Automatic observation method of tillering stage of sugarcane

Abstract

The invention discloses an automatic observation method for the tillering stage of sugarcane, which comprises the following steps: obtaining a sugarcane image in real time; pre-processing the sugarcane image; extracting the image of a sugarcane plant; calculating the green pixel coverage FVC of the sugarcane plant; calculating the number of the apexes of the sugarcane plant; calculating the numberof leaves and stem crossing points of the sugarcane plant; if the green pixel coverage FVC>=coverage threshold FVCg, the number of apexes of the sugarcane plant>=the threshold of the value of the apexes of the sugarcane plant, and the number of intersections of the leaves and the stem of the sugarcane plant>= the threshold of number of intersections of the leaves and the stem of the sugarcane plant, determining that the sugarcane plants in the sugarcane images will enter the tillering stage. The green pixel coverage of the sugarcane plant, the number of the apexes of the sugarcane plant and the number of the intersecting points of the leaves and stem of the sugarcane plant are calculated by using image recognition technology to finish automatic observation of the tillering stage of sugercane, which has higher objectivity, higher timeliness and practicability than traditional manual observation method, and reduces labor intensity and production cost.

Description

technical field [0001] The invention relates to the technical field of sugarcane planting. More specifically, the present invention relates to an automatic observation method for sugarcane tillering stage. Background technique [0002] Sugarcane is one of the most important sugarcane crops in the world, and the sucrose extracted from sugarcane accounts for more than 90% of the sugar production in my country. The development of precision agriculture is an effective measure to realize the sustainable development of sugarcane industry, and the real-time and accurate acquisition of sugarcane growth period information is one of the key foundations for the implementation of precision sugarcane industry. Through accurate observation of information such as the speed and progress of sugarcane development, it is possible to scientifically guide production activities such as irrigation, fertilization, and pest control to increase production. It can be applied to sugarcane growth model...

AI Technical Summary

Problems solved by technology

In 2012, Zhong Chu et al. published "Prediction of Sugarcane Physiological Development Time and Growth Period" in "Saccharaceae of China" by using the crop physiological development time to predict the germination period, seedling period and stem elongation period of sugarcane. The relative errors of the prediction results were 7, 3, 6, and 15 days, the prediction effect is better, and it has certain practicability, but because the growth period is twice a day, manual inspections and observations have caused certain errors, and there is no prediction for the tillering stage of sugarcane

The above related literature does not provide a real-time detection method for sugarcane tillering stage, and the satellite remote sensing image has a long imaging distance and low resolution, and the imaging quality is greatly affected by the weather, so it cannot actively monitor changes, and is more suitable for the analysis of the overall growth status of crops on a large scale , not applicable to the selection of sugarcane tillering observation method

Images