Granular material sorting classifying method based on visual sense recognition

Abstract

The invention discloses a particulate material separation and classification method based on visual identity. The method comprises obtaining sample images, image feature information preprocessing, designing classifiers, establishing feature information expert database, proposed material separation to match with the feature information expert database and carrying out separation and classification. The separation and classification method overcomes the shortcomings of current poor technical accuracy, low efficiency and being easy to cause the secondary pollution, etc. and the separation and classification method ensures the product quality, realizes fast intelligence and improves the separation and classification effect of particulate materials.

Description

Technical field [0001] The invention belongs to a sorting and grading method for granular materials, in particular to an intelligent visual recognition-based sorting and grading method for granular materials using embedded visual image technology and pattern recognition technology. Background technique [0002] With the development of my country’s grain industry and the development of food deep processing technology, color sorting technology and near-infrared spectroscopy analysis technology have been gradually applied to the classification and sorting of agricultural and sideline products such as rice. Essentially, the classification and sorting requirements of these agricultural materials are It has a certain similarity, that is, it belongs to the granular materials that need to be sorted and classified according to color and intrinsic quality. However, due to the different spectral characteristics of these sorted materials, and their sorting and grading standards are also very ...

AI Technical Summary

Problems solved by technology

[0005] Due to the characteristics of the silicon photodetection device itself, when some colors are very rich and the colors of the normal materials and the materials to be rejected are very close (such as green and light yellow raisins in raisins), the resolution accuracy is limited. Therefore, based on this detection The sorting device of the device may not be able to effectively sort some materials

Images