Method for estimating tree root diameter based on ground penetrating radar A-scan data

Abstract

The invention discloses a method for estimating the diameter of a tree root based on A-scan data of a ground penetrating radar, and belongs to the technical field of nondestructive testing of woods. According to the method, the sandy soil content S, the water content M and the tree root depth D in a soil environment are estimated through a trained neural network according to ground penetrating radar A-scan data, then S, M and D are fused with the A-scan data to obtain a new input vector, and the new input vector is input into a trained random forest network to obtain an estimated value of the tree root diameter to be estimated; the problem that the parameter space is too large and the solving time is long due to the fact that Hough transform needs to be carried out when the estimated value of the tree root diameter is solved through a hyperbola function in the prior art is solved, the hyperbola does not need to be segmented, errors caused by deviation between the fitted hyperbola and the actual value are avoided, and the estimation precision is further improved; according to the method, the maximum error of root depth estimation is + / -3cm, and the root radius estimation precision is + / -10mm.

Description

technical field [0001] The invention relates to a method for estimating tree root diameter based on ground-penetrating radar A-scan data, and belongs to the technical field of non-destructive detection of forest trees. Background technique [0002] GPR, as a well-known and convenient non-destructive testing technique, has been widely used in near-surface research to detect objects located within a few meters below the surface. In the field of forest nondestructive testing, ground penetrating radar has been successfully applied to tree root detection and root biomass estimation at home and abroad. [0003] The basic working principle of ground penetrating radar is that during the detection process along the direction perpendicular to the root of the tree, the radar signal transmitter continuously transmits electromagnetic wave signals (A-scan) to the ground, and the receiver receives multiple A-scan signals to form B-scan, which makes underground tubular targets appear as hy...

AI Technical Summary

Problems solved by technology

[0006] In order to solve the current inaccurate problem of estimating tree root diameter based on ground penetrating radar, the present invention provides a method for estimating tree root diameter based on ground penetrating radar A-scan data, and the method inputs ground penetrating radar A-scan data respectively The three trained neural networks with the same structure but with different weights between nodes in each layer can obtain the sand content S, water content M and root depth D of the soil environment where the tree root is to be estimated, and obtain the sand content S , moisture content M and tree root depth D, fuse it with the A-scan data to get a new input vector, and input the new input vector into the trained random forest network with 100 trees to get the tree root to be estimated Estimated value of diameter

Images