Task-Driven Learned Hyperspectral Data Reduction Using End-to-End Supervised Deep Learning
An important challenge in hyperspectral imaging tasks is to cope with the large number of spectral bins. Common spectral data reduction methods do not take prior knowledge about the task into account. Consequently, sparsely occurring features that may be essential for the imaging task may not be pre...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-12-01
|
| Series: | Journal of Imaging |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2313-433X/6/12/132 |
| _version_ | 1797545935694200832 |
|---|---|
| author | Mathé T. Zeegers Daniël M. Pelt Tristan van Leeuwen Robert van Liere Kees Joost Batenburg |
| author_facet | Mathé T. Zeegers Daniël M. Pelt Tristan van Leeuwen Robert van Liere Kees Joost Batenburg |
| author_sort | Mathé T. Zeegers |
| collection | DOAJ |
| description | An important challenge in hyperspectral imaging tasks is to cope with the large number of spectral bins. Common spectral data reduction methods do not take prior knowledge about the task into account. Consequently, sparsely occurring features that may be essential for the imaging task may not be preserved in the data reduction step. Convolutional neural network (CNN) approaches are capable of learning the specific features relevant to the particular imaging task, but applying them directly to the spectral input data is constrained by the computational efficiency. We propose a novel supervised deep learning approach for combining data reduction and image analysis in an end-to-end architecture. In our approach, the neural network component that performs the reduction is trained such that image features most relevant for the task are preserved in the reduction step. Results for two convolutional neural network architectures and two types of generated datasets show that the proposed Data Reduction CNN (DRCNN) approach can produce more accurate results than existing popular data reduction methods, and can be used in a wide range of problem settings. The integration of knowledge about the task allows for more image compression and higher accuracies compared to standard data reduction methods. |
| first_indexed | 2024-03-10T14:22:01Z |
| format | Article |
| id | doaj.art-1018fe423f94464dafb9bc0059fba8f1 |
| institution | Directory Open Access Journal |
| issn | 2313-433X |
| language | English |
| last_indexed | 2024-03-10T14:22:01Z |
| publishDate | 2020-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Imaging |
| spelling | doaj.art-1018fe423f94464dafb9bc0059fba8f12023-11-20T23:14:37ZengMDPI AGJournal of Imaging2313-433X2020-12-0161213210.3390/jimaging6120132Task-Driven Learned Hyperspectral Data Reduction Using End-to-End Supervised Deep LearningMathé T. Zeegers0Daniël M. Pelt1Tristan van Leeuwen2Robert van Liere3Kees Joost Batenburg4Centrum Wiskunde & Informatica, Science Park 123, 1098 XG Amsterdam, The NetherlandsCentrum Wiskunde & Informatica, Science Park 123, 1098 XG Amsterdam, The NetherlandsCentrum Wiskunde & Informatica, Science Park 123, 1098 XG Amsterdam, The NetherlandsCentrum Wiskunde & Informatica, Science Park 123, 1098 XG Amsterdam, The NetherlandsCentrum Wiskunde & Informatica, Science Park 123, 1098 XG Amsterdam, The NetherlandsAn important challenge in hyperspectral imaging tasks is to cope with the large number of spectral bins. Common spectral data reduction methods do not take prior knowledge about the task into account. Consequently, sparsely occurring features that may be essential for the imaging task may not be preserved in the data reduction step. Convolutional neural network (CNN) approaches are capable of learning the specific features relevant to the particular imaging task, but applying them directly to the spectral input data is constrained by the computational efficiency. We propose a novel supervised deep learning approach for combining data reduction and image analysis in an end-to-end architecture. In our approach, the neural network component that performs the reduction is trained such that image features most relevant for the task are preserved in the reduction step. Results for two convolutional neural network architectures and two types of generated datasets show that the proposed Data Reduction CNN (DRCNN) approach can produce more accurate results than existing popular data reduction methods, and can be used in a wide range of problem settings. The integration of knowledge about the task allows for more image compression and higher accuracies compared to standard data reduction methods.https://www.mdpi.com/2313-433X/6/12/132hyperspectral imagingfeature extractioncompressionmachine learningdeep learningconvolutional neural network |
| spellingShingle | Mathé T. Zeegers Daniël M. Pelt Tristan van Leeuwen Robert van Liere Kees Joost Batenburg Task-Driven Learned Hyperspectral Data Reduction Using End-to-End Supervised Deep Learning Journal of Imaging hyperspectral imaging feature extraction compression machine learning deep learning convolutional neural network |
| title | Task-Driven Learned Hyperspectral Data Reduction Using End-to-End Supervised Deep Learning |
| title_full | Task-Driven Learned Hyperspectral Data Reduction Using End-to-End Supervised Deep Learning |
| title_fullStr | Task-Driven Learned Hyperspectral Data Reduction Using End-to-End Supervised Deep Learning |
| title_full_unstemmed | Task-Driven Learned Hyperspectral Data Reduction Using End-to-End Supervised Deep Learning |
| title_short | Task-Driven Learned Hyperspectral Data Reduction Using End-to-End Supervised Deep Learning |
| title_sort | task driven learned hyperspectral data reduction using end to end supervised deep learning |
| topic | hyperspectral imaging feature extraction compression machine learning deep learning convolutional neural network |
| url | https://www.mdpi.com/2313-433X/6/12/132 |
| work_keys_str_mv | AT mathetzeegers taskdrivenlearnedhyperspectraldatareductionusingendtoendsuperviseddeeplearning AT danielmpelt taskdrivenlearnedhyperspectraldatareductionusingendtoendsuperviseddeeplearning AT tristanvanleeuwen taskdrivenlearnedhyperspectraldatareductionusingendtoendsuperviseddeeplearning AT robertvanliere taskdrivenlearnedhyperspectraldatareductionusingendtoendsuperviseddeeplearning AT keesjoostbatenburg taskdrivenlearnedhyperspectraldatareductionusingendtoendsuperviseddeeplearning |