Deep fully-convolutional neural network-based fast segmentation method of field rice panicles

Abstract

The invention discloses a deep fully-convolutional neural network-based fast segmentation method of field rice panicles. According to the method, a deep fully-convolutional neural network-based segmentation network of the field rice panicles is designed. A to-be-segmented field image is firstly divided into a plurality of sub-graphs which are suitable for an input size of the deep fully-convolutional neural network, pixel-level semantics segmentation is carried out on all the sub-graphs by the deep fully-convolutional neural network, and then all the sub-graphs are spliced for obtaining a segmentation result of the same size as the input image. According to the method, impacts of huge differences of colors, shapes, sizes, postures and textures of the rice panicles of different varieties and growth stages, severe irregularity of rice panicle edges, panicle leaf color aliasing and uneven and varying illumination, occlusion and wind blowing factors in a field can be overcome, and fast andaccurate segmentation on the field rice panicles of the different varieties and growth stages can be realized. Compared with the prior art, the method has the technical advantages of high precision,high applicability and high processing speed.

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 method for rapidly segmenting rice spikes in a field based on a deep fully convolutional neural network. 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 field of rice breeding and cultivation, it is necessary to measure the yield of a large number of candidate samples in different environments, so as to provide a scientific basis for cultivating high-yield, high-quality, and stress-resistant rice varieties. Rice panicle is the organ where rice grains grow, and the traits of panicle are directly related to rice yield. Rice ears also play a very important role in the detection of rice diseases and ins...

AI Technical Summary

Problems solved by technology

[0006] In order to overcome the problem of low segmentation accuracy and long time-consuming in the prior art for different varieties and growth stages of the field rice ear segmentation method, the present invention provides a fast segmentation of field rice ears based on a deep fully convolutional neural network method to realize the segmentation of multi-variety rice ears at different growth stages in a complex field environment

Images