Corn seed investigating device based on machine vision

Abstract

The invention discloses a corn seed investigating device based on machine vision, comprising an object stage, an image collection unit and two reflectors. The image collection unit is opposite to the object state; the two reflectors are placed at the other side of the object stage opposite to the image collection unit and are arranged in the position where the two reflectors can image the part of the corn ear part to be tested, which can not be directly collected by the image collection unit. The device has the advantages of time and labor saving, rapidness, high efficiency, and the like; images at three sides of the corn ear can be imaged at one time through two reflectors; the object stage has the weighing function without other auxiliary weighing instruments; three linear laser light sources can be used for distinguishing the boundary of the corn overlapping part; the collected picture can be used for extracting multiple appearance parameters of progeny length, progeny row, progeny grains, grain number of each row, bald length, bald rate, and the like; and the defects for artificial seed investigation at present is improved.

Description

technical field [0001] The invention relates to the technical field of corn seed testing, in particular to a corn seed testing device with machine vision. Background technique [0002] my country's corn planting area and total output both rank second in the world, and corn production occupies a very important strategic position in my country's grain production and food security. The appearance parameters and ear weight of corn ear directly reflect the growth status of corn and also play an important role in the identification of new corn varieties. The maize test by machine vision has the advantages of objectivity, high efficiency and low cost. [0003] In 2009, Zhao Chunming et al. published an article "Research on the Application of Image Processing in the DUS Character Test of Corn Ears". In the article, the CCD camera was fixed on the tripod. The shooting environment is a closed box, exposed with a whiteboard, and the rotation angle around the longitudinal axis is unifo...

AI Technical Summary

Problems solved by technology

This method has the following defects: the corn is hung vertically above the airtight box, which is inconvenient to operate, and it is easy to cause the corn to fall off due to unstable fixation or loose corn, which will have a certain impact on taking pictures and waste time; To collect a side image of corn, if you want to collect multiple sides, you must rotate the camera three times, 120° each time to collect three side images of corn for research. The rotation process is easy to cause errors, and the accuracy of the rotation angle is not easy to control. Human subjective influence is large, and there will be different degrees of offset; third: because the three-sided images taken will overlap, and the boundaries cannot be distinguished, the number of ear rows of corn cannot be measured through the plane image, and the method of scanning the corn cross-section

[0005] Neither of the above two methods can measure the weight of corn ears, only images can be collected

Images