Abstract
Bright objects on a dark background, such as cells in microscopy images, can sometimes be modeled as maxima of sufficient dynamic, called h-maxima. Such a model could be sufficient to count these objects in images, provided we know the dynamic threshold that tells apart actual objects from irrelevant maxima. In this paper we introduce a neural architecture that includes a morphological pipeline counting the number of h-maxima in an image, preceded by a classical CNN which predicts the dynamic h yielding the right number of objects. This is made possible by geodesic reconstruction layers, already introduced in previous work, and a new module counting connected components. This architecture is trained end-to-end to count melanocytes in microscopy images. Its performance is close to the state of the art CNN on this dataset, with much fewer parameters (1/100) and an increased interpretability.
S. Velasco-Forero—This work was granted access to the HPC resources of IDRIS under the allocation 2023-AD011012212R2 made by GENCI.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The dataset is available at https://bit.ly/melanocytesTRP1.
- 2.
Code available at https://github.com/peter12398/DGMM2024-comptage-cellule.
- 3.
Convolutional layers are implemented as a double convolution with twice the same number of filters.
- 4.
Tested values: \(\alpha = 1\) and \(\beta \in \{0\} \cup 10^{\{-4,-3,-2,-1,0,1,2\}} \cup 5\times 10^{\{-4,-3,-2\}}\).
References
Duval, C., Cohen, C., Chagnoleau, C., Flouret, V., Bourreau, E., Bernerd, F.: Key regulatory role of dermal fibroblasts in pigmentation as demonstrated using a reconstructed skin model: impact of photo-aging. PLoS ONE 9(12), e114182 (2014)
Guan, S., Loew, M.: Understanding the ability of deep neural networks to count connected components in images. In: 2020 IEEE Applied Imagery Pattern Recognition Workshop (AIPR), pp. 1–7. IEEE (2020)
He, S., Minn, K.T., Solnica-Krezel, L., Anastasio, M.A., Li, H.: Deeply-supervised density regression for automatic cell counting in microscopy images. Med. Image Anal. 68, 101892 (2021)
Lazard, T., et al.: Applying deep learning to melanocyte counting on fluorescent TRP1 labelled images of in vitro skin model. Image Anal. Stereol. (2022)
Najman, L., Talbot, H.: Mathematical Morphology: From Theory to Applications. Wiley, Hoboken (2013)
Rall, L.B.: Automatic Differentiation: Techniques and Applications. Springer, Heidelberg (1981)
Sangalli, M., Blusseau, S., Velasco-Forero, S., Angulo, J.: Scale equivariant neural networks with morphological scale-spaces. In: Lindblad, J., Malmberg, F., Sladoje, N. (eds.) DGMM 2021. LNCS, vol. 12708, pp. 483–495. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-76657-3_35
Soille, P., et al.: Morphological Image Analysis: Principles and Applications, vol. 2. Springer, Cham (1999)
Tarjan, R.: Depth-first search and linear graph algorithms. SIAM J. Comput. 1(2), 146–160 (1972)
Tieleman, T., Hinton, G., et al.: Lecture 6.5-RMSPROP: divide the gradient by a running average of its recent magnitude. COURSERA: Neural Netw. Mach. Learn. 4(2), 26–31 (2012)
Velasco-Forero, S.: Morpholayers (2020). https://github.com/Jacobiano/morpholayers
Velasco-Forero, S., Angulo, J.: Learnable empirical mode decomposition based on mathematical morphology. SIAM J. Imaging Sci. 15(1), 23–44 (2022)
Velasco-Forero, S., Rhim, A., Angulo, J.: Fixed point layers for geodesic morphological operations. In: British Machine Vision Conference (2022)
Vincent, L.: Morphological grayscale reconstruction in image analysis: applications and efficient algorithms. IEEE Trans. Image Process. 2(2), 176–201 (1993)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Liu, X., Blusseau, S., Velasco-Forero, S. (2024). Counting Melanocytes with Trainable h-Maxima and Connected Component Layers. In: Brunetti, S., Frosini, A., Rinaldi, S. (eds) Discrete Geometry and Mathematical Morphology. DGMM 2024. Lecture Notes in Computer Science, vol 14605. Springer, Cham. https://doi.org/10.1007/978-3-031-57793-2_32
Download citation
DOI: https://doi.org/10.1007/978-3-031-57793-2_32
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-57792-5
Online ISBN: 978-3-031-57793-2
eBook Packages: Computer ScienceComputer Science (R0)