Tea tender shoot segmentation and identification method based on color and region growth

Abstract

The invention provides a novel tea tender shoot segmentation and identification method based on color and region growth, relating to the technical field of image processing. The method comprises the following steps of: firstly, converting the original tea RGB (Red, Green, Blue) color image into an HIS color space, and carrying out initial seed selection according to the parameters of hue H and saturation S therein; then carrying out region growth on a seed region according to the color similarity and region contiguity; carrying out region growth and combination by combining with color distance and edge distance to segment tea tender shoots; then extracting the shape feature parameters of binarized color and tender shoot images; and finally finishing identification through an improved hereditary neural network. Through a segmentation and identification experiment on the tea tender shoots in the tea image, the result shows that the algorithm can well separate the tea tender shoots from the tea image, well store the outline information of the tea tender shoots and obtain very good identification results.

Description

technical field [0001] The invention relates to an image processing technology, which is a method for segmenting and identifying tea buds in an on-site tea image based on color and region growth. Background technique [0002] The application of electronic technology and computer image processing technology in the production of agricultural farmland environment, greenhouses and orchards is the requirement of modern agriculture and precision agriculture. Although there have been some studies on the analysis and identification of tea color and quality using computer technology, but in the past Most of the research is on the shape and color of the tea leaves that have been made or the soup color of the tea leaves. There are very few corresponding studies on the segmentation and identification of tea buds in the field tea trees in China. The growth state and maturity of tea buds play an important role in the timely picking of tea and the quality of tea products in the later stage...

AI Technical Summary

Problems solved by technology

It simulates the reproduction, mating and mutation phenomena that occur in natural selection and natural inheritance. The convergence speed of the commonly used BP neural network algorithm is very slow, it is easy to fall into the local minimum and the global search ability is weak, and the numerical value is high when the learning rate is large. Features such as poor stability

Images