A two-stage framework of target detection in high-resolution hyperspectral images
- 91 Downloads
Hyperspectral images (HSIs) have been widely used in various areas, especially in the remote sensing field. Compared with traditional remote sensing HSI, the high-resolution HSI (HRHSI) owns high resolution in both spatial and spectral domains. Since most existing methods with automatic target detection are merely suitable for traditional remote sensing HSI with low resolution, they are unable to be applied directly on HRHSI. Therefore, this paper proposes a novel automatic target detection framework for HRHSI. The framework includes two stages: in Stage one, a new spectral curve spatial pyramid matching model called SC-SPM is established to search the potential target regions; in Stage two, a novel bounding box dilation is proposed to detect the target region precisely. Extensive experiments are carried out in different scenes, and the results demonstrate the superiority of proposed framework compared to those traditional target detection methods.
KeywordsHigh-resolution hyperspectral image Target detection SPM Dilation
This work was partially supported by National Natural Science Foundation of China (No. 61771319), Natural Science Foundation of Guangdong Province (No. 2017A030313343), Shenzhen Science and Technology Project (No. JCYJ20180507182259896).
- 4.Ziemann, A., Theiler, J.: Variable target detection using Simplex ACE. In: IEEE Geoscience and Remote Sensing Symposium, Fort Worth, TX, USA, 23–28 July, 2017; pp. 1347–1350Google Scholar
- 13.Wang, H., Celik, T.: Sparse representation-based hyperspectral image classification. SIViP 11, 1–9 (2018)Google Scholar
- 14.Li, Y., Shi, W., Liu, A.: A Harris corner detection algorithm for multispectral images based on the correlation. In 6th International conference on wireless, mobile and multi-media, Beijing, China, 20–23 November, 2015, pp. 161–165Google Scholar
- 15.Li, Y., Shi, W.: A new framework of hyperspectral image classification based on spatial spectral interest point. In: IEEE 13th International Conference on Signal Processing, Chengdu, China, 6–10 November, 2016, pp. 733–736Google Scholar
- 18.Lazebnik, S., Schmid, C., Ponce, J.: Beyond bags of features: spatial pyramid matching for recognizing natural scene categories. In: 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, New York, USA, 17–22 June, 2006, pp. 2169–2178Google Scholar
- 19.Kim, J., Liu, C., Sha, F., et al.: Deformable spatial pyramid matching for fast dense correspondences. In: Proceedings of the IEEE Conference on Computer vision and Pattern Recognition, Portland, OR, USA, 23–28 June, 2013, pp. 2307–2314Google Scholar
- 22.Huang, C.T., Lin, H.H., Su, C.H.: Polygonal object dilation and packing in vector space for die-cutting machine tools. In: 2017 IEEE International Conference on Applied System Innovation, Sapporo, Japan, 13–17 May, 2017, pp. 1793–1796Google Scholar
- 23.Kudo, Y., Aoki, Y.: Dilated convolutions for image classification and object localization. In: 2017 IEEE Fifteenth IAPR International Conference on Machine Vision Applications, Nagoya, Japan, 8–12 May, 2017, pp. 452–455Google Scholar
- 28.Redmon, J., Farhadi, A.: YOLO9000: better, faster, stronger. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA, 21–26 July, 2017, pp. 6517–6525Google Scholar