Answers are grounded in this paper and cite their sources.
Subpixel mapping techniques aim to obtain the spatial location and distribution of subpixels by transforming the information coming from a set of input abundance maps into a classification result with higher spatial resolution. However, traditional subpixel mapping algorithms generally ignore the possible errors that are due to abundance estimation inaccuracies by spectral unmixing techniques. In this paper, we propose a new genetic algorithm-based subpixel mapping technique that solves the subpixel mapping problem by correcting the potential errors in the estimated abundance fractions used as input to the subpixel mapping process. The proposed algorithm has been compared with other two genetic subpixel mapping methods, using both synthetic and real hyperspectral images. Our experimental results demonstrate that the proposed approach outperforms traditional subpixel mapping algorithms, thus providing an effective option to improve the accuracy of subpixel mapping for remotely sensed hyperspectral images.
No takes yet. Share an insight, caveat, or question.
Tong et al. (2016) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: