Tea yield prediction system and method

Abstract

The invention provides a tea yield prediction system and method. The system comprises an image splicing module used for obtaining an orthographic mosaic image of a target tea garden; a tea bud classification and recognition module used for inputting the orthographic mosaic pattern of the target tea garden into a tea recognition model to obtain tea buds of different grades in the target tea garden,wherein a tea bud recognition model is obtained by training based on tea bud samples of different grades and preset labels of the tea bud samples; and a yield calculation module used for obtaining the total yield of the tea buds in the target tea garden according to the number of the tea buds corresponding to each grade and the single mass of the tea buds corresponding to each grade. The system adopts a non-contact and non-destructive mode with tea leaves, the tea bud quantity can be counted without picking tea buds, the tea buds are identified and counted in a machine learning mode, nondestructive monitoring is realized, the real-time performance is high, and the data precision is relatively high.

Description

technical field [0001] The invention relates to the technical field of machine vision, in particular to a tea yield prediction system and method. Background technique [0002] China is a big tea production and export country in the world, with many varieties of tea and a large planting area. The amount of tea buds picked has always been a concern of tea farmers, and it is also an important factor affecting market returns. The traditional statistics of the amount of tea buds are generally divided into two ways, one is to rely on manual experience to pick, and then weigh to realize the statistics of the output, which has many deficiencies. [0003] The existing statistical yield method is based on climatic and environmental factors combined with the empirical model of historical data for general estimation. The result of such analysis is firstly affected by the accuracy of the model, because the accuracy of the model is affected by the accuracy of the climatic and environment...

AI Technical Summary

Problems solved by technology

[0005] The embodiment of the present invention provides a tea production forecasting system and method, which is used to solve the defect of inaccurate tea production forecast in the prior art, so as to improve the accuracy of tea production forecast

Images