Cotton image segmentation method and device in natural environment

Abstract

The invention relates to a segmentation method and device for a cotton image in a natural environment. The method comprises steps of: transforming a natural environment cotton image to an image in a YIQ color space, in an Extra color space and in a L *a * b color space, and extracting a Q component image, an Extra Green component image and an a*component image; via a PCNN model, respectively obtaining an object zone and background segmentation binary image corresponding to the Q component image, an object zone and background segmentation binary image corresponding to the Extra Green component image, and an object zone and background segmentation binary image corresponding to the a*component image; performing an intersection operation based on the object zone and background segmentation binary images respectively corresponding to the Q component image, the Extra Green component image and the a*component image; and obtaining a cotton image segmentation result based on an intersection operation image and the natural environment cotton image. The segmentation method and device are advantageous in that, accurate automatic segmentation of a cotton target and a complex background, and accurate extraction of a crop area in the natural environment are achieved, and great significance is provided for the field of crop growth state judgment, etc.

Description

technical field [0001] The invention relates to the technical field of agricultural information, in particular to a cotton image segmentation method and device in a natural environment. Background technique [0002] Cotton is one of the most important crops in the world. It has the characteristics of high yield, wide distribution and multiple uses. As an important economic crop in China, the cultivation of cotton has played an important role in adjusting the structure of agricultural production and ensuring the income of farmers. However, cotton is susceptible to various diseases and insect pests during its growth and development. Image processing methods and computer vision technology can be used to identify and judge images of cotton diseases and insect pests, and automatic segmentation of cotton and complex backgrounds is the first step in image processing and analysis, and it is also a crucial step. The quality of the segmentation results directly affects It affects th...

AI Technical Summary

Problems solved by technology

[0005] In view of the currently commonly used crop image segmentation method that does not consider the image is affected by illumination and complex backgrounds, and has poor robustness, the present invention proposes the following technical solutions:

Images