Aidash, Inc.

Nitin Das

Shantanu Rajora

An example method includes receiving georeferenced satellite images of a geographic region that includes electrical assets. Segmentation maps are generated by providing the georeferenced satellite images to fully convolutional networks to classify pixels of the georeferenced satellite images as either trees or non-trees. Rasters of the geographic region are generated based on the segmentation maps and vectors are generated based on the rasters. A vector includes one or more polygons, a polygon representing a tree and having a set of coordinates defining the polygon. Canopy height models are generated based on received digital surface models. The canopy height models include heights of trees in the geographic region. Heights of trees are associated with polygons. A height of a polygon is compared to a distance between the polygon and an electrical asset to identify a tree as a potential hazard. Notifications thereof are provided.

Growing community of inventors

Haryana, India

Nitin Das