Abstract
Remote sensing is a vital technique for detecting and monitoring the physical characteristics of an area from a distance: typically, from an aircraft or satellite. Accurate object detection in remote sensing images is crucial since security, transportation, environmental monitoring, and rescue applications in the military and civilian sectors demand comprehensive, precise analysis and utilization of these images. Despite recent advances, small object detection in satellite images is challenging due to the lack of information raised from the area covered by small objects in the image. To address the problem of small object detection, this paper discusses the importance of feature space analysis. The need for distinctive feature space is equally important along with the choice of the model architecture. Focusing on feature space analysis, deep learning is used for the extraction of the granularity of data. Further, spectral context is combined with spatial features to reinforce feature information and thus, enhance the detection. Our strategy involves comparing two units: the Fourier transform unit and the Wavelet packet transform unit to extract spectral features. These units are placed in the EfficientDet to enhance the feature space and assist small object detection by providing context. Three models were trained and tested on DIOR dataset, which is a publicly available dataset. The intuition of adding the transform units to the existing model has shown a remarkable improvement of 20% in the mAP. With limited changes in the number of parameters and without much hyper parameter tuning, the mAP shows improvement. Initially an experiment was conducted on a small custom dataset to test the hypothesis of the impact of feature space in object detection and the models containing the transform units showed an improvement of 4% in the mAP.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Girshick, R.R., Donahue, J., Darrell, T., Malik, J.: Rich feature hierarchies for accurate object detection and semantic segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 580–587 (2014)
Girshick, R.: Fast R-CNN. In: Proceedings of the IEEE international conference on computer vision, 2015, pp. 1440–1448. [3] S. Ren, K. He, R. Girshick, and J. Sun, “Faster rcnn: Towards real-time object detection with region proposal networks,” Advances in neural information processing systems, vol. 28, 2015
Gotovac, S., Papić, V., Marusic, Z.: Analysis of saliency object detection algorithms for search and rescue operations (2016). https://doi.org/10.1109/SOFTCOM.2016.7772118
Liu, W., et al.: SSD: single shot multibox detector. In: European conference on computer vision. Springer, 2016, pp. 21–37. [5] J. Redmon, S. Divvala, R
Girshick, Farhadi, A.: Y ou only look once: unified, real-time object detection. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 779–788 (2016)
Lin, T.-Y., Dollár, P., Girshick, R., He, K., Hariharan, B., Belongie, S.: Feature pyramid networks for object detection. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2117–2125 (2017)
Li, Y., Huang, Q., Pei, X., Chen, Y., Jiao, L., Shang, R.: Cross-layer attention network for small object detection in remote sensing imagery. IEEE J. Sel. Top. Appl. Earth Observ. Remote Sens. 14, 2148–2161 (2021). https://doi.org/10.1109/JSTARS.2020.3046482
Fang, X., et al.: Small object detection in remote sensing images based on super-resolution. Pattern Recogn. Lett. 153, 107–112 (2022). ISSN 0167-8655. https://doi.org/10.1016/j.patrec.2021.11.027
Chen, H.M., Luo, L., Chen, X., Wu, K.: Effective Feature Fusion Network in BIFPN for small object detection. In: 2021 IEEE International Conference on Image Processing (ICIP), 2021, pp. 699–703 (2021). https://doi.org/10.1109/ICIP42928.2021.9506347
Qingyun, F., Lin, Z., Zhaokui, W.: An efficient feature pyramid network for object detection in remote sensing imagery. IEEE Access 8, 93 058–93 068 (2020)
Deng, C., Wang, M., Liu, L., Liu, Y., Jiang, Y.: Extended feature pyramid network for small object detection. IEEE Trans. Multimedia 24, 1968–1979 (2022). https://doi.org/10.1109/TMM.2021.3074273
Tan, M., Pang, R., Le, Q.V.: Efficientdet: scalable and efficient object detection. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 10781–10790 (2020)
Xu, K., Qin, M., Sun, F., Wang, Y., Chen, Y.-K., Ren, F.: Learning in the Frequency Domain, Computer Vision and Pattern Recognition (cs.CV). Computer and information sciences, arXiv (2020). https://doi.org/10.48550/arxiv.2002.12416
Chi, L., Jiang, B., Mu, Y.: Fast fourier convolution. Adv. Neural. Inf. Process. Syst. 33, 4479–4488 (2020)
Suvorov, R., et al.: Resolution robust large mask inpainting with Fourier convolutions. In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pp. 2149–2159 (2022)
Katznelson, Y.: An introduction to Harmonic Analysis. Cambridge University Press, Cambridge (2004)
Bruna, J., Mallat, S.: Invariant scattering convolution networks, computer vision and pattern recognition (cs.CV), Comput. Inf. Sci. arXiv (2012). https://doi.org/10.48550/arxiv.1203.1513
Wolter, M., Blanke, F., Heese, R., Garcke, J.: Wavelet-packets for deepfake image analysis and detection. In: Computer Vision and Pattern Recognition, 17 June 2021
Fortuna-Cervantes, J.M., Ramírez-Torres, M.T., Martínez-Carranza, J., et al.: Object detection in aerial navigation using wavelet transform and convolutional neural networks: a first approach. Program Comput. Soft. 46, 536–547 (2020). https://doi.org/10.1134/S0361768820080113
Yu, B., Lu, H., Tao, H.: A deep learning method with wavelet packet transform for infrared target recognition. In: 2021 IEEE International Conference on Signal Processing, Communications and Computing (ICSPCC), pp. 1–5 (2021). https://doi.org/10.1109/ICSPCC52875.2021.9564410
Li, K., Wan, G., Cheng, G., Meng, L., Han, J.: Object detection in optical remote sensing images: A survey and a new benchmark. ISPRS Journal of Photogrammetry and Remote Sensing, vol. 159, pp. 296–307 (2020). ISSN 0924-2716, https://doi.org/10.1016/j.isprsjprs.2019.11.023
Arya, R., Singh, N., Agrawal, R.K.: A novel hybrid approach for salient object detection using local and global saliency in frequency domain. Multimed. Tools Appl. 75(14), 8267–8287 (2015). https://doi.org/10.1007/s11042-015-2750-y
Zhang, L.-G., Wang, L., Jin, M., Geng, X.-S., Shen, Q.: Small object detection in remote sensing images based on attention mechanism and multi-scale feature fusion. Int. J. Remote Sens. 43(9), 3280–3297 (2022). https://doi.org/10.1080/01431161.2022.2089539
Wang, G., et al.: FSoD-Net: full-scale object detection from optical remote sensing imagery. IEEE Trans. Geosci. Remote Sens. 60, 1–18 (2022). Art no. 5602918 https://doi.org/10.1109/TGRS.2021.3064599
Sun, X., Deng, H., Liu, G., Deng, X.: Combination of spatial and frequency domains for floating object detection on complex water surfaces. Appl. Sci. 9, 5220 (2019). https://doi.org/10.3390/app9235220
Li, Z., et al.: Deep learning-based object detection techniques for remote sensing images: a survey. Remote Sens. 14, 2385 (2022). https://doi.org/10.3390/rs14102385
Tang, Y., Zhang, X., Hu, X., Wang, S., Wang, H.: Facial expression recognition using frequency neural network. IEEE Trans. Image Process. 30, 444–457 (2021). https://doi.org/10.1109/TIP.2020.3037467
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 IFIP International Federation for Information Processing
About this paper
Cite this paper
Giridharan, U., Ramprasad, N., Roy, S., Omkar, S.N. (2023). Impact of Spectral Domain Features for Small Object Detection in Remote Sensing. In: Mercier-Laurent, E., Fernando, X., Chandrabose, A. (eds) Computer, Communication, and Signal Processing. AI, Knowledge Engineering and IoT for Smart Systems. ICCCSP 2023. IFIP Advances in Information and Communication Technology, vol 670. Springer, Cham. https://doi.org/10.1007/978-3-031-39811-7_15
Download citation
DOI: https://doi.org/10.1007/978-3-031-39811-7_15
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-39810-0
Online ISBN: 978-3-031-39811-7
eBook Packages: Computer ScienceComputer Science (R0)