Detection and Classification of Floating and Submerged Polluted Targets at Turbid Water Using Remote Sensing Hyperspectral Imaging

Hyperspectral imaging (HI) is gaining prominence for its rapid and precise remote sensing in aquatic environments to identify and categorize a number of targets. The foundation for a physical laboratory setup for images obtained with a hyperspectral imager has been laid out in this study. Utilizing hyperspectral imagery and an accompanying image processing algorithm, a diffuse reflectance spectral signature for the target theatre of operations (polymer, wood, and metal with water) was developed as part of the suggested testing setup to differentiate between them. The second experiment is a case study in which the classified samples and their recorded spectral datasets are submerged in a muddy water vessel in order to validate our spectral data results. This information might then be used to cluster different materials using the imaging approach we’ve suggested. So, a high-resolution hyperspectral image dataset of a specific mixture of these materials and sand in a drive to strengthen the scientific underpinnings of diffuse reflectance-based floating and submerged material identification in an intertidal zone or shallow water is presented. Our objective was to build a high-resolution library of spectral patterns for these materials. Our research bolsters ongoing attempts to estimate the probability that these compounds would turn up in different murky marine habitats, which could affect applications for aqueous pollution detection. Aerial HI-remote sensing systems can be built using the knowledge in this research for trustworthy marine polymers identifications.