The specific features of a object can be extracted from its spectral curve through the wavelet transform at a variable scale. The Wavelet Feature Correlation Coefficient (WFCC) is a concept to describe the similarities between curves after wavelet transform. This technique can be used to spectroscopy and photoacoustic imaging.
Zhaohui Wang. “Unsupervised Wavelet-Feature Markov Clustering Algorithm for Remotely Sensed Images,” 2020 IEEE International Symp. on Signal Processing and Information Tech., 2020.
Zhaohui Wang, Pei-ling Zhou. “Greedy clustering algorithm and its application for the classification and compression of remotely sensed images,” Journal of University of Science and Technology of China, 33(1), 52-59 (2003).
Zhaohui Wang, Pei-ling Zhou. “Wavelet Features Extraction and Markov Clustering Algorithm for Remotely Sensed Images,” Journal of Data Acquisition & Processing, 18(1), 1-7 (2003).
Zhaohui Wang, Pei-ling Zhou. “Fast Clustering Based on Spectral Wavelet Features Extraction and Simulated Annealing Algorithm for Multispectral Images,” Journal of Image and Graphics, 7A(12), 1257-1262 (2002).
With the help of an improved K-means clustering algorithm to accelerate the clustering process, multi-level clustering with initial S+P (Sequential transform + Prediction) integer wavelet transformation can not only remove the spatial and structural redundancy, but also delete the residual-data redundancy, realizing the breakthrough of lossless compression for multi-spectral images.
Zhaohui Wang, “Entropy Analysis for Clustering Based Lossless Compression of Remotely Sensed Images,” 2021 IEEE Big Data Conference, 2021.
Zhaohui Wang, “Residual clustering based lossless compression for remotely sensed images,” 2018 IEEE International Symposium on Signal Processing and Information Technology, 2018, pp.536-539.
Zhaohui Wang. “Fast multi-level clustering lossless compression algorithm for remotely sensed images,” Journal of Image and Graphics, 8(7), 843-848 (2003).
Zhaohui Wang. “Fast clustering lossless compression algorithm for hyperspectral images,” Journal of Remote Sensing, 7(5), 400-406 (2003).
Figure 1. Spectrum of targets
Figure 2. Wavelet filter
Figure 3. Wavelet feature classification results