Image Description Attained Through Remote Sensing and a Word-Embedding System

The main motto of this is used to showcase the effects of rapidly changing land which was ready to use on soil erosion, segmentation, and also used to change the detection to qualify the changing of land which was used on soil erosion risk. The mainstream was based on a statement perfect which was functional to the expected space and it was used to access the effect on soil destruction. Based on the previous research, the total soil erosion from towns is mostly expected to increase nowadays.so, resultant support earlier reports were increased the soil erosion risk in urban areas. This gives the interaction between the changes in the land which was used in corrosion possible. The soil destruction hazard was taken on land which can be qualified by the incorporation similarly existed in physical data in remote sensing, which was used to serve as a land management tool for a small change in erosion risk. This work provides an end-to-end picture description generating method using word embedding to classify and identify remote sensing photos. First, co-occurrence matrices and global vectors for word representation are used to handle category ambiguity in remote sensing photographs.