可學習的形態學層
選讀・IP 09・形態學影像處理 的補充資料(第 9/9 則)。主線不依賴這則補充;引用編號對應本頁底部的參考文獻。
導覽:本頁與本系列: 導覽:本頁與本系列
最大池化(max pooling)本來就是平坦灰階膨脹再加上降取樣,所以兩者的連結很自然。有好幾個團隊設計了結構元素可以用梯度下降學習的網路層。Mondal 等人研究由可學習的膨脹與侵蝕神經元組成的網路,證明兩個「膨脹與侵蝕後接線性組合」的區塊就能逼近任意連續函數 [13]。Nogueira 等人提出 DeepMorphNet,以可學習的形態學運算取代卷積,並比較它學到的特徵與一般卷積網路的差異 [14]。實務上的難處是 min 與 max 的梯度很稀疏。Kirszenberg 等人 [16] 與 Hermary 等人 [15] 用 max 與 min 的可微近似(銳利程度由可學習參數控制的廣義平均與 soft-max 函數)建構平滑的形態學層,改進了較早的 p-convolution 層,並探討這些層何時真的能學回侵蝕與膨脹。這些仍是研究工具:形態學層並不是主流視覺骨幹網路的標準元件。
參考文獻
- R. C. Gonzalez and R. E. Woods, Digital Image Processing, 4th ed., Pearson, 2018, Ch. 9. publisher page
- G. Matheron, Random Sets and Integral Geometry, Wiley, 1975. library record
- J. Serra, Image Analysis and Mathematical Morphology, Academic Press, 1982. library record
- R. M. Haralick, S. R. Sternberg, and X. Zhuang, “Image Analysis Using Mathematical Morphology,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. PAMI-9, pp. 532–550, 1987. doi
- P. Soille, Morphological Image Analysis, Springer, 2004. doi
- T. Y. Zhang and C. Y. Suen, “A Fast Parallel Algorithm for Thinning Digital Patterns,” Communications of the ACM, vol. 27, pp. 236–239, 1984. doi
- L. Vincent, “Morphological Grayscale Reconstruction in Image Analysis: Applications and Efficient Algorithms,” IEEE Trans. Image Processing, vol. 2, pp. 176–201, 1993. doi
- P. Maragos, “Pattern Spectrum and Multiscale Shape Representation,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 11, no. 7, pp. 701–716, 1989. doi
- E. J. Breen and R. Jones, “Attribute Openings, Thinnings, and Granulometries,” Computer Vision and Image Understanding, vol. 64, no. 3, pp. 377–389, 1996. doi
- P. Salembier and M. H. F. Wilkinson, “Connected Operators: A Review of Region-Based Morphological Image Processing Techniques,” IEEE Signal Processing Magazine, vol. 26, no. 6, pp. 136–157, 2009. doi
- E. Carlinet and T. Géraud, “A Comparative Review of Component Tree Computation Algorithms,” IEEE Trans. Image Processing, vol. 23, pp. 3885–3895, 2014. doi
- S. van der Walt et al., “scikit-image: Image Processing in Python,” PeerJ, 2014. arXiv
- R. Mondal, S. Santra, S. S. Mukherjee, and B. Chanda, “Morphological Network: How Far Can We Go with Morphological Neurons?,” BMVC, 2022. arXiv
- K. Nogueira, J. Chanussot, M. Dalla Mura, and J. A. dos Santos, “An Introduction to Deep Morphological Networks,” arXiv:1906.01751, 2019. arXiv
- R. Hermary, G. Tochon, É. Puybareau, A. Kirszenberg, and J. Angulo, “Learning Grayscale Mathematical Morphology with Smooth Morphological Layers,” Journal of Mathematical Imaging and Vision, vol. 64, pp. 736–753, 2022. doi
- A. Kirszenberg, G. Tochon, É. Puybareau, and J. Angulo, “Going Beyond p-convolutions to Learn Grayscale Morphological Operators,” arXiv:2102.10038, 2021. arXiv
- S. Shit et al., “clDice — A Novel Topology-Preserving Loss Function for Tubular Structure Segmentation,” CVPR, 2021. arXiv