Identifying the Changes of Mine Water Bodies from Landsat 8 OLI Images in Automated Manner: A Case Study in Jharia, India

Identifying and monitoring water bodies have been active research areas because of their multi-fold effects on the environment and society. Water bodies are detected in the literature by several indexes such as normalized difference water index (NDWI), modified normalized water body index (MNDWI), automated water extraction index (AWEI), and others from multi-spectral satellite images. Identifying and separating different types of water bodies using such indexes have been found complex due to their multiple homogeneous features. The high mineral abundance of the surrounding regions can be the distinguishing attribute of a mine water body. This idea has been used in the past to separate mine water bodies from other kinds of water bodies. However, a certain limitation has been reported as other high mineral abundance regions can be present close to a water body, which are not associated with mining. Further, these mine water bodies change frequently due to their uses. Monitoring such water bodies has several applications in the mining industry, water pollution, and health. Hence, automated detection of the changes in mine water bodies needs more extensive attention. In this work, this research gap has been addressed. First, mine water bodies are separated. Further, their translation, rotation, and shearing changes are computed using the coherent point drift technique.