Hyperspectral image ground-object identification method based on spectral segment fusion combination and depth residual network
Source abstract
Abstract
An algorithm based on the spectral segment fusion combination and deep residual network is proposed to improve the recognition accuracy of the objects of interest in the WHU-Hi dataset, particularly for cruciferous plants. The accuracy of the objects of interest was effectively improved, as well as the recognition accuracy of other ground objects, and the time efficiency was improved as well. The optimal combination of spectral segments was determined, and spatial and spectral information was extracted from the deep residual network for ground object recognition research. Experimental results showed that the classification accuracy of the cruciferous plants of interest, namely, pak choi, Brassica chinensis, and small Brassica chinensis, increased from 81.36%, 84.2%, and 83.8% to 98.32%, 99.22%, and 98.35%, respectively. In addition, the accuracy of interested trees and grass also increased from 77.6% and 89.09% to 99.12% and 98.33%, respectively, and the overall accuracy, KAPPA, and average accuracy of the three datasets were all improved. The time efficiency was also improved by an order of magnitude.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Remote-Sensing Image Classification · Remote Sensing and Land Use · Advanced Image Fusion Techniques
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
College of Astronautics, Nanjing University of Aeronautics and Astronautics 1 , Nanjing, Jiangsu 210016, China
Key Laboratory of Space Photoelectric Detection and Perception, Nanjing University of Aeronautics and Astronautics 2 , Nanjing, Jiangsu 210016, China
National Innovation Institute of Defense Technology, Academy of Military Science, Beijing 3 , Beijing 650500, China
Funding
National Defense Science and Technology Foundation Strengthening Program · National Natural Science Foundation of China · Fundamental Research Funds for the Central Universities
References (24)
Show source-provided references
1. The automatic recognition and classification of ground objects in hyperspectral images based on the projection pursuit method
2. Improving spatial–spectral endmember extraction in the presence of anomalous ground objects10.1109/tgrs.2011.2163160 ↗
3. Robustness tests for object identification algorithms in hyperspectral imagery10.1117/12.683219 ↗
4. A dimension reduction-based method for classification of hyperspectral and LIDAR data10.5194/isprsarchives-xl-1-w5-1-2015 ↗
5. Hyperspectral imaging technology combined with deep learning for hybrid okra seed identification10.1016/j.biosystemseng.2021.09.010 ↗
6. Identification of plant leaf phosphorus content at different growth stages based on hyperspectral reflectance10.1186/s12870-020-02807-4 ↗
7. Deep learning earth observation classification using ImageNet pretrained networks10.1109/lgrs.2015.2499239 ↗
8. A color-texture-structure descriptor for high-resolution satellite image classification10.3390/rs8030259 ↗
9. Transferring pre-trained deep CNNs for remote scene classification with general features learned from linear PCA network10.3390/rs9030225 ↗
10. Multi-scale guided feature extraction and classification algorithm for hyperspectral images10.1038/s41598-021-97636-2 ↗
11. Deep pyramidal residual networks for spectral–spatial hyperspectral image classification10.1109/TGRS.2018.2860125 ↗
12. Deep learning ensemble for hyperspectral image classification10.1109/jstars.2019.2915259 ↗
13. Hyperspectral band selection for multispectral image classification with convolutional networks
14. Hyperspectral band selection via region-aware latent features fusion based clustering10.1016/j.inffus.2021.09.019 ↗
15. A multi-strategy integrated multi-objective artificial bee colony for unsupervised band selection of hyperspectral images10.1016/j.swevo.2020.100806 ↗
16. Hyperspectral band selection using attention-based convolutional neural networks10.1109/access.2020.2977454 ↗
17. Multiscale residual network with mixed depthwise convolution for hyperspectral image classification10.1109/tgrs.2020.3008286 ↗
18. Hyperspectral image classification based on hierarchical fusion of residual networks10.3390/rs13030335 ↗
19. WHU-hi: UAV-borne hyperspectral with high spatial resolution (H2) benchmark datasets for hyperspectral image classification
20. WHU-hi: UAV-borne hyperspectral with high spatial resolution (H2) benchmark datasets and classifier for precise crop identification based on deep convolutional neural network with CRF10.1016/j.rse.2020.112012 ↗
21. Spectral-spatial hybrid mechanism for feature detection using spectral correlation
22. A correlation based band selection approach for hyperspectral image classification
23. Libsvm: A library for support vector machines10.1145/1961189.1961199 ↗
24. A support vector conditional random fields classifier with a mahalanobis distance boundary constraint for high spatial resolution remote sensing imagery10.1109/jstars.2013.2290296 ↗