Adhesive corn ear segmentation method and device used in cell corn ear species testing

Abstract

The invention provides an adhesive corn ear segmentation method and device used in cell corn ear species testing. The method comprises the steps of: obtaining a two-dimensional colored picture of cell corn ears collected by a two-dimensional image collection device, and processing the two-dimensional colored picture into a binary picture including at least two binary sub-images (hereinafter referred to as sub-images); obtaining a blank profile diagram of any one sub-image, and calculating the perimeter and area of the blank profile diagram; according to the perimeter and area, calculating the adhesive coefficient of the sub-image; if the adhesive coefficient is no more than a preset value or the area is no less than a preset value, generating an image and a background mark line; and according to the mark line and the binary picture, generating a mask for segmenting the sub-images, the segmenting the sub-images according to the mask, and obtaining profile diagrams of the sub-images. The device comprises a binary processing unit, a profile diagram obtaining unit, a calculation unit, a mark line generation unit, a mask generation unit and an image segmentation unit. According to the invention, the speed is high, the efficiency is high, and the adhesive corn ear segmentation method and device are applicable to species testing of various kinds of adhesive corn ears and are universal.

Description

technical field [0001] The invention relates to the technical field of corn ear seed testing, in particular to a method and a device for segmenting cohesive corn ear seeds in the corn ear seed test in a plot. Background technique [0002] At present, domestic corn seed testing mainly relies on manual identification, and the quantity of each batch of corn is very large. The traditional seed test process is often to test each ear one by one. The process needs to last for more than one month, and there is a huge labor cost And the waste is serious, the work efficiency is low and the precision is not high and many other drawbacks. [0003] The current methods for segmenting the contours of cohesive corn ears can only segment regularly arranged series-type cohesions, but cannot segment other types of cohesions. [0004] For this reason, it is necessary to provide a method and device for segmenting cohesive corn ears in the testing of corn cobs in small plots. The method is appli...

AI Technical Summary

Problems solved by technology

[0002] At present, domestic corn seed testing mainly relies on manual identification, and the quantity of each batch of corn is very large. The traditional seed testing process is often to test each ear one by one. The process needs to last for more than one month, and there is a large labor cost And there are many disadvantages such as serious waste, low work efficiency and low precision.

[0003] The current method for segmenting the contours of cohesive corn ears can only segment regularly arranged series cohesions, but cannot segment other types of cohesions

Images