. 6(1), 57–97 (1974). https://doi.org/10.1145/356625.356627
Gonzalez, R., Woods, R.: Digital Image Processing, 3rd edn. Prentice-Hall Inc, Upper Saddle River (2006)
Goodfellow, I., Bengio, Y., Courville, A.: Deep Learning. MIT, Cambridge (2016)
Goutsias, J.: Morphological analysis of discrete random shapes. J. Math. Imaging Vis. (1992). https://doi.org/10.1007/BF00118590
Haralick, R., Sternberg, S., Zhuang, X.: Image analysis using mathematical morphology. IEEE Trans. Pattern Anal. Mach. Intell. (1987). https://doi.org/10.1109/TPAMI.1987.4767941
Harvey, N., Marshall, S.: The use of genetic algorithms in morphological filter design. Signal Process. Image Commun. (1996). https://doi.org/10.1016/0923-5965(95)00033-X
Hashimoto, R., Barrera, J.: From the sup-decomposition to sequential decompositions. J. Math. Imaging Vis. (2001). https://doi.org/10.1023/A:1012276422769
Hashimoto, R., Barrera, J.: Sup-compact and Inf-compact representations of W-Operators. Fundam. Inform. 45(4), 283–294 (2001)
Hashimoto, R., Barrera, J.: A note on Park and Chin’s algorithm. IEEE Trans. Pattern Anal. Mach. Intell. (2002). https://doi.org/10.1109/34.982891
Hashimoto, R., Barrera, J.: A greedy algorithm for decomposing convex structuring elements. J. Math. Imaging Vis. (2003). https://doi.org/10.1023/A:1022847527229
Hashimoto, R., Barrera, J., Ferreira, C.: A combinatorial optimization technique for the sequential decomposition of erosions and dilations. J. Math. Imaging Vis. (2000). https://doi.org/10.1023/A:1008373522375
Hastie, T., Tibshirani, R., Friedman, J.: The Elements of Statistical Learning: Data Mining, Inference, and Prediction. Springer, New York (2009)
Heijmans, H.: Morphological Image Operators. Academic Press, Boston, MA, USA (1994)
Hils, D.: Visual languages and computing survey: data flow visual programming languages. J. Vis. Lang. Comput. 3(1), 69–101 (1992). https://doi.org/10.1016/1045-926X(92)90034-J
Hirata, N.: Multilevel training of binary morphological operators. IEEE Trans. Pattern Anal. Mach. Intell. (2009). https://doi.org/10.1109/TPAMI.2008.118
Hirata, N., Barrera, J.: A unifying view for stack filter design based on graph search methods. Pattern Recognit. (2005). https://doi.org/10.1016/j.patcog.2005.02.018
Hirata, N.S.T., Papakostas, G.A.: Omacn hine-learning morphological image operators. Mathematics (2021). https://doi.org/10.3390/math9161854
Hirata, N., Barrera, J., Dougherty, E.: Design of statistically optimal stack filters. In: XII Brazilian Symposium on Computer Graphics and Image Processing, pp. 265–274 (1999). https://doi.org/10.1109/SIBGRA.1999.805734
Hirata, N., Dougherty, E., Barrera, J.: A switching algorithm for design of optimal increasing binary filters over large windows. Pattern Recognit. (2000). https://doi.org/10.1016/S0031-3203(99)00165-X
Hirata, N., Dougherty, E., Barrera, J.: Iterative design of morphological binary image operators. Opt. Eng. (2000). https://doi.org/10.1117/1.1327178
Hirata, R., Jr., Dougherty, E., Barrera, J.: Aperture filters. Signal Process. (2000). https://doi.org/10.1016/S0165-1684(99)00162-0
Hirata, N., Barrera, J., Terada, R., Dougherty, E.: The incremental splitting of intervals algorithm for the design of binary image operators. In: Talbot, H., Beare, R. (eds.) Proceedings of 6th International Symposium on Mathematical Morphology, pp. 219–228. CSIRO Publishing, Clayton (2002)
Hirata, R., Barrera, J., Hashimoto, R., Dantas, D., Esteves, G.: Segmentation of microarray images by mathematical morphology. Real-Time Imaging 8(6), 491–505 (2002). https://doi.org/10.1006/rtim.2002.0291
Hirata, R., Jr., Brun, M., Barrera, J., Dougherty, E.R.: Multiresolution design of aperture operators. J. Math. Imaging Vis. (2002). https://doi.org/10.1023/A:1020377610141
Julca-Aguilar, F., Hirata, N.: Image operator learning coupled with CNN classification and its application to staff line removal. In: 14th IAPR International Conference on Document Analysis and Recognition (2017). https://doi.org/10.1109/ICDAR.2017.18
Karp, R.: Reducibility among combinatorial problems. In: R. Miller, J. Thatcher (eds.) Proceedings of a symposium on the Complexity of Computer Computations. Plenum Press, New York (1972). https://doi.org/10.1007/978-1-4684-2001-2_9
Kleitman, D., Markowsky, G.: Dedekind’s problem: the number of isotone Boolean functions. II. Trans. Am. Math. Soc. 213, 373–390 (1975). https://doi.org/10.1090/S0002-9947-1975-0382107-0
Konstantinides, K., Rasure, J.R.: The Khoros software development environment for image and signal processing. IEEE Trans. Image Process. (1994). https://doi.org/10.1109/83.287018
LeCun, Y., Boser, B., Denker, J.S., Henderson, D., Howard, R.E., Hubbard, W., Jackel, L.D.: Backpropagation applied to handwritten zip code recognition. Neural Comput. 1(4), 541–551 (1989). https://doi.org/10.1162/neco.1989.1.4.541
Long, J., Shelhamer, E., Darrell, T.: Fully convolutional networks for semantic segmentation. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (2015). https://doi.org/10.1109/CVPR.2015.7298965
Maragos, P.A.: A unified theory of translation-invariant systems with applications to morphological analysis and coding of images. PhD thesis, School of Electrical and Computer Engineering (ECE), Georgia Institute of Technology (1985)
Marcondes, D., Simonis, A., Barrera, J.: Learning the hypotheses space from data: Learning space and u-curve property (2020). arxiv:2001.09532. ArXiv preprint
Marcondes, D., Simonis, A., Barrera, J.: Learning the hypotheses space from data, part II: convergence and feasibility. arXiv preprint (2020). arxiv:2001.11578
Martins-Jr., D., Cesar-Jr., R., Barrera, J.: W-operator window design by minimization of mean conditional entropy. Pattern Anal. Appl. (2006). https://doi.org/10.1007/s10044-006-0031-0
Matheron, G.: Random Sets and Integral Geometry. Wiley, New York (1975)
Matheron, G., Serra, J.: The birth of mathematical morphology. In: Talbot, H., Beare, R. (eds.) Proceedings of 6th International Symposium on Mathematical Morphology, pp. 1–16. CSIRO Publishing, Clayton (2002)
Mondal, R., Purkait, P., Santra, S., Chanda, B.: Morphological networks for image de-raining. In: Couprie, M., Cousty, J., Kenmochi, Y., Mustafa, N. (eds.) Discrete Geometry for Computer Imagery. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-14085-4_21
Montagner, I., Hirata Jr, R., Hirata, N., Canu, S.: Kernel approximations for w-operator learning. In: 29th SIBGRAPI Conference on Graphics, Patterns and Images, pp. 386–393 (2016)
Montagner, I., Hirata, N., Hirata, R., Jr.: Staff removal using image operator learning. Pattern Recognit. (2017). https://doi.org/10.1016/j.patcog.2016.10.002
Park, H., Chin, R.: decomposition of arbitrarily shaped morphological structuring elements. IEEE Trans. Pattern Anal. Mach. Intell. (1995). https://doi.org/10.1109/34.368156
Pavlidis, T.: Algorithms for Graphics and Image Processing. Computer Science Press, Rockville (1982)
Reis, M.: Minimização de funções decomponíveis em curvas em U definidas sobre cadeias de posets – algoritmos e aplicações. PhD thesis, Instituto de Matemática e Estatística, Universidade de São Paulo, Brazil (2012). https://doi.org/10.11606/T.45.2012.tde-05022013-123757
Reis, M., Barrera, J.: Solving problems in mathematical morphology through reductions to the u-curve problem. In: International Symposium on Mathematical Morphology and Its Applications to Signal and Image Processing. Springer (2013). https://doi.org/10.1007/978-3-642-38294-9_5
Reis, M., Estrela, G., Ferreira, C., Barrera, J.: featsel: a framework for benchmarking of feature selection algorithms and cost functions. SoftwareX 6, 193–197 (2017). https://doi.org/10.1016/j.softx.2017.07.005
Reis, M., Estrela, G., Ferreira, C., Barrera, J.: Optimal boolean lattice-based algorithms for the U-curve optimization problem. Inf. Sci. (2019). https://doi.org/10.1016/j.ins.2018.08.060
Ris, M., Barrera, J., Martins-Jr, D.: U-curve: A branch-and-bound optimization algorithm for U-shaped cost functions on Boolean lattices applied to the feature selection problem. Pattern Recognit. (2010). https://doi.org/10.1016/j.patcog.2009.08.018
Ronneberger, O., Fischer, P., Brox, T.: U-net: Convolutional networks for biomedical image segmentation. In: International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer (2015). https://doi.org/10.1007/978-3-319-24574-4_28
Ronse, C.: Why mathematical morphology needs complete lattices. Signal Process. 21(2), 129–154 (1990). https://doi.org/10.1016/0165-1684(90)90046-2
Ronse, C.: A lattice-theoretical morphological view on template extraction in images. J. Vis. Commun. Image Represent. (1996). https://doi.org/10.1006/jvci.1996.0024
Ruderman, D.: The statistics of natural images. Netw. Comput. Neural Syst. (1994). https://doi.org/10.1088/0954-898X_5_4_006
Santos, C., Hirata, N., Hirata, R., Jr.: An information theory framework for two-stage binary image operator design. Pattern Recognit. Lett. (2010). https://doi.org/10.1016/j.patrec.2009.03.019
Schmitt, M.: Mathematical morphology and artificial intelligence: an automatic programming system. Signal Process. 16(4), 389–401 (1989). https://doi.org/10.1016/0165-1684(89)90032-7
Serra, J.: Image Analysis and Mathematical Morphology. Academic Press, Cambridge, MA, USA (1982)
Serra, J.: Image Analysis and Mathematical Morphology. Vol. 2: Theoretical Advances. Academic Press, Cambridge (1988)
Serra, J., Vincent, L.: An overview of morphological filtering. Circuits Syst. Signal Process. 11(1), 47–108 (1992). https://doi.org/10.1007/BF01189221
Soille, P.: Morphological Image Analysis. Springer, Berlin (2003)
Sternberg, S.: Grayscale morphology. Comput. Vis. Graph. Image Process. 35(3), 333–355 (1986). https://doi.org/10.1016/0734-189X(86)90004-6
Van Droogenbroeck, M., Talbot, H.: Fast computation of morphological operations with arbitrary structuring elements. Pattern Recognit. Lett. 17(14), 1451–1460 (1996). https://doi.org/10.1016/S0167-8655(96)00113-4
Vincent, L.M.: Efficient computation of various types of skeletons. In: Loew, M.H. (ed.) Medical Imaging V: Image Processing, vol. 1445, pp. 297–311. International Society for Optics and Photonics, SPIE, Bellingham (1991). https://doi.org/10.1117/12.45227
Wilson, S.S.: Training structuring elements in morphological networks. In: Dougherty, E.R. (ed.) Mathematical Morphology in Image Processing, Chap. 1, pp. 1–41. Marcel Dekker, New York (1993). https://doi.org/10.1201/9781482277234
Zhang, A., Lipton, Z., Li, M., Smola, A.: Dive into Deep Learning. Self-published, Open Source Book (2020). https://d2l.ai
Acknowledgements
The authors thank Professors George Matheron and Jean Serra for their scientific interaction with Barrera and Hashimoto, Professor Gerald J. F. Banon for pioneering Mathematical Morphology in Brazil and inspiring young researchers, Professor Roberto Lotufo for introducing Khoros/Cantata programming language and collaborating with our group for many years, Professor Edward R. Dougherty for his inspiring work on the statistical design of image operators and the fruitful collaboration with our group for so long. In particular, Professor Banon advised Barrera’s Doctorate and collaborated with IME’s group for many years at the beginning of our group. The authors also acknowledge all their former collaborators and former students for their contributions to develop methods and tools for the automatic programming of image operators. The authors acknowledge CNPq, FAPESP, and Olivetti do Brasil for the financial support that allowed the development of this research. This study was also financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Finance Code 001.
Funding
This study was supported by FAPESP (Grants 2019/21619-5, 2017/25835-9, 2015/22308-2, 2013/07467-1).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Additional information
Communicated by Yoshiharu Kohayakawa.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Barrera, J., Hashimoto, R.F., Hirata, N.S.T. et al. From Mathematical Morphology to machine learning of image operators. São Paulo J. Math. Sci. 16, 616–657 (2022). https://doi.org/10.1007/s40863-022-00303-1
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s40863-022-00303-1
Keywords
- Mathematical Morphology
- Image operator
- Machine learning
- Automatic design
- Operator representation
- Operator decomposition