Wheat morphology double-sided identification device

Abstract

The invention discloses a wheat morphology double-sided identification device, which comprises a camera shooting device, a wheat grain laying device, a wheat grain placing device and a computer processing system, wherein the camera shooting device can obtain images of the front side and the back side of each wheat grain, a plurality of weighing devices are arranged on the wheat grain placing plate, independent weight data of each wheat grain can be obtained, the data is transmitted to the computer processing system for numbering, the wheat grain weight numbers are in one-to-one correspondencewith the image numbers of the corresponding wheat grains, the computer processing system can screen out wheat grain numbers which are not in a standard weight range in advance, and then an image recognition module of the computer processing system only recognizes the images corresponding to the wheat grain numbers which are not in the standard weight range, recognizes the abnormal types of the wheat grains with the non-standard weight, and carries out statistics on each disease and damage to obtain multi-parameter quality indexes of the group of wheat.

Description

technical field [0001] The invention relates to a wheat detection device for laboratory use, in particular to a double-sided identification device for wheat morphology. Background technique [0002] In the process of wheat production and planting, various factors will cause defects that affect its quality, such as insect erosion, disease, damage, germination, moldy kernels, etc. Therefore, it is necessary to detect the quality of wheat quickly, accurately and objectively. In the traditional inspection process, simple appearance observation is often used to judge the above-mentioned quality defects. [0003] Compared with traditional manual detection methods, a method of detecting the appearance of wheat through machine vision technology has appeared in the laboratory. However, the image recognition device in the prior art can only recognize the wheat image on one side because it only takes the wheat image from the upper part. Due to the uncertainty of the location of wheat...

AI Technical Summary

Problems solved by technology

However, the image recognition device in the prior art can only recognize the wheat image on one side because it only takes the wheat image from the top.

Due to the uncertainty of the location of wheat diseases, etc., only one-sided identification will lead to inaccurate results

[0004]At the same time, due to the limitations of hardware and computing power, it is still unable to meet the timely and rapid detection requirements in the actual grain purchase process

Adding multi-angle images of wheat morphology increases the difficulty of image processing

Images