KR20190042180A - Cover-glass analyzing method - Google Patents

Abstract

The present invention relates to a cover glass analyzing method and, specifically, to a cover glass analyzing method for quickly performing processing on an image of a photographed cover glass so as to quickly decide a defect included in the cover glass. The cover glass analyzing method of the present invention comprises: a step of extracting blobs from a pre-processed binarized image of the cover glass; a step of controlling the number of blobs by determining the blob to be post-processed on the basis of the size and number of the extracted blobs; a step of calculating basic data for the adjusted number of blobs; a step of calculating a stain index, which is a brightness ratio between the blob and a background image, to determine the noise; and a step of classifying each blobs.

Description

{COVER-GLASS ANALYZING METHOD}

The present invention relates to a cover glass analysis method, and more particularly, to a cover glass analysis method for quickly performing processing on an image of a captured cover glass in order to quickly determine a defect included in a cover glass.

As the IT industry develops, portable electronic products such as smart phones and tablet PCs, and LCD monitors, and household appliances such as TVs are increasingly used. Research is also actively proceeding.

If a blob such as a foreign object or a stain is included on the surface of such a cover glass, the light generated from the backlight module is distorted. Therefore, the process proceeds to the commercialization stage and the display screen becomes blurry or burnt The brightness non-uniformity with the phenomenon appears, which leads to a deterioration of the merchantability and a user's inconvenience, and at the same time, there is no method of repairing the quality of the cover glass surface in the manufacturing process. Therefore, it is necessary to confirm the defects of the cover glass before delivery.

At this time, the unevenness may be caused by dust and foreign matter during the manufacturing process, scratches caused by the carelessness of the inspector, fingerprints, and the like. Until now, these stains have been directly inspected by the naked eye for the inspection of the stain on the cover glass. However, when the inspection is performed by the naked eye, there is a disadvantage that not only the same standards are not completely perfect for each person but also labor and time are required.

To overcome these drawbacks, machine vision systems that automatically detect defects using image processing are becoming increasingly important. Machine vision systems are systems that automatically inspect these cover glass surfaces,

An image acquisition step of capturing an image of a product to be inspected by a non-human computer and importing the image to a computer, an image processing step of performing processing necessary for the input image using the software, The surface of the cover glass is automatically inspected through the image analyzing step to determine whether or not the cover glass surface exists. Applying a professional - level detection algorithm using such a machine vision system is expected to have a great effect on the time and cost side rather than detecting human direct defects.

The conventional automatic blotch defect detection system and method, which is Korean Patent Registration No. 10-1677070, detects the blot using morphological image processing, but the process of detecting each defect is complicated and time consuming, There is a problem that it takes a considerable time to proceed with the data processing without consideration. In addition, the above-mentioned prior art patent has a problem that defects having a small influence on the display quality of the electronic device on which the cover glass is mounted are determined as defects, and the time required for the defect analysis is considerably long.

According to the present invention, a blob is restored (extracted) from a pre-processed image by a simpler method, a progress of post-processing is judged according to the number and size of defects, and a blob index is calculated, The present invention provides a cover glass analysis method that can reduce the amount or process of data to be processed as a whole.

A method for analyzing a cover glass according to the present invention includes the steps of extracting a blob from a pre-processed binarized image of a cover glass, determining a blob to be post-processed based on the size and number of extracted blobs to adjust the number of blobs, Calculating basic data for the blobs of the blobs of the blobs, calculating a blob index which is the ratio of the brightness between the blobs and the background image to determine the noise, and classifying each blob.

According to the present invention, a blob is restored (extracted) from a pre-processed image by a simpler method, a progress of post-processing is judged according to the number and size of defects, a blob index is calculated, It is possible to reduce the amount and process of the data to be processed as a whole.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram of a cover glass analysis apparatus for carrying out a cover glass analysis method according to the present invention. FIG.
2 is a flowchart of a method for analyzing a cover glass according to the present invention.
3A to 3E are images or data processed in step S1 of FIG.
4A to 4C are explanatory diagrams of the process in step S7 of FIG.

Hereinafter, the present invention will be described in detail with reference to the embodiments and the drawings.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram of a cover glass analysis apparatus for carrying out a cover glass analysis method according to the present invention. FIG.

The cover glass analysis apparatus according to the present invention includes an input unit 1 for obtaining input from a user (for example, start / end of line scan, power on / off, etc.) (3) for displaying an image of the object (for example, the position / size / number of foreign objects) of the cover glass, and a cover glass An image pickup section 7 for acquiring an image of a cover glass while maintaining a reference angle in a cover glass, and an image pickup section 7 for moving the image pickup section 7 at different positions The first to third illuminators 9a to 9c for irradiating light at different angles in the cover glass 5 and the cover glass 5 to control the above described components and to control the conveying unit 5 to be placed on the conveying unit 5 and / (7) and the first to third And sequentially acquires the image of the cover glass by the light from the first to third illuminators 9a to 9c from the image pickup section 7 by sequentially controlling the operations of the lights 9a to 9c, And a control unit 20 for determining whether foreign substances are present on the outer surface or inside. The power supply unit (not shown), the input unit 1, the display unit 3, the transfer unit 5, the image pickup unit 7, and the like, which supply power to the respective components of the cover glass analysis apparatus, Is a well-known technology to those skilled in the art, and detailed description thereof is omitted.

The first to third illuminators 9a to 9c are for illuminating the outside and inside of the cover glass. For example, they irradiate light at different angles to the cover glass at different angles. The first illumination 9a is perpendicular to the surface of the cover glass or the direction of movement of the cover glass or the moving direction of the imaging region of the imaging unit 7, The upper side of the glass). The second illumination 9b is positioned at the same position as the imaging section 7 (for example, on the upper side of the cover glass) while maintaining a certain angle (for example, less than 90 degrees) on the surface of the cover glass. The third illuminator 9c is disposed at a position symmetrical with respect to the cover glass 7 relative to the image sensing unit 7 while being perpendicular to the surface of the cover glass or the direction of movement of the cover glass or the direction of movement of the image sensing area of the image sensing unit 7 For example, the lower side of the cover glass).

The control unit 20 acquires an image of the cover glass at different angles and positions from the imaging unit 7 by sequentially operating only one of the first to third illuminations 9a to 9c. In addition, the control unit 20 processes the obtained image to determine whether there is a defect including a blob or the like in the cover glass.

In addition, the control unit 20 uses a preprocessing process such as noise reduction for the original image of the obtained cover glass and an image in which binarization is completed.

2 is a flowchart of a method for analyzing a cover glass according to the present invention. Hereinafter, the control unit 20 processes the binarized image (the binarized image) that has been pre-processed by a conventionally known method.

In step S1, the control unit 20 performs data conversion in units of blobs in the binarized image. FIG. 3A is an original image taken by the imaging unit 7, and FIG. 3B is a binarized image in which the control unit 20 performs preprocessing and completes the binarization. Figures 3a and 3b show cases where blobs B1 and B2 are included. The control unit 20 processes the binarized image in a contour manner to obtain five points B11, B12, B13 and B14 of the blob B1 and five points B21, B22, B23, B24, B25) and stores the coordinate data. Also, the controller 20 processes the binarized images in the LABELLING format and assigns the same numbers to the same bubbles B1 or B2, as shown in FIG. 3D, And stores the labeled image in which the number B2 is differently numbered. Next, the control unit 20 determines the number in the labeled image corresponding to the coordinates of at least one point of each of the bubbles Bl and B2 in the coordinate data, and outputs the labeled image with only the determined number AND operation to extract only each of the bubbles (B1) and (B2). 3E, when extracting the block B1, the control unit 20 determines whether the image including the labeled image LImage and the image containing only the number (i.e., 1) corresponding to the coordinate data of the block B1 (B1LImage) and extracts a block (B1) image including only the block (B1). In the case of extracting the block B2, an image including only the number (that is, 2) corresponding to the coordinate data of the block B2 may be used. In the above-described manner, the control unit 20 extracts all blobs included in the binarized image.

In step S3, the control unit 20 performs post-processing of the binarized image only when the number of blobs larger than the reference size is equal to or less than the reference number, in order to increase the operation speed in the post-processing. That is, the control unit 20 does not perform the post-processing so that the number of blobs larger than the reference size exceeds the reference number, Does not perform post-processing. That is, the control unit 20 adjusts the number of blobs to proceed to step S5 only for the binarized images including the number of blobs larger than the reference size among the binarized images not larger than the reference number. In step S3, the control unit 20 determines whether to post-process the binarized image based on the number of blobs extracted in step S1 and the size of each blob. The control unit 20 calculates the size of each blob using coordinate data. For example, when the minimum value is subtracted from the maximum value in the x direction, the width is calculated. When the minimum value is subtracted from the maximum value in the y direction, the height is calculated. Therefore, the controller multiplies the width and height to calculate the area (size).

In step S5, the control unit 20 calculates basic data (e.g., blob size, area, density, and the like) for classifying the blobs according to their characteristics in step S9.

In step S7, the control unit 20 calculates a blob index to additionally remove blobs that can not be perceived by the human eye. First, the control unit 20 expands each bubble at a predetermined ratio to generate a background mask, and calculates the brightness (average value) FI (FI) of each bubble in the original image and the background (Average value of brightness) (BI) of the portion corresponding to the brightness index (I) is calculated, and the speckle index (I) is calculated according to the following Expression (1).

If the blob index I is less than or equal to the reference exponent value, the control unit 20 classifies the blob as noise and, if the blob exceeds the reference exponent value, classifies the blob type in step S9.

More specifically, the control unit 20 expands the blob image as shown in FIG. 4A to generate a background image as shown in FIG. 4B. The controller 20 extracts the brightness of the portion UB included in the background image but outside the blob image from the original image as shown in FIG. 4C.

In step S9, in addition to the noise classification processing in step S7, the control unit 20 stores blobs, which are not noise, classified as point defects, linear defects, etc. according to their characteristics.

It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or scope of the general inventive concept as defined by the appended claims. It is to be understood that modifications are possible and that such modifications are within the scope of the claims.

1: input unit 3: display unit
5: transfer unit 7:
9a, 9b, 9c: first to third illumination 20:

Claims (1)

Extracting the blob from the pre-processed binarized image of the cover glass; Determining the number of blobs to be processed based on the size and number of extracted blobs;
Computing basic data for the adjusted number of blobs;
Calculating a speckle index which is a ratio of brightness between the blob and the background image to determine noise;
And classifying each bubble. ≪ RTI ID = 0.0 > 11. < / RTI >

Images