Automated processing of unmanned aerial vehicles images based on conceptual modeling of objective tasks. (May 2020)

• Record Type: Journal Article
• Authors: Vasendina, I; Shoshina, K; Guriev, A; Aleshko, R
• Is Part Of: Journal of physics. Volume 1553(2020)
• Publication Date: 2020-05
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/1553/1/012019
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: In this article, the authors proposed a methodology for processing images of unmanned aerial vehicles, based on the conceptual modeling of objective tasks, which allows to increase the accuracy of decoding of forest resources due to the unified description of the conceptual model of the process. The approach is based on a hierarchical structural representation of the process of decrypting the image. The result of the application of the method is the determination of individual contours of tree crowns. The developed method involves the use of the watershed many times for each contour, which made it possible to determine the contours of each tree and at the same time not to miss trees with small crowns. Using the error matrix and the formula for calculating the Kappa Cohen index, the reliability of thematic decryption was evaluated. The developed methodology allowed to improve the identification of objects by 18.6%.