Detection of Forest Area Density using Satellite Images

The proposed work focuses on the utilization of satellite image processing techniques for the effective detection and monitoring of forest area density. Through a series of image processing steps, we analyze satellite images obtained from the designated dataset, emphasizing a region containing forests. The workflow includes key stages such as pre-processing, segmentation, feature extraction, and density prediction. The paper culminates in the evaluation of the Support Vector Regressor model’s performance, including metrics such as Mean Squared Error (MSE), Mean Absolute Error (MAE), and R 2 Error. The proposed method achieves a mean squared error (MSE) of 3.17, a mean absolute error (MAE) of 1.32, and an R² score of 0.0056 in a 60:40 train-test split. Also, the precision, recall, F1-score, and Intersection over Union (IoU) metrics were calculated, besides which strong F1-score of 0.89 and IoU of 0.76 were received, illustrating the viability of the segmentation and the feature extraction. The advised model is very accurate in forest density determination, which is very important in deforestation monitoring, biological diversity preservation, and carbon stock assessment, and also other issues. The main goal of the model is to increase the forest density measurement, and it in turn will enable the policy makers with the environment to manage the changes in the forest cover as required.