Compression Supports Spatial Deep Learning
In the last decades, the domain of spatial computing became more and more data driven, especially when using remote sensing-based images. Furthermore, the satellites provide huge amounts of images, so the number of available datasets is increasing. This leads to the need for large storage requiremen...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9969909/ |
| _version_ | 1797975322505773056 |
|---|---|
| author | Gabriel Dax Srilakshmi Nagarajan Hao Li Martin Werner |
| author_facet | Gabriel Dax Srilakshmi Nagarajan Hao Li Martin Werner |
| author_sort | Gabriel Dax |
| collection | DOAJ |
| description | In the last decades, the domain of spatial computing became more and more data driven, especially when using remote sensing-based images. Furthermore, the satellites provide huge amounts of images, so the number of available datasets is increasing. This leads to the need for large storage requirements and high computational costs when estimating the label scene classification problem using deep learning. This consumes and blocks important hardware recourses, energy, and time. In this article, the use of aggressive compression algorithms will be discussed to cut the wasted transmission and resources for selected land cover classification problems. To compare the different compression methods and the classification performance, the satellite image patches are compressed by two methods. The first method is the image quantization of the data to reduce the bit depth. Second is the lossy and lossless compression of images with the use of image file formats, such as JPEG and TIFF. The performance of the classification is evaluated with the use of convolutional neural networks (CNNs) like VGG16. The experiments indicated that not all remote sensing image classification problems improve their performance when taking the full available information into account. Moreover, compression can set the focus on specific image features, leading to fewer storage needs and a reduction in computing time with comparably small costs in terms of quality and accuracy. All in all, quantization and embedding into file formats do support CNNs to estimate the labels of images, by strengthening the features. |
| first_indexed | 2024-04-11T04:33:42Z |
| format | Article |
| id | doaj.art-78d769b8871d4202919686d046ba544c |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-04-11T04:33:42Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-78d769b8871d4202919686d046ba544c2022-12-29T00:00:07ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352023-01-011670271310.1109/JSTARS.2022.32265639969909Compression Supports Spatial Deep LearningGabriel Dax0https://orcid.org/0000-0002-1511-5770Srilakshmi Nagarajan1Hao Li2Martin Werner3https://orcid.org/0000-0002-6951-8022Department of Aerospace and Geodesy, TUM School of Engineering and Design, Professorship of Big Geospatial Data Management, Technical University of Munich, München, GermanyDepartment of Aerospace and Geodesy, TUM School of Engineering and Design, Professorship of Big Geospatial Data Management, Technical University of Munich, München, GermanyDepartment of Aerospace and Geodesy, TUM School of Engineering and Design, Professorship of Big Geospatial Data Management, Technical University of Munich, München, GermanyDepartment of Aerospace and Geodesy, TUM School of Engineering and Design, Professorship of Big Geospatial Data Management, Technical University of Munich, München, GermanyIn the last decades, the domain of spatial computing became more and more data driven, especially when using remote sensing-based images. Furthermore, the satellites provide huge amounts of images, so the number of available datasets is increasing. This leads to the need for large storage requirements and high computational costs when estimating the label scene classification problem using deep learning. This consumes and blocks important hardware recourses, energy, and time. In this article, the use of aggressive compression algorithms will be discussed to cut the wasted transmission and resources for selected land cover classification problems. To compare the different compression methods and the classification performance, the satellite image patches are compressed by two methods. The first method is the image quantization of the data to reduce the bit depth. Second is the lossy and lossless compression of images with the use of image file formats, such as JPEG and TIFF. The performance of the classification is evaluated with the use of convolutional neural networks (CNNs) like VGG16. The experiments indicated that not all remote sensing image classification problems improve their performance when taking the full available information into account. Moreover, compression can set the focus on specific image features, leading to fewer storage needs and a reduction in computing time with comparably small costs in terms of quality and accuracy. All in all, quantization and embedding into file formats do support CNNs to estimate the labels of images, by strengthening the features.https://ieeexplore.ieee.org/document/9969909/Deep learningimage compressionimage quantizationremote sensing imagescene classificationtransfer learning |
| spellingShingle | Gabriel Dax Srilakshmi Nagarajan Hao Li Martin Werner Compression Supports Spatial Deep Learning IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Deep learning image compression image quantization remote sensing image scene classification transfer learning |
| title | Compression Supports Spatial Deep Learning |
| title_full | Compression Supports Spatial Deep Learning |
| title_fullStr | Compression Supports Spatial Deep Learning |
| title_full_unstemmed | Compression Supports Spatial Deep Learning |
| title_short | Compression Supports Spatial Deep Learning |
| title_sort | compression supports spatial deep learning |
| topic | Deep learning image compression image quantization remote sensing image scene classification transfer learning |
| url | https://ieeexplore.ieee.org/document/9969909/ |
| work_keys_str_mv | AT gabrieldax compressionsupportsspatialdeeplearning AT srilakshminagarajan compressionsupportsspatialdeeplearning AT haoli compressionsupportsspatialdeeplearning AT martinwerner compressionsupportsspatialdeeplearning |