I have 2D thermal image data. How to extract the features such as standard deviation, mean, entropy, skewness, kurtosis, mean absolute deviation from thermal image data?
As your question as formulated is very open (it is not clear whether the question is more specifically about mathematical formulas, software solutions, data analysis methodology, or anything else...), I propose to answer you by addressing these three different aspects.
About mathematical formulas, refer to :
https://www.mdpi.com/2076-3417/10/18/6493/s1
GNU Octave, Descriptive Statistics - https://docs.octave.org/latest/Descriptive-Statistics.html
Armi et al., "TEXTURE IMAGE ANALYSIS AND TEXTURE CLASSIFICATION METHODS - A REVIEW", 2019 - https://arxiv.org/ftp/arxiv/papers/1904/1904.06554.pdf
About data analysis methodology, refer to:
Armi et al., "TEXTURE IMAGE ANALYSIS AND TEXTURE CLASSIFICATION METHODS - A REVIEW", 2019 - https://arxiv.org/ftp/arxiv/papers/1904/1904.06554.pdf
de Siqueira et al., "Multi-scale gray level co-occurrence matrices for texture description", 2013 - http://www.ssig.dcc.ufmg.br/wp-content/uploads/2014/09/2013-Multi-Scale-Gray-Level-Co-Occurrence-Matrices-for-Texture-Description.pdf
About software tools, refer to :
Image Processing Toolbox by MathWorks - https://uk.mathworks.com/help/images/index.html?s_tid=CRUX_lftnav
GNU Octave, Descriptive Statistics - https://docs.octave.org/latest/Descriptive-Statistics.html
I have the approximation and detailed coefficients obtained from the MATLAB software. Now I have to plot graphs between the STD vs Skew, STD vs Kurtosis, STD vs Entropy, and STD vs MAD at every level. How should I have to proceed?