Method for automatically detecting three-true-leaf period of cotton

Abstract

The invention provides a method for automatically detecting a three-true-leaf period of cotton. The method includes adopting a real-time downward-looking image acquired in a cotton field as an object; dividing cotton plants by the aid of color characteristics; and judging whether the cotton reaches the three-true-leaf period or not by the aid of extracted characteristics of the image. Important parameters representing the growth condition of the cotton are used as judgment basis in the method, the growth period of the cotton is judged in the field, and the accuracy of a detection result is high. In addition, the method has an important guidance significance for evaluation of agricultural metrological conditions for growth of the cotton and farming activities of the cotton.

Description

technical field [0001] The invention belongs to the field of combination of digital image processing and agricultural meteorological observation, and in particular relates to a method for detecting whether cotton has reached the third true leaf stage by using image features to take cotton field down-view sequence images as objects. Background technique [0002] Cotton is one of the main economic crops in my country, and the planting area is very extensive. Cotton development period, yield and quality are vulnerable to climate change. For a long time, the observation of cotton development period has been mainly through manual observation, which is greatly affected by the subjective factors of observers. At the same time, due to the long observation period and wide area, it is not economical to use manpower to observe. Therefore, it is very necessary to observe the developmental stages of cotton fields by taking video images of cotton fields and using image processing methods...

AI Technical Summary

Problems solved by technology

This method uses hyperspectral data to monitor the growth information of cotton. Although the growth information of the growth period is obtained, it does not detect the specific growth period of cotton.

Moreover, hyperspectral data, like remote sensing data, is susceptible to interference and does not directly reflect the growth status of plants

[0004] The above monitoring of cotton growth, area, yield and growth period information is determined using satellite images, remote sensing data or hyperspectral data, but these images are not conducive to accurately judging the growth status of crops due to their own limitations

Images