Abstract
In this article, we propose a method for designing loss functions based on component trees that can be optimized by gradient descent algorithms and are therefore usable in conjunction with recent machine learning approaches such as neural networks. The nodes of this tree are the connected components of the upper level sets of an image and the leaves represent the regional maxima (or regional minima if the dual tree is considered) of the image, i.e., connected sets of bright pixels surrounded by darker pixels. The proposed loss function is thus defined at the level of connected components rather than at the level of individual pixels, which allows for the optimization of higher semantic level quantities such as topological features. We show how the altitudes associated with the nodes of such hierarchical image representations can be differentiated with respect to the values of the image pixels. This property is used to design a generic loss function that can select or discard image maxima based on various attributes, such as extinction values based on the contrast or the size of the maxima. The possibilities of the proposed method are demonstrated on simulated and real image filtering.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Ballester, C., Caselles, V., Monasse, P.: The tree of shapes of an image. In: ESAIM: Control, Optimisation and Calculus of Variations, vol. 9, pp. 1–18 (2003)
Boutry, N., Géraud, T., Najman, L.: An equivalence relation between morphological dynamics and persistent homology in 1D. In: Burgeth, B., Kleefeld, A., Naegel, B., Passat, N., Perret, B. (eds.) ISMM 2019. LNCS, vol. 11564, pp. 57–68. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-20867-7_5
Chierchia, G., Perret, B.: Ultrametric fitting by gradient descent. In: NeurIPS, pp. 3181–3192 (2019)
Clough, J., Byrne, N., Oksuz, I., Zimmer, V., Schnabel, J., King, A.: A topological loss function for deep-learning based image segmentation using persistent homology. IEEE TPAMI (2020)
Clough, J.R., Oksuz, I., Byrne, N., Schnabel, J.A., King, A.P.: Explicit topological priors for deep-learning based image segmentation using persistent homology. In: Chung, A.C.S., Gee, J.C., Yushkevich, P.A., Bao, S. (eds.) IPMI 2019. LNCS, vol. 11492, pp. 16–28. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-20351-1_2
Cousty, J., Najman, L.: Incremental algorithm for hierarchical minimum spanning forests and saliency of watershed cuts. In: Soille, P., Pesaresi, M., Ouzounis, G.K. (eds.) ISMM 2011. LNCS, vol. 6671, pp. 272–283. Springer, Heidelberg (2011). https://doi.org/10.1007/978-3-642-21569-8_24
Dalla Mura, M., Benediktsson, J., Waske, B., Bruzzone, L.: Morphological attribute profiles for the analysis of very high resolution images. IEEE TGRS 48(10), 3747–3762 (2010)
Donoser, M., Bischof, H.: Efficient maximally stable extremal region (MSER) tracking. In: IEEE CVPR, vol. 1, pp. 553–560 (2006)
Gabrielsson, R., Nelson, B., Dwaraknath, A., Skraba, P.: A topology layer for machine learning. In: AISTATS, pp. 1553–1563. PMLR (2020)
Géraud, T., Carlinet, E., Crozet, S., Najman, L.: A quasi-linear algorithm to compute the tree of shapes of nD images. In: Hendriks, C.L.L., Borgefors, G., Strand, R. (eds.) ISMM 2013. LNCS, vol. 7883, pp. 98–110. Springer, Heidelberg (2013). https://doi.org/10.1007/978-3-642-38294-9_9
Girshick, R., Donahue, J., Darrell, T., Malik, J.: Rich feature hierarchies for accurate object detection and semantic segmentation. In: IEEE CVPR, pp. 580–587 (2014)
Hu, X., Li, F., Samaras, D., Chen, C.: Topology-preserving deep image segmentation. In: NeurIPS, pp. 5657–5668 (2019)
Jones, R.: Connected filtering and segmentation using component trees. CVIU 75(3), 215–228 (1999)
Kingma, D., Ba, J.: Adam: a method for stochastic optimization. In: Bengio, Y., LeCun, Y. (eds.) ICLR (2015)
Paszke, A., et al.: Pytorch: an imperative style, high-performance deep learning library. In: NeurIPS, pp. 8026–8037 (2019)
Perret, B., Chierchia, G., Cousty, J., Guimarães, S., Kenmochi, Y., Najman, L.: Higra: hierarchical graph analysis. SoftwareX 10, 100335 (2019)
Perret, B., Cousty, J., Guimaraes, S., Maia, D.: Evaluation of hierarchical watersheds. IEEE TIP 27(4), 1676–1688 (2017)
Robic, J., Perret, B., Nkengne, A., Couprie, M., Talbot, H.: Self-dual pattern spectra for characterising the dermal-epidermal junction in 3D reflectance confocal microscopy imaging. In: Burgeth, B., Kleefeld, A., Naegel, B., Passat, N., Perret, B. (eds.) ISMM 2019. LNCS, vol. 11564, pp. 508–519. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-20867-7_39
Salembier, P., Oliveras, A., Garrido, L.: Anti-extensive connected operators for image and sequence processing. IEEE TIP 7(4), 555–570 (1998)
Teeninga, P., Moschini, U., Trager, S., Wilkinson, M.: Statistical attribute filtering to detect faint extended astronomical sources. Math. Morphol.-Theory Appl. 1 (2016)
Vachier, C., Meyer, F.: Extinction value: a new measurement of persistence. In: IEEE Workshop on Nonlinear Signal and Image Processing, vol. 1, pp. 254–257 (1995)
Xu, Y., Carlinet, E., Géraud, T., Najman, L.: Hierarchical segmentation using tree-based shape spaces. IEEE TPAMI 39(3), 457–469 (2016)
Xu, Y., Monasse, P., Géraud, T., Najman, L.: Tree-based Morse regions: a topological approach to local feature detection. IEEE TIP 23(12), 5612–5625 (2014)
Acknowledgements
This work was supported by the French ANR grant ANR-20-CE23-0019.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 Springer Nature Switzerland AG
About this paper
Cite this paper
Perret, B., Cousty, J. (2022). Component Tree Loss Function: Definition and Optimization. In: Baudrier, É., Naegel, B., Krähenbühl, A., Tajine, M. (eds) Discrete Geometry and Mathematical Morphology. DGMM 2022. Lecture Notes in Computer Science, vol 13493. Springer, Cham. https://doi.org/10.1007/978-3-031-19897-7_20
Download citation
DOI: https://doi.org/10.1007/978-3-031-19897-7_20
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-19896-0
Online ISBN: 978-3-031-19897-7
eBook Packages: Computer ScienceComputer Science (R0)