Paper Key : IRJ************749
Author: Prof.nibe A.a.
Date Published: 09 Apr 2024
Abstract
The Ecological diversity is extremely dependent on the forest cover. The plants are one of the most essential sources that provides abundant nutrition and nourishment. The trees are considered as the carbon sinks of the world as they absorb atmospheric carbon and cool down the environment. But due to rapid urbanization and development, there has been increased incidences of deforestation that have been observed across the world. The loss of forest cover drastically harms the environment and further degrades the living conditions. The cutting down of forests is considered illegal in most countries but the process of monitoring of the deforestation activities is very difficult. Therefore, this research paper defines an effective framework for the detection of deforestation through the use of image processing approaches. The proposed approach utilizes the deforestation image dataset for image normalization and training through the Channel Boosted Convolutional Neural Networks and classification through Decision making. The experimental analysis of the approach has achieved successful results.