A novel denoised aerial image segmentation for disaster management based on deep learning

Abstract

Natural disasters such as Hurricanes, tsunamis, floods, and other newlinenatural catastrophes are examples of abrupt, catastrophic events that happen newlineall year long in many regions of the world. They danger both life and newlineproperty. For emergency management and recovery operations, the capacity newlineto map out damage distribution as soon as a natural catastrophe occurs is newlinecrucial. The impact of natural calamities is immeasurable, and it is often newlineimpossible to reverse the damage caused by them; however, humans can newlinechange the way they respond to them. Damage can be reduced with the help newlineof a good disaster management plan, preparation, and forecast. For optimal newlinedisaster management, consistent and quantitative damage assessment tools are newlineurgently needed to facilitate an efficient decision-making process. Identifying newlinethe severity of the damage and segmenting Natural Disaster images remain newlinechallenging issues in the field of Image Processing (IP) because there is no newlinewidely acknowledged way of doing so after decades of effort. Further newlineresearch is necessary before these tools may be used to make important newlineemergency response decisions, as the earlier research recognizes. newline

Description

Keywords

Citation

item.page.endorsement

item.page.review

item.page.supplemented

item.page.referenced