Environmental Exploration and Monitoring of Vegetation Cover using Deep Convolutional Neural Network in Gombe State

In the previous years, human assessments of
satellite data were possible because to the relatively
modest volume of images accessible; however, this is no
longer the case. Additionally, traditional software like
ARGIS, EDARS, ILWIS, and other time-consuming and
ineffective tools are heavily used by environmental
organizations in Nigeria, such as the National Centre for
Remote Sensing. With today's large number of data,
relevant information extraction from photos thus
becomes a challenge. Differentiating between classes
with comparable visual qualities is another problem, as
seen when attempting to categorize a green pixel as
grass, shrubs, or a tree. However, as seen in other
computer vision fields, machine learning approaches
have shown to be a strong answer in this case. We
proposed the introduction of a novel three-dimensional
based architecture that is specialized to multispectral
pictures and addresses the majority of the challenging
features of Deep Learning for Remote Sensing as a
solution to the issues raised. This work's major goal is to
create a deeper CNN architecture (U-Net) model that is
efficient for semantically segmenting remote sensing
imagery with additional multi-spectral feature classes.
In comparison to traditional remote sensing software,
our innovative 3D CNN architecture can analyze the
spatial and spectral components simultaneously with
true 3D convolutions. our results in better, faster, and
more efficient segmentation.