A sugarcane plant number automatic identification method based on a crop canopy aerial photography top view

Abstract

The invention provides a sugarcane plant number automatic identification method based on a crop canopy aerial photography top view, and the method comprises the steps: taking a field sugarcane canopyimage which is taken by an unmanned aerial vehicle under a natural light condition in a top view as a research object, and carrying out the removal of a field complex background and the preliminary extraction of sugarcane leaf white meridians through employing a digital image processing technology; Extracting end points of the white meridians after connecting the broken white meridians with the same root to obtain an end point distribution map of the white meridians; and finally, identifying sugarcane plant centers from the end point map of the white meridians by utilizing a DBSCAN clusteringalgorithm according to the distribution characteristic of dense end points in a plant center region. Calculating the average value of various plant center areas to be approximately used as plant center coordinates, and counting the number of the coordinates to obtain the plant number, thereby realizing automatic identification of the field sugarcane plant number. When the method is used for counting field sugarcanes, the flexibility is high, the precision is good, the implementation cost is low, statistical mistakes and omissions caused by artificial subjective factors can be effectively reduced, and the automatic detection degree of the sugarcane early-stage plant number is improved.

Description

technical field [0001] The invention relates to an automatic identification method for the number of sugarcane strains, in particular to an automatic identification method for the number of sugarcane strains based on aerial photography of crop canopies, and belongs to the technical fields of digital image processing, image segmentation and machine vision. Background technique [0002] Sugarcane is an important sugar crop in the world. Sugarcane cultivation believes that reasonable planting specifications can promote the rational allocation of ecological factors such as water, fertilizer, light and heat, so as to obtain higher yield and better quality. Therefore, obtaining the number of sugarcane plants in the field is of great significance for optimizing the planting density of different sugarcane varieties and predicting yield. At present, the main method of sugarcane counting is still based on artificial field statistics, that is, before the sugarcane is harvested, the nu...

AI Technical Summary

Problems solved by technology

However, there are relatively few studies on the counting method of sugarcane, which has different attitudes, overlapping and irregular objects that are difficult to segment, and most of them are based on remote sensing images to predict sugarcane yield, and the implementation cost is high.

Images