Tea leaf recognition method and tea leaf sorting device using same

Abstract

The invention provides a tea leaf recognition method and a tea leaf sorting device using the same. The tea leaf recognition method includes following steps: step 1, acquiring an image of known tea leaves; step 2, preprocessing the acquired image of the known tea leaves; step 3, extracting image features of the preprocessed image of the known tea leaves; step 4, screening and classifying the imagefeatures, and building a prediction recognition model; step 5, referring to the prediction recognition model to judge a variety of to-be-recognized tea leaves. The tea leaf recognition method has theadvantages that by means of combining intelligent hardware with a mobile phone operation system development platform, tea leaf variety recognition rate is increased effectively; the image is acquiredthrough a photographing function of mobile equipment, and corresponding image recognition processing is performed, so that the problem that image recognition equipment is inconvenient to carry can besolved effectively; through optimization and innovation of a conventional recognition algorithm, tea leaf variety recognition accuracy is improved effectively, and recognition effect which is more convenient, quick and effective is acquired.

Description

technical field [0001] The invention belongs to the technical direction of digital image recognition and processing of improved pattern recognition algorithms, and specifically relates to the field of tea recognition and tea sorting, in particular to a tea recognition method for automatic recognition based on indicators such as tea type and shape and tea leaves using the method Sorting device. Background technique [0002] Tea refers to the leaves or buds of the tea tree. It is one of the three major beverages in the world. It has the functions of clearing away heat and detoxification, fighting cancer, and clearing the mind. It is deeply loved by the public. At the same time, tea planting has a long history and the planting area is very extensive. At present, there are four first-level tea areas in China, namely, South China tea area, Southwest tea area, Jiangnan tea area, and Jiangbei tea area. The tea planting area is very wide, which is enough to prove that Tea is very p...

AI Technical Summary

Problems solved by technology

But at the same time, due to the wide variety of tea leaves with different shapes and colors, some tea leaves have small differences in leaf size, shape and color. If you continue to use traditional characteristic parameters, such as leaf area, perimeter, and color, it is no longer possible to meet the current needs. There is an urgent need to improve the traditional recognition algorithm and create a new algorithm so that it can distinguish tea leaves of different types, colors and sizes more carefully, thereby improving the recognition accuracy.

Images