The study of the seafloor is important for living and non-living resource estimation along with the related processes identification. To understand the fine-scale seafloor processes, various methods such as application of acoustic remote sensing, seafloor photographic and geological samplings are well established. Among these, the high-frequency single beam echo-sounding system (SBES) and multi-beam echo-sounding system (MBES) became more familiar due to their rapid data acquisition advantages. These systems are extensively used to study the seafloor morphology etc. Seafloor acoustic backscatter information provides fine-scale seafloor roughness and associated sediment processes. The angular and normal incidence backscatter strength data can be utilized to estimate seafloor roughness parameters using physics based numerical inversion models. However, for such applications, the segmentation of the backscatter data is essential, especially before initiating any numerical based models to characterize the seafloor. Under such situations, the employment of the soft computational techniques e.g., artificial neural networks (ANNs) are found to be suitable for seafloor acoustic data segmentation and classifications. Seafloor studies are carried out at the National Institute of Oceanography, Goa during the last two decades employing similar techniques, and study results related to the seafloor classification and characterizations are documented in this research review.