Abstract
Hyperspectral and multispectral information processing systems and technologies have demonstrated their value in enhancing agricultural productivity and practices by providing farmers and crop managers with valuable information on factors influencing crop condition and growth. These technologies play a crucial role in various agricultural applications, such as crop management, crop yield forecasting, crop disease detection, and monitoring soil, water, and land usage. To enhance the process of agriculture through effective crop management and assistance to farmer, using an advanced image transformation techniques the study delves into the exploration of techniques for harmonizing cross-view image transformation, with a focus on integrating both local and global insights. Further the study proposes the application of the 2D-DWT (Two-Dimensional Discrete Wavelet Transform) technique for image data preprocessing and the Extreme Learning Machine (ELM) algorithm for image classification in the context of hyperspectral images. Hyperspectral information sensing allows for the coverage of the electromagnetic spectrum in a single acquisition with several hundred spectral bands, resulting in a data cube containing significant spectral and spatial information ELM is particularly effective for classification problems because of its quick training time and capacity to operate with hidden nodes whose parameters are randomly assigned rather than modified iteratively. This comprehensive review aims to provide valuable insights and a critical analysis of the suggested method, shedding light on its potential contributions to the advancement of image transformation techniques for agricultural development.
References
X. Liu, B. Lou, C. Jiang, T. Li, S. B. Hussain, G., H. Zhai, Y. Shen, and B. Shen, "Large-scale crop mapping using Google Earth engine with multisource remote sensing images," IEEE J. Sel. Subjects Appl. Earth Watch. Vol., Remote Sens. 13, no. 13, 2020, pages 414–427.
S. D. Harle, T. K. Dao, I. Andonovic, J. Ren, S., H. Vu, C. Tachtatzis, P. Murray, and D. Fabiyi. Marshall, "Using RGB and hyperspectral images for variabilty classification of rice seeds," IEEE Access, vol. 8, 2020, pp. 22493–22505.
Ze Z. Li, T.; Chen, G. Zhang, "A hybrid CNN-transformer approach for multitemporal multisensor image crop classification," IEEE J. Sel. Subjects Appl. Earth Watch. Vol., Remote Sens. 13, no. 13, 2020, pp. 847–859.
Xin, Zhou, Sun Jun, Tian Yan, Chen Quansheng, Wu Xiaohong, and Hang Yingying. "A deep learning based regression method on hyperspectral data for rapid prediction of cadmium residue in lettuce leaves." Chemometrics and intelligent laboratory systems 200 (2020): 103996.
Han, Zhongzhi, and Jiyue Gao. "Pixel-level aflatoxin detecting based on deep learning and hyperspectral imaging." Computers and Electronics in Agriculture 164 (2019): 104888.
S. Workman, R. Souvenir, and N. Jacobs, ‘‘Wide-area image geo localization with aerial reference imagery,’’ in Proc. IEEE Int. Conf. Compute. Vis. (ICCV), Dec. 2023, pp.3961–3969.
Xin, Zhou, Sun Jun, Tian Yan, Chen Quansheng, Wu Xiaohong, and Hang Yingying. "A deep learning based regression method on hyperspectral data for rapid prediction of cadmium residue in lettuce leaves." Chemometrics and intelligent laboratory systems 200 (2020): 103996.
Gao, Jiyue, Jiangong Ni, Dawei Wang, Limiao Deng, Juan Li, and Zhongzhi Han. "Pixel-level aflatoxin detecting in maize based on feature selection and hyperspectral imaging." Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 234 (2019): 118269.
S. D. Fabiyi, H. Vu, C. Tachtatzis, P. Murray, D. Harle, T. K. Dao, I. Andonovic, J. Ren, and S. Marshall, ‘‘Varietal classification of rice seeds using RGB and hyperspectral images,’’ IEEE Access, vol. 8, pp. 22493–22505, 2020.
Z. Tang, S. Li, L. Luo, M. Fu, H. Peng, and Q. Zhou, ‘‘Image super resolution via simplified dense network with non-degenerate layers,’’ IEEE, Access, vol. 7, pp. 24775–24787, 2021.