|
|
Learnable three-dimensional Gabor convolutional network with global affinity attention for hyperspectral image classification |
Hai-Zhu Pan(潘海珠)1,†, Mo-Qi Liu(刘沫岐)1, Hai-Miao Ge(葛海淼)1, and Qi Yuan(袁琪)2 |
1 College of Computer and Control Engineering, Qiqihar University, Qiqihar 161000, China; 2 College of Telecommunication and Electronic Engineering, Qiqihar University, Qiqihar 161000, China |
|
|
Abstract Benefiting from the development of hyperspectral imaging technology, hyperspectral image (HSI) classification has become a valuable direction in remote sensing image processing. Recently, researchers have found a connection between convolutional neural networks (CNNs) and Gabor filters. Therefore, some Gabor-based CNN methods have been proposed for HSI classification. However, most Gabor-based CNN methods still manually generate Gabor filters whose parameters are empirically set and remain unchanged during the CNN learning process. Moreover, these methods require patch cubes as network inputs. Such patch cubes may contain interference pixels, which will negatively affect the classification results. To address these problems, in this paper, we propose a learnable three-dimensional (3D) Gabor convolutional network with global affinity attention for HSI classification. More precisely, the learnable 3D Gabor convolution kernel is constructed by the 3D Gabor filter, which can be learned and updated during the training process. Furthermore, spatial and spectral global affinity attention modules are introduced to capture more discriminative features between spatial locations and spectral bands in the patch cube, thus alleviating the interfering pixels problem. Experimental results on three well-known HSI datasets (including two natural crop scenarios and one urban scenario) have demonstrated that the proposed network can achieve powerful classification performance and outperforms widely used machine-learning-based and deep-learning-based methods.
|
Received: 11 April 2022
Revised: 26 August 2022
Accepted manuscript online: 26 August 2022
|
PACS:
|
07.05.Mh
|
(Neural networks, fuzzy logic, artificial intelligence)
|
|
93.85.Pq
|
(Remote sensing in exploration geophysics)
|
|
93.90.+y
|
(Other topics in geophysical observations, instrumentation, and techniques)
|
|
95.75.Mn
|
(Image processing (including source extraction))
|
|
Fund: Project supported by the Fundamental Research Funds in Heilongjiang Provincial Universities (Grant No. 145109218) and the Natural Science Foundation of Heilongjiang Province of China (Grant No. LH2020F050). |
Corresponding Authors:
Hai-Zhu Pan
E-mail: panhaizhu@qqhru.edu.cn
|
Cite this article:
Hai-Zhu Pan(潘海珠), Mo-Qi Liu(刘沫岐), Hai-Miao Ge(葛海淼), and Qi Yuan(袁琪) Learnable three-dimensional Gabor convolutional network with global affinity attention for hyperspectral image classification 2022 Chin. Phys. B 31 120701
|
[1] Bishop C A, Liu J G and Mason P J 2011 Int. J. Remote Sens. 32 2409 [2] Stuart M B, McGonigle A J S and Willmott J R 2019 Sensors 19 3071 [3] Roessner S, Segl K, Heiden U and Kaufmann H 2001 IEEE Trans. Geosci. Remote Sens. 39 1525 [4] Audebert N, Saux B Le and Lefévre S 2019 IEEE Trans. Geosci. Remote Sens. 7 159 [5] Li S T, Song W W, Fang L Y, Chen Y S, Ghamisi P and Benediktsson J A 2019 IEEE Trans. Geosci. Remote Sens. 57 6690 [6] Duan Y, Huang H and Wang T 2021 IEEE Trans. Geosci. Remote Sens. 60 1 [7] Luo F L, Zou Z H, Liu J M and Lin Z P 2021 IEEE Trans. Geosci. Remote Sens. 60 1 [8] Kuo B C, Ho H H, Li C H, Hung C C and Taur J S 2013 IEEE J. Sel. Top Appl. Earth Obs. Remote Sens. 7 317 [9] Ma L, Crawford M M and Tian J W 2010 IEEE Trans. Geosci. Remote Sens. 48 4099 [10] Zhang Y Q, Cao G, Li X S, Wang B S and Fu P 2019 Remote Sens. 11 2974 [11] Gu Y F, Liu T Z, Jia X P, Benediktsson J A and Chanussot J 2016 IEEE Trans. Geosci. Remote Sens. 54 3235 [12] Wang Q, He X and Li X L 2018 IEEE Trans. Geosci. Remote Sens. 57 911 [13] Rajadell O, Garcia-Sevilla P and Pla F 2012 IEEE Geosci. Remote Sens. Lett. 10 860 [14] Li W and Du Q 2014 IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 7 1012 [15] Chen C, Li W, Su H J and Liu K 2014 Remote Sens. 6 5795 [16] Shen L L and Jia S 2011 IEEE Trans. Geosci. Remote Sens. 49 5039 [17] Jia S, Lin Z J, Deng B, Zhu J S and Li Q Q 2019 IEEE Trans. Neural. Netw. Learn Syst. 31 1638 [18] Hong D F, Yokoya N, Chanussot J and Zhu X X 2018 IEEE Trans. Image Process. 28 1923 [19] Chen Y S, Lin Z H, Zhao X, Wang G and Gu Y F 2014 IEEE J. Sel. Top Appl. Earth Obs. Remote Sens. 7 2094 [20] Hu W, Huang Y Y, Wei L, Zhang F and Li H C 2015 J. Sensors 2015 12 [21] Yu S Q, Jia S and Xu C Y 2017 Neurocomputing 219 88 [22] Paoletti M E, Haut J M, Plaza J and Plaza A 2018 ISPRS J. Photogramm. Remote Sens. 145 120 [23] Jia S, Lin Z J, Xu M, Huang Q, Zhou J, Jia X P and Li Q Q 2020 IEEE Trans. Geosci. Remote Sens. 59 4150 [24] Ghaderizadeh S, Abbasi-Moghadam D, Sharifi A, Zhao N and Tariq A 2021 IEEE J. Sel. Top Appl. Earth Obs. Remote Sens. 14 7570 [25] Ge Z X, Cao G, Li X S and Fu P 2020 IEEE J. Sel. Top Appl. Earth Obs. Remote Sens. 13 5776 [26] Bhatti U A, Yu Z Y, Chanussot J, Zeeshan Z, Yuan L W, Luo W, Nawaz S A, Bhatti M A, Ain Q U and Mehmood A 2021 IEEE Trans. Geosci. Remote Sens. 60 1 [27] Chen Y S, Zhu L, Ghamisi P, Jia X P, Li G Y and Tang L 2017 IEEE Geosci. Remote Sens. Lett. 14 2355 [28] Krizhevsky A, Sutskever I and Hinton G E 2012 Adv. Neural Inf. Process. Syst. 60 84 [29] Jiang C Z and Su J B 2018 Proceedings of 25th IEEE International Conference on Image Processing, October 7-10, 2018, Megaron Athens, Greece, p. 3408 [30] Luan S Z, Chen C, Zhang B C, Han J G and Liu J Z 2018 IEEE Trans. Image Process. 27 4357 [31] Jia S, Liao J H, Xu M, Li Y, Zhu J S, Sun W W, Jia X P and Li Q Q 2022 IEEE Trans. Geosci. Remote Sens. 60 1 [32] Sun H, Zheng X T, Lu X Q and Wu S Y 2019 IEEE Trans. Geosci. Remote Sens. 58 3232 [33] Mou L C and Zhu X X 2019 IEEE Trans. Geosci. Remote Sens. 58 110 [34] Fang B, Li Y, Zhang H K and Chan J C W 2019 Remote Sens. 11 159 [35] Hu J, Shen L, Albanie S, Sun G and Wu E H 2018 Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, June 19-21, 2018, Salt Lake City, Utah, USA, p. 7132 [36] Woo S, Park J, Lee J Y and Kweon I S 2018 Proceedings of the European Conference on Computer Vision, September 8-14, 2018, Munich, Germany, p. 3 [37] Ma W P, Yang Q F, Wu Y, Zhao W and Zhang X R 2019 Remote Sens. 11 1307 [38] Fu J, Liu J, Tian H J, Li Y, Bao Y J, Fang Z W and Lu H Q 2019 Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, June 16-20, 2019, Los Angeles CA, USA, p. 3146 [39] Li R, Zheng S Y, Duan C X, Yang Y and Wang X Q 2020 Remote Sens. 12 582 [40] Hubel D H and Wiesel T N 1959 J. Physiol. 148 574 [41] Yun T and Guan L 2013 Pattern Recognit. 46 529 [42] Jain A K and Farrokhnia F 1991 Pattern Recognit. 24 1167 [43] Wu S X, Chen Y C, Li X, Wu A C, You J J and Zheng W S 2016 IEEE Winter Conference on Applications of Computer Vision, March 7-10, 2016, Lake Placid, NY, USA, p. 1 [44] Zhu Q Q, Zhong Y F, Zhang L P and Li D R 2018 IEEE Trans. Geosci. Remote Sens. 56 6180 [45] Liang J, Zhou J, Tong L, Bai X and Wang B 2018 Pattern Recognit. 76 476 [46] Zhao C R, Lv X B, Zhang Z, Zuo W M, Wu J and Miao D Q 2020 IEEE Trans. Multimedia 22 3180 [47] Zhang Z Z, Lan C L, Zeng W J, Jin X and Chen Z B 2020 Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, June 14-19, 2020, Seattle, WA, USA, p. 3186 [48] Ge H M, Pan H Z, Wang L G, Liu M Q and Li C 2021 J. Appl. Remote. Sens. 15 042604 [49] Chen Y S, Jiang H L, Li C Y, Jia X P and Ghamisi P 2016 IEEE Trans. Geosci. Remote Sens. 54 6232 |
No Suggested Reading articles found! |
|
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
Altmetric
|
blogs
Facebook pages
Wikipedia page
Google+ users
|
Online attention
Altmetric calculates a score based on the online attention an article receives. Each coloured thread in the circle represents a different type of online attention. The number in the centre is the Altmetric score. Social media and mainstream news media are the main sources that calculate the score. Reference managers such as Mendeley are also tracked but do not contribute to the score. Older articles often score higher because they have had more time to get noticed. To account for this, Altmetric has included the context data for other articles of a similar age.
View more on Altmetrics
|
|
|