Extraction Method of Corn Crop Row Skeleton Based on Maximum Square Principle

Abstract

The invention discloses a corn crop row skeleton extraction method based on the maximum square principle, the steps of which are as follows: collecting RGB color images of corn crop rows; using an improved overgreen grayscale method to grayscale the corn crop row images; The obtained grayscale image is segmented by using median filter; binarization operation is performed on the image filtered by the middle finger; morphological operation is performed on the binarized image to obtain the crop row outline image; The skeleton extraction algorithm is used to extract the skeleton line of the corn crop row; the crop row navigation line is obtained by using the commonly used random Hough transform algorithm. The invention can provide accurate position information for the navigation line, and has strong robustness to the edge noise of the crop row, not only maintains the singleness and stability of the corn crop row skeleton, but also has the least number of redundant branches for extracting the skeleton, It also has good adaptability and anti-interference ability to images with complex backgrounds.

Description

technical field [0001] The invention relates to the technical field of crop row detection, in particular to a method for extracting corn crop row skeletons based on the maximum square principle, which is mainly used in the detection of crop row navigation lines for automatic walking of agricultural machinery or automatic alignment of precision pesticide application machines. Background technique [0002] In the research of modern precision agriculture system, precision pesticide application technology occupies an important position, but the extraction of crop row skeleton line plays an important role in the acquisition of pesticide application machine navigation information. Digital image processing algorithm has great advantages in automatic identification, and it is the basis and key technology of modern precision pesticide application technology. Although the crop row skeleton extraction algorithms studied in the past have strong topological properties to a large extent, ...

AI Technical Summary

Problems solved by technology

Although the above studies on skeleton line extraction are largely topological, it is difficult to guarantee the stability and pixel unity of the skeletons extracted by these algorithms for objects with complex backgrounds, so further research is needed. research and experiments

[0004] The Chinese invention patent application "Crop Row Detection Method Based on Minimum Tangent Circle and Morphological Principle" filed on June 30, 2015 with the application number 2015103642669 uses the principle of minimum tangent circle to extract the central crop row skeleton, which is well connected However, there are many redundant branches in the skeleton extracted by morphological refinement, which affects the accuracy of navigation line fitting and is not adaptable.

Images