Rice spike segmentation method of big paddy field based on deep learning and super-pixel segmentation

Abstract

The invention discloses a rice spike segmentation method of a big paddy field based on deep learning and super-pixel segmentation. According to the invention, by use of simple linear iteration clustering method in the super-pixel segmentation technology, adjacent pixels with similar characteristics form image blocks, i.e., super-pixels; based on automatic annotation and selection of large-scale training samples, the type of the super-pixels is discriminated through a convolution neural network in the deep learning technology so that initial segmentation of rice spikes is achieved; and based on the entropy rate-based super-pixel segmentation method, initial segmentation results are optimized. Thus, effects caused by great difference in colors, shapes, sizes, poses and textures of different varieties of rice spikes in different growth periods, serious irregular edges of the rice spikes, color aliasing of spike leaves, and uneven and changeable light, shielding and wind blowing in the field can be overcome; precise segmentation of different varieties of rice spikes in different growth periods is achieved; and the method is applicable to segmentation of rice spikes in the pot plant environment. Compared with the prior art, the method is advantaged by high precision and applicability.

Description

technical field [0001] The invention belongs to the field of agricultural automation, and in particular relates to automatic measurement of rice phenotypic parameters, in particular to a field rice ear segmentation method based on deep learning and superpixel segmentation. Background technique [0002] The production and distribution of rice is related to the food security of more than half of the world's population. High yield has always been one of the important goals of rice breeding and cultivation. In the research of rice breeding and cultivation related fields, it is necessary to measure the yield of a large number of candidate samples under different environments, so as to provide a scientific basis for cultivating high-quality, high-yield, and stress-resistant rice varieties. Field rice yield measurement usually selects several representative small field plots according to certain principles in the plot or large area. After harvesting, threshing, drying, cleaning, a...

AI Technical Summary

Problems solved by technology

[0006] In order to overcome the large difference in the prior art for different varieties, different growth periods and different environments, the rice ear segmentation methods are quite different, and there is no such problem that there is no multi-variety rice ear segmentation method applicable to different growth periods in a complex field environment. A rice ear segmentation method based on deep learning and superpixel segmentation is provided to realize the segmentation of multi-variety rice ears at different growth stages in a complex field environment

Images