Method of counting sheet materials

Patent metadata

Field	Value
Publication number	US-11094052-B2
Application number	US-202016909129-A
Country	US
Kind code	B2
Filing date	Jun 23, 2020
Priority date	Jul 12, 2019
Publication date	Aug 17, 2021
Grant date	Aug 17, 2021

Abstract

A method of counting sheet materials applied to a pile of sheet materials, comprising the steps of: receiving an image of the pile of sheet materials; obtaining a grayscale value of a plurality of pixels along a first image axis direction of the image to form an one dimensional first array; performing binarization of the first elements of the first array with a first threshold value to form an one dimensional second array; obtaining the number of the second elements of a first value appearing between two second elements of a second value in the second array to form a third array; dividing the elements of the third array into a first cluster and a second cluster with a second threshold value; counting the number of the third elements belonging to the first cluster and defining said number as the number of the first sheet materials.

First claim

What is claimed is:

1. A method of counting sheet materials, which is applied to a pile of sheet materials comprising a plurality of first sheet materials and a plurality of second sheet materials, which are staggeringly arranged along an axial direction, wherein each of the second sheet materials is adjacent to each of the first sheet materials, and a thickness of each of the first sheet materials is greater than a thickness of each of the second sheet materials; the method of counting sheet materials is performed by an image processing module, comprising the steps of:
A. receiving an image of the pile of sheet materials;
B. obtaining a plurality of grayscale values of a plurality of pixels along a first image axial direction corresponding to the axial direction of the image, to form a first array which is one-dimensional, wherein the first array comprises a plurality of first elements; values of the plurality of first elements are the grayscale values respectively;
C. performing binarization of the first elements of the first array with a first threshold value to form a second array which is one-dimensional, wherein the second array comprises a plurality of second elements; a value of each of the plurality of second elements is one of a first value and a second value, wherein the first value is greater than the second value;
D. obtaining a number of the second elements of the first value appearing between two of the second elements of the second value in the second array, to form a third array, wherein the third array comprises a plurality of elements which comprises a plurality of third elements; a value of each of the plurality of third elements is the number of the second elements of the first value, which appears;
E. dividing the elements of the third array into a first cluster and a second cluster with a second threshold value, wherein values of the third elements belonging to the first cluster are not less than the second threshold value;
F. counting a number of the third elements belonging to the first cluster, and defining the number of the third elements belonging to the first cluster as a number of the first sheet materials.

2. The method of claim 1 , wherein step C comprises a first threshold value setting step which is provided to set the first threshold value.

3. The method of claim 2 , wherein the first threshold value setting step comprises the steps of:
using a classification algorithm to divide the first elements into two groups, and calculating a mean value of the first elements of each of the two groups, wherein the mean value of one of the two groups is greater than the mean value of the other group;
taking the smallest value among the first elements in one of the two groups with greater mean value as a first predetermined value, and taking the largest value among the first elements in the other group with smaller mean value as a second predetermined value; and
setting the first threshold value as a value between the first predetermined value and the second predetermined value.

4. The method of claim 3 , wherein the first threshold value is a mean value of the first predetermined value and the second predetermined value.

5. The method of claim 3 , wherein the classification algorithm is k-means algorithm.

6. The method of claim 1 , wherein step E comprises a second threshold value setting step, which is provided to set the second threshold value.

7. The method of claim 6 , wherein the second threshold value setting step comprises the steps of:
using a classification algorithm to divide the elements of the third array into two groups, and calculating a mean value of the elements of each of the two groups, wherein the mean value of one of the two groups is greater than the mean value of the other group;
taking the smallest value among the elements of the third array in one of the two groups with greater mean value as a first predetermined value, and taking the largest value among the elements of the third array in the other group with smaller mean value as a second predetermined value; and
setting the second threshold value as a value between the first predetermined value and the second predetermined value.

8. The method of claim 7 , wherein the second threshold value is a mean value of the first predetermined value and the second predetermined value.

9. The method of claim 7 , wherein the classification algorithm is k-means algorithm.

10. The method of claim 7 , wherein step D comprises the step of:
adding a plurality of fourth elements in the third array, wherein a number of the plurality of fourth elements is equal to a number of the third elements; a value of each of the plurality of fourth elements is less than any one of values of the third elements.

11. The method of claim 10 , wherein the value of each of the plurality of fourth elements is 0.

12. The method of claim 1 , wherein step B comprises the steps of:
Taking a target area in a second image axial direction perpendicular to the first image axial direction in the image, wherein the target area has a plurality of vertical pixels groups which are arranged side by side along the first image axial direction;
calculating a mean value of grayscale values of each of the vertical pixels groups along the first image axial direction in the target area, to obtain the grayscale values of the plurality of pixels of the first image axials direction, and thus to form the first array.

13. The method of claim 12 , wherein the target area is a middle area in the second image axial direction of the image.

14. The method of claim 13 , wherein in the step B, the image is divided into a first area, a second area, and a third area in order on the second image axial direction; the second area forms the target area; along the second image axial direction, a number of pixels of the second area accounts for 30% to 50% of a total number of pixels from the first area to the third area.

Assignees

• Globalwafers Co Ltd

Inventors

• Chang Wei-Cheng
• Lee Chia-Yeh
• Wu han-zong

Classifications

• G06V10/762

  using clustering, e.g. of similar faces in social networks · CPC title

• G06F18/243

  relating to the number of classes · CPC title

• G06F18/23213

  with fixed number of clusters, e.g. K-means clustering · CPC title

• G06F18/24

  Classification techniques · CPC title

• G06F18/23

  Clustering techniques · CPC title