The purpose of this study is to examine the applicability of CG images reproducing detailed urban greenery in the evaluation of river landscape. The acquisition of greenery information was based on drone technology and deep learning. We obtained information on the height and location of greenery. In addition, regarding physical indices and psychological evaluation, we examined the coincidence between CG images and photographs. As a result, we were able to obtain the data that could express the shape of individual trees and match photographs with a high degree of accuracy in the psychological evaluation items related to shape.