Multi-Scale Spectral Analysis Of Vegetation And Health Indices Using Dwt
Oladipo Samuel Adeyemi
Department of Electrical and Electronic Engineering, Federal University of Technology, Akure, Nigeria
Abstract
Socio-physics reveals that non-physical things of society can be understood and explained by the laws of physics. Health of people is one of the main parameters that decide the development of a country. Vegetables are the main and clean source of nutrition to nurture the human body and mind. Vegetables provide multivitamins and minerals that are essential for the health of human beings and have manifold health benefits. Wavelet transforms is a powerful and computational efficient tool which has wide applications in field of data analysis and signal processing. We have taken health and vegetable data of India from Jan. 2013 to Aug. 2021 as raw data. Wavelet transforms of this data is performed by software dyadwaves using Haar wavelet, decomposition level-5. In first level of decomposition, the data is decomposed into approximation and wavelet coefficients. The approximation represents average behaviour and detail represents differential behaviour of the data. Approximation is the slowest part of data which describes trend of the data. In each further level of decomposition, the approximation is further decomposed into next level approximation and detail. It has been found that wavelet analytical results are strongly consistent with statistical analytical results.