Image recognition-based staple crop plant height measurement method

Abstract

The invention discloses an image recognition-based staple crop plant height measurement method. The method comprises the following steps of: obtaining a field staple crop image with a color label foronce in a unit time in a growth period of a staple crop, wherein the staple crop image comprises the color label and a background crop; preprocessing the staple crop image to obtain a complete staplecrop image with the color label; extracting the color label in the complete staple crop image to obtain a color label image; denoising the color label image to obtain a color label binary image; calculating a pixel height of the color label according to coordinate information at the bottom and top point of a connected domain in the color label binary image; and calculating a staple crop plant height according to the pixel height of the color label and a relative position relationship in which the color label is shielded by the background crop. According to the image recognition-based staple crop plant height measurement method, automatic continuous observation for staple crop plant heights is realized.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a method for measuring plant height of bulk crops based on image recognition. Background technique [0002] In the process of agricultural production, timely and accurate grasp of crop growth information can provide guidance for field production and management. Crop height, as one of the important growth parameters of crops, is an important indicator for measuring crop growth speed and predicting crop yield. At present, the observation of plant height of field crops is still done by agricultural meteorological observers using a tape measure in the field. This method has disadvantages such as heavy workload, inability to achieve continuous observation during the whole process of crop growth, and the observation process is easily affected by subjective factors. [0003] Realizing the automatic observation of crop growth parameters is the basis and premise of efficient and...

AI Technical Summary

Problems solved by technology

Many scholars have carried out a series of research in this field. Gao Wei, Chen Jin, etc. published "Development of Crop Height Automatic Measuring Device" in "Agricultural Mechanization Research". In this paper, ultrasonic ranging technology was used to develop a highly automatic However, due to the large beam divergence and low orientation accuracy of ordinary ultrasonic waves, it has not achieved good development in the field of crop height measurement.

In the article "Remote Non-destructive Measurement System for Plant Height Based on Network Camera" published by Jiang Pu and Geng Nan, the principle of triangulation is used to operate the returned monitoring video stream and calculate the rotation angle of the optical axis to realize the partial growth period of corn. plant height measurement, but this method requires that the plant cannot be tilted during growth, otherwise it will cause large errors. Therefore, this type of method generally has high requirements for the experimental environment and cannot be applied to the real environment of field farmland.

Chinese patent CN 105869152 A discloses a method for measuring the spatial distribution of crop plant height by UAV remote sensing. The target image sequence collected by the UAV airborne sensor is used to construct the crop canopy spatial information of the crop coverage area in the target area respectively. The data surface model (DSM) of the target area and the base surface model (BSM) corresponding to the bottom of the crop, using the difference between the DSM and the BSM, obtains the spatial distribution of the crop height in the target area. problem, but this method has the problem of not being able to observe continuously

Chinese patent CN 106871799 A discloses a fully automatic crop height measurement method and device, which realizes non-contact and automatic measurement of crop height by combining laser scanning data and camera image data. Large, increasing the maintenance cost of field equipment

Images