An efficient No-Line-of-Sight learning approach for Prediction of Static Victim Detection Using Genetic Algorithm and Kth Nearest Neighbor Data Classification

In emergencies, among the most effective methods for saving lives is spotting and locating victims. While time is against them, quick actions are essential for victims. The first 48 hours, commonly referred to as "the golden hours," are vital for rescue efforts. The likelihood of survivability dramatically decreases beyond this point. Hence, the core of these systems is the constant tracking of the exact location, which is a difficult task in general but is particularly difficult in no-line-of-sight (NLOS) situations. So, in these urgent situations, a quick and effective deployment is crucial. Systems for NLOS person detection have been proposed using a wide range of machine learning (ML) approaches and algorithms. However, because of the dimensionality of the data, difficulties with accuracy, dependability, and scalability still exist. In this study, a genetic algorithm feature selection approach is applied to the complex input feature space in order to reduce the number of features while maintaining the data. The best way to extract attributes from data is to reduce it to latent components. The aim is to reduce prosperity by utilizing all dimensionality reduction approaches available for NLOS human detection categorization. Additionally, to improve the accuracy and scalability of the victim's detection analysis, the classifier applies Kth Nearest neighbor (KNN) to the smaller dataset. The classification accuracy is 86.3% which is used to compare the model's performances to other approaches that have been proposed in the literature. The obtained experimental results show promise for relevant data analysis classification and feature selection and is more effective for the prediction of human targets behind walls. This shows that the strategy can complement existing machine-learning techniques. Also, this work may helpful with human identification during search and rescue efforts after a disaster.