Machine vision-based agricultural product volume rapid measurement device and measurement method

Abstract

The invention discloses a machine vision-based agricultural product volume rapid measurement device and measurement method. According to the machine vision-based agricultural product volume rapid measurement device and measurement method, three CCD industrial cameras are arranged according to a certain angle; the three CCD industrial cameras are connected with a camera driver and an image acquisition processor; the three CCD industrial cameras are utilized to parallelly acquire the image information of a training sample; segmentation and edge feature extraction are performed on images; the maximum length dimension of the three images and the maximum length dimension of the three images in the orthogonal directions of the three images are obtained through a bounding rectangle method; PLSR (partial least squares regression) modeling is carried out, so that an agricultural product volume prediction model can be obtained; the six dimension parameters of the three images of an agricultural product to be measured are substituted into the volume prediction model; and the volume of the agricultural product can be obtained fast. Since the volume prediction model is built in advance, so that the low real-time performance of a three-dimensional virtual reproduction process can be neglected; and in a volume measurement stage, the volume of the agricultural product to be measured is quickly obtained through the volume prediction model, and therefore, a necessary classification basis can be provided for agricultural product intelligent sorting based on a plurality of features such as color and volume.

Description

Technical field [0001] The invention belongs to the field of agricultural product screening, in particular to a device and method for quickly measuring the volume of agricultural products based on machine vision. Background technique [0002] At present, the color feature detection technology of agricultural products based on machine vision is relatively mature, and the color feature detection technology has been successfully applied to commercial agricultural product classification and screening devices, such as rice color sorter, corn color sorter, etc., according to different colors and mildew, The degree of reflection to light such as belly white can remove damaged grains, heterochromatic grains and other impurities in agricultural products. The existing agricultural product classification and screening methods are mainly based on the color selection technology based on a single color feature, and the appearance, volume, especially the volume and other characteristics of the ...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the defect that the current sorting method of agricultural products mainly uses a single color feature as the basis for sorting, the present invention provides a rapid volume measurement device for agricultural products based on machine vision, which uses three CCD industrial cameras at angles to construct a visual system for The synergy of rapid volume measurement of agricultural products, and the use of image acquisition processor and camera driver to grayscale the image, image segmentation and edge feature extraction, so as to obtain the appearance characteristics of the agricultural products to be measured, combined with the volume prediction model of agricultural products, quickly obtain The size of the produce

Images