%0 Conference Paper %1 KukushkinBogdanSchmid2024 %A Kukushkin, Maksim %A Bogdan, Martin %A Schmid, Thomas %B Sixteenth International Conference on Machine Vision (ICMV 2023) %D 2024 %E Osten, Wolfgang %I SPIE %K classification computer deep hyperspectral image learning mathematical morphological morphology networks neuronal nopdf remote sensing vision %P 1307202 %R 10.1117/12.3023593 %T On optimizing morphological neural networks for hyperspectral image classification %U https://doi.org/10.1117/12.3023593 %V 13072 %X Convolutional Neural Networks have become an important tool for various Computer Vision tasks. Yet, increasing complexity of such architectures drives computational costs. To this end, we propose two measures to achieve similar classification results as state-of-the-art architectures while at the same time reducing model complexity significantly. Firstly, we describe a novel type of non-linear parameter-efficient morphological layers inspired by concepts that are well-known and widely used with convolutions. Secondly, we present a set of simple network architectures, organized as optimization framework, which is enhanced by neural architecture search and hyperparameter optimization. In experiments with hyperspectral remote sensing data, we demonstrate that the identified optimal morphological architecture produces results not only comparable with other architectures from the optimization framework, but also comparable or better than selected state-of-the-art neural network architectures for image classification. Depending on the performed task, the proposed optimized architecture requires up to 25 times fewer parameters than actual state-of-the-art networks.

@inproceedings{KukushkinBogdanSchmid2024, abstract = {Convolutional Neural Networks have become an important tool for various Computer Vision tasks. Yet, increasing complexity of such architectures drives computational costs. To this end, we propose two measures to achieve similar classification results as state-of-the-art architectures while at the same time reducing model complexity significantly. Firstly, we describe a novel type of non-linear parameter-efficient morphological layers inspired by concepts that are well-known and widely used with convolutions. Secondly, we present a set of simple network architectures, organized as optimization framework, which is enhanced by neural architecture search and hyperparameter optimization. In experiments with hyperspectral remote sensing data, we demonstrate that the identified optimal morphological architecture produces results not only comparable with other architectures from the optimization framework, but also comparable or better than selected state-of-the-art neural network architectures for image classification. Depending on the performed task, the proposed optimized architecture requires up to 25 times fewer parameters than actual state-of-the-art networks.}, added-at = {2025-01-07T10:50:58.000+0100}, author = {Kukushkin, Maksim and Bogdan, Martin and Schmid, Thomas}, biburl = {https://puma.scadsai.uni-leipzig.de/bibtex/2d525396f5e263b8b1a2d3b40b053c528/scadsfct}, booktitle = {Sixteenth International Conference on Machine Vision (ICMV 2023)}, doi = {10.1117/12.3023593}, editor = {Osten, Wolfgang}, interhash = {c00b1f8263eb395db64c3c4879fbc299}, intrahash = {d525396f5e263b8b1a2d3b40b053c528}, keywords = {classification computer deep hyperspectral image learning mathematical morphological morphology networks neuronal nopdf remote sensing vision}, organization = {International Society for Optics and Photonics}, pages = 1307202, publisher = {SPIE}, timestamp = {2025-07-29T10:50:45.000+0200}, title = {On optimizing morphological neural networks for hyperspectral image classification}, url = {https://doi.org/10.1117/12.3023593}, volume = 13072, year = 2024 }