A Coarse and Fine Secondary Paper Counting Method Based on Spatial and Gray Features

Abstract

Disclosed is a coarse-fine two-stage paper counting method based on space and gray features. First, an image of a bundle of paper collected is grayed, the tilt of the image is corrected using a Hough change method, the paper image is projected longitudinally to get a projection curve, the projection curve is processed using a fuzzy threshold method, a group of peak and trough features with better nature is decided, and the paper is counted according to the number of peaks or troughs of the processed curve to get the initial number of the bundle of paper. In view of the fact that paper may be adhered when arranged and placed and the number of pieces of paper cannot be detected using a coarse counting method, a classification method is designed according to the features of abnormal change of gray and width at the paper adhesion part. The width of the paper and the gray difference between the trough points of the projection curve and the mean of the projection curve are calculated respectively; then, the average gray difference of the projection curve and the average width of the paper are calculated, and the paper width and the gray difference are used to constitute a 2D feature vector; and finally, the feature vector is classified using a K-means clustering method to judge the number of pieces of paper adhered.

Description

technical field [0001] The invention relates to a method for counting paper by using paper space and gray scale characteristic paper, and its main application fields are printing and packaging and papermaking industry. Background technique [0002] In the printing and packaging industry, in order to accurately grasp the company's paper inventory and shipment quantity, it is often necessary to count the number of paper. Due to the large scale of paper in the modern printing and papermaking industry and people's advocacy of the concept of green environmental protection and high requirements for the quality of printing paper in modern life, the methods of counting the production, printing and packaging of paper in today's society are mostly Both use manual counting or mechanical equipment to count paper. Although the accuracy of manual counting of paper or banknotes is guaranteed to a certain extent, the speed of paper counting is slow, especially when the number of paper excee...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to provide a paper counting method based on spatial and grayscale features. The method collects the imaging effect map of the paper by an image acquisition device, and after obtaining the image of the paper, first analyzes and predicts the paper image features. Processing, using fuzzy threshold and detection curve peak-to-valley point algorithm to calculate the number of papers, which fundamentally solves the shortcomings of slow manual counting and cannot be carried out for a long time. Disadvantage of taking a long time in many cases

Images