| Tree root plays a very important role in the ecosystem, however, as the root is generally grown below ground, with the result that the research breadth of the tree root is far behind the aboveground of tree. The traditional research methods for tree root exists many deficiencies. A detection method with lossless characteristics needs to be found urgently. Based on the non-destructive detection, this paper proposes a research method using ground penetrating radar to detect tree root. The Ground penetrating radar is used to detect tree the root, simple, easy, non-destructive, we can long-term, repeated detect the tree root. In the research process this paper received the strong support of national natural science fund project (31270757/C201337) and the Guangzhou Institute of Landscape Gardening. The contents of this paper are as follows:(1) The tree root characteristics are analyzed briefly. The radar data of tree root is analyzed. Clutters in the radar image were classified. Each type of the clutter characteristics are analyzed in this paper, presenting a different type of Clutter suppression method for different types of clutter. As the more targeted, the suppression effect is the better.(2) In order to detect the tree root target, analyzes the radar images tree root. The hyperbolic echo feature is one of the most important elements for root recognition in ground penetrating radar (GPR) image. By intensive analysis of the hyperbola detected accurately from tree root image by GPR, an improved method was presented to detect the target curve. The new method includes two aspects of content:According to the particularity of GPR image’s wavelet spectrum information, an extraction method of region of interest (ROI) was proposed based on GPR gradient magnitude. Furthermore, in order to make the hyperbolic characteristics in the gradient magnitude diagram more outstanding, the way of differential value was optimized by gradient magnitude method to achieve the rapid extraction of hyperbolic. Hough transform was conducted to detect the hyperbola in ROI. By using location information and amplitude information comprehensively voting, the accuracy of original. Hough transform was improved, and the background clutters and false targets can be removed effectively. Noise immunity of the method proposed was strong. In addition, the stability of the method was high. Thereby the accurate extraction of the target hyperbola is realized.(3) GPR wave velocity is an important parameter, this paper describes simply three wave velocity estimation methods commonly used, proposed based on the characteristics of the hyperbolic tree root Wave velocity estimation method. In this paper, using Hough transform based on the GPR image velocity estimation, detection of target position, size and the hyperbolic edge noise and target hyperbola discrete derivative method targets on the forming of the estimation result will affect the resulting estimation error. In order to solve the error, the algorithm proposed in this paper is optimized by using weighted derivatives and Lagrange multiplier method, effectively restrain the influence of noise on the hyperbolic edge velocity estimation, which can enhance the hyperbolic discrete function derivative calculation accuracy, making the estimation results more accurate.In this paper, the method of clutter suppression and improved Hough transform algorithm are proposed for the ground penetrating radar root radar image processing, so as to achieve the extraction of forest root target hyperbolas and velocity estimation, prepared for the project of forest root location, root diameter and root biomass estimation. |
PDF Full Text Request