CN110108584B - Device and method for automatically detecting hardness of ceramic tile - Google Patents

Abstract

The invention discloses a device and method for automatically detecting the hardness of ceramic tiles. The device comprises: a grinding wheel grinder is installed on one side of a conveyor belt, an industrial camera is installed on the other side, an industrial control computer receives and analyzes the signal of the sensor of the industrial camera; a cleaning device is installed on the In the middle of the conveyor belt; use the hardest diamond groove grinding wheel grinder to grind the tiles with two grooves, and use the tile cleaning device to clean the polished tiles; use four bar light sources to be located around the tiles And the relative position of the tiles is vertical; the information is analyzed and calculated through industrial cameras and industrial computers, and the hardness of the tiles is accurately reflected in real time. The method includes: using wavelet noise reduction to denoise the image, and automatically obtaining the optimal threshold by calculating the information entropy; regularizing and approximating the identified groove image to obtain a more accurate groove length; The detected groove length determines the hardness of the tile.

Description

Device and method for automatic detection of ceramic tile hardness

technical field

The invention relates to the field of automatic detection, in particular to a device and method for automatically detecting the hardness of ceramic tiles.

Background technique

For ceramic tile manufacturers, it is an important task to detect the hardness of the tiles produced in real time. The hardness level of the tiles directly determines whether the tiles are qualified and the use of the tiles.

At present, the detection methods and standards for measuring ceramic tiles in my country are not scientific and reasonable enough. The commonly used method is to use standard modules to carve the glaze and observe which level of module the glaze is damaged by; or beat the tiles to judge the hardness of the tiles by sound. Among them, the standard module engraving method, due to the difference in the strength of different inspectors, even if the same standard module is used to engrave the same tile, different inspectors will get different inspection conclusions. The tile tapping method can only be vaguely judged by the experience of the inspectors.

In addition, these two methods can only obtain general detection results, and cannot classify ceramic tiles in detail, which brings inconvenience to manufacturers, inspectors and users of ceramic tiles.

SUMMARY OF THE INVENTION

The invention provides a device and method for automatically detecting the hardness of ceramic tiles. The equipment provided by the invention is simple, the detection is fast, the maintenance cost is low, and the real-time detection of the hardness of ceramic tiles is realized. See the following description for details:

A device for automatically detecting the hardness of ceramic tiles, the device comprises: a groove grinding device, a cleaning device, a bar light source, an industrial camera, an industrial control computer, and three position sensors;

Among them, the grinding wheel grinder is installed on one side of the conveyor belt, the industrial camera is installed on the other side of the conveyor belt, and the industrial computer receives and analyzes the signal of the industrial camera sensor; the cleaning device is installed in the middle of the conveyor belt;

Use the diamond groove grinding wheel grinder with the hardest strength to grind the tiles with two grooves, and use the tile cleaning device to clean the polished tiles;

Use four bar light sources to be located around the tiles and vertically to the relative position of the tiles;

The information is analyzed and calculated through industrial cameras and industrial computers, and the hardness of the tiles is accurately reacted in real time.

Among them, the position of the industrial camera and the tile surface is kept fixed, and the distance sensor is used to ensure that the position of the industrial camera and the tile surface is fixed.

A method for automatic detection of ceramic tile hardness, the method comprises the following steps:

Use four strip light sources in different positions to illuminate the tile grooves, and obtain the images after grinding the grooves through an industrial camera and an industrial computer;

According to the image obtained by the industrial camera, use wavelet noise reduction to denoise the image, and automatically obtain the best threshold by calculating the information entropy;

The identified groove image is regularized and approximated to obtain a more accurate groove length; the hardness of the tile is judged according to the detected groove length.

The process of regularizing and approximating the identified groove image is as follows:

至

作为该条瓷砖凹痕l₁；A regular tile dent can be obtained by redrawing the graph through four boundary points.

to

as the tile dent l ₁ ;

Install a laser distance sensor on the industrial camera, continuously adjust the distance from the industrial camera to the surface of the tile through the laser distance sensor, and fix the distance from the industrial camera to the tile as d;

The correspondence table between the position of the pixel points on the image and the actual distance is constructed, and the pixel correspondence table is designed according to the size of the tile, and the detected groove length is converted into the real length.

The beneficial effects of the technical scheme provided by the present invention are:

1. The present invention uses the slotted sheet of diamond material to carry out groove grinding on ceramic tiles of any hardness, and analyzes and calculates the information through industrial cameras and industrial control computers, and accurately detects the hardness of ceramic tiles in real time;

2. This testing method not only measures the hardness of polished tiles, but also can measure the hardness of glazed tiles;

3. The detection device and the calculation process have small errors, high detection accuracy, high speed, and low maintenance cost.

4. The present invention uses an industrial camera to measure the size, the measurement is at the pixel level, and the measurement accuracy can be accurate to millimeters. Not only the hardness of polished tiles can be measured, but also the hardness of glazed tiles.

Description of drawings

1 is a schematic structural diagram of an automatic detection device for automatic detection of ceramic tile hardness provided by the present invention;

Fig. 2 is the obtained tile image schematic diagram after grinding;

Fig. 3 is the flow chart of noise reduction;

Fig. 4 is the result graph after noise reduction;

Fig. 5 is the flow chart of identifying groove;

Fig. 6 is the result graph of identifying groove;

FIG. 7 is a result diagram of regularization and approximation processing of the results.

In the accompanying drawings, the list of components represented by each number is as follows:

1: tile; 2, 4, 6: position sensor;

3: Groove grinding device; 5: Cleaning device;

7: Identification device; 8: Conveyor belt;

9: Grinding wheel grinding machine; 10: Disc shaped brush;

11: Rinse water pipe; 12: Industrial camera;

13: Bar light source; 14: Distance sensor;

15: Industrial computer.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the embodiments of the present invention are further described in detail below.

Example 1

The embodiment of the present invention provides an automatic detection device for automatic detection of tile hardness. Referring to FIG. 1, the detection device includes: a groove grinding device 3, a cleaning device 5, a bar light source 13, an industrial camera 12, an industrial computer 15, three Position sensor (represented by the labels 2, 4, 6);

The grinding wheel grinder 9 is installed on one side of the conveyor belt 8 , the industrial camera 12 is installed on the other side of the conveyor belt 8 , and the industrial computer 15 receives and analyzes the signals of the industrial camera sensor 12 ; the cleaning device 5 is installed in the middle of the conveyor belt 8 .

Preferably, there are two grinding wheel grinders 9, one industrial camera 12, four bar light sources 13, and three position sensors.

In specific implementation, the ceramic tile 1 is subjected to groove grinding with a diamond groove grinding wheel grinder 9 having the hardest strength. In order to obtain a recognition result with higher accuracy, two groove grinding treatments are performed on the ceramic tile 1 . Since the tile 1 has some stains after being polished, these stains will inevitably have a certain impact on the identification of the grooves in the later stage, so a tile cleaning device 5 is provided. For example, the polished tiles 1' can be cleaned by using the flushing water pipe 11 and the disc special-shaped brush 10.

During the specific implementation, considering that the purpose of the embodiment of the present invention is to identify the length of the groove, in order to highlight the groove more clearly, four strip light sources 13 are used, which are respectively located around the tile 1" and in a relative position to the tile 1". placed vertically.

In specific implementation, since the embodiment of the present invention uses a monocular camera to measure the length of the groove, in order to accurately measure the length of the groove, it is necessary to keep the position of the industrial camera 12 and the surface of the ceramic tile fixed. Considering the different thicknesses of the ceramic tiles produced by the production line, In the embodiment of the present invention, the distance sensor 14 is used to ensure that the position of the industrial camera 12 and the surface of the tile is fixed.

To sum up, in the embodiment of the present invention, the grooved sheet of diamond material can be used for groove grinding on tiles of any hardness, and the information is analyzed and calculated through an industrial camera and an industrial computer, and the hardness of the tiles can be accurately reflected in real time. , the device and the calculation process have small errors, fast detection speed and low maintenance cost.

Example 2

The embodiment of the present invention provides an automatic detection method for automatically detecting the hardness of ceramic tiles, and the method includes the following steps:

101: Grinding the grooves of the tile 1 to be detected;

Wherein, the step 101 includes: initializing the equipment, and grinding the tile 1 with two grooves by using a grooved sheet made of diamond material.

102: Use clean water and a disc special-shaped brush to clean the polished tiles;

103: Use four strip light sources 13 at different positions to perform illumination processing on the tile grooves, and obtain images after grinding the grooves through the industrial camera 12 and the industrial control computer; (as shown in FIG. 2 );

104: According to the image obtained by the industrial camera 12, apply an image processing method to perform image de-drying processing;

Wherein, this step 104 includes: using the wavelet noise reduction method to denoise the image, and automatically obtaining the optimal threshold by calculating the information entropy; Precise groove length.

105: Identify the groove length of tile 1 using the designed length measurement technology;

That is, it includes: judging the hardness of the tile according to the detected groove length.

106: Obtain the hardness of the tile according to the identified groove length.

As shown in FIG. 3 , step 104 includes acquiring an image of the polished tile through the black and white industrial camera 12 , and FIG. 2 is an image acquired by the industrial camera 12 after being irradiated by a strip light source 13 ; The obtained image has the characteristics of salt and pepper noise, select db5 as the wavelet function and set the number of decomposition layers to 4 layers; select the improved threshold function to process high-frequency coefficients:

为处理后的小波系数，w_j，k为原始信号，λ为阈值。λ采用不同层次上不同阈值的计算方法：in,

is the processed wavelet coefficient, w _{j, k} is the original signal, λ is the threshold. λ adopts the calculation methods of different thresholds at different levels:

Among them, δ ² is the variance of the noise, N is the length of the sampled signal, and j is the number of decomposition layers.

An inverse transformation is performed on the processed high-frequency coefficients to obtain a noise reduction diagram as shown in FIG. 4 .

As shown in FIG. 5 , step 105 includes: for the denoised image, using the method of binarization threshold to identify the length of the polished groove, wherein the calculation method of the binarization threshold is:

(1) Calculate the probability of each gray value appearing in the denoised grayscale image:

In the formula, N and M represent the length and width of the image, respectively, G _m = {0,1,...,(m-1)} represents the gray value that appears in a picture, and f _i represents each gray value The number of occurrences, m is 256.

(2) For any gray value s, the following two normalized parts can be obtained by calculation:

表示从灰度值为0至s所出现的概率和，p_s为灰度值s出现的概率，A、B为所计算获得的两部分信息。In the formula,

It represents the sum of the probability of occurrence from gray value 0 to s, ps is the probability of occurrence of gray value _s , and A and B are the two parts of information obtained by calculation.

(3) For the tile image segmentation problem involved in this study, as shown in Figures 2 and 4, the area where the white grooves to be recognized appear is small; in addition, for the image segmentation problem, the basic idea is to make the target and the background The distinction of s is maximized, so the overall information entropy that can be provided by part A and part B after dividing by s as the threshold is:

In the formula, C _A (s) and C _B (s) are the overall information entropy that can be provided by part A and part B, respectively.

(4) In order to maximize the separation of the tile groove target and the background studied by this method, the selected threshold TC(s) is the maximized value:

Among them, TC(s ^* ) is the maximum value in TC(s).

As shown in Figure 6, the tile grooves can be separated from the background according to the calculated threshold:

至

作为该条瓷砖凹痕l₁。Since the obtained shape is an irregular shape after identification, the obtained shape is regularized. As shown in Figure 7, find the coordinates of the four boundary points of the identified irregular figure A(x _a , y _a ), B(x _b , y _b ), C(x _c , y _c ), D(x _d , y _d ). Redrawing the graph through the four boundary points results in a regular tile dent (white rectangle in Figure 7). select

to

as the tile indent l ₁ .

作为瓷砖1硬度的评价值。Considering the different thicknesses of the tiles to be detected, in order to obtain the actual length of the groove more conveniently, a laser distance sensor 14 is installed on the industrial camera 12 . The distance between the industrial camera 12 and the surface of the tile 1 is continuously adjusted by the laser distance sensor 14, and the distance between the industrial camera 12 and the tile 1 is fixed as d. Since the industrial camera 12 has distortion, in order to reduce the influence of the distortion, Table 1 is constructed to perform the correspondence between the position of the pixel point on the image and the actual distance. According to the size of the tile 1 , the resolution of the industrial camera 12 used in the embodiment of the present invention is 1628×1236, and a 10×10 pixel correspondence table is designed to convert the detected groove length l ₁ into the real length L ₁ . Similarly, the remaining three groove lengths L ₂ , L ₃ , and L ₄ can be obtained by turning on the other bar light sources 13 respectively. Find the average of the four lengths

As the evaluation value of the hardness of the tile 1.

Table 1. Correspondence between camera pixel points and actual length

Table 2 Corresponding relationship between groove length L and hardness obtained by testing

As shown in Table 2, step F includes dividing the hardness level of the tiles 1 produced by the production line according to the correspondence table between the length L of the groove and the hardness. There is no specific standard for the corresponding relationship between the constructed groove length L and the hardness of the tiles. A standard version can be set, and a standard diamond material slotted sheet is used to grind the grooves of tiles of any hardness to obtain a correspondence between the length and the hardness. Develop detailed standard specifications for the division of hardness grades.

To sum up, the embodiments of the present invention provide an automatic detection method for automatic detection of tile hardness. The method uses an industrial camera to measure the size, the measurement is at the pixel level, and the measurement accuracy can be accurate to millimeters. The method not only measures the hardness of polished tiles, but also can measure the hardness of glazed tiles.

In the embodiment of the present invention, the models of each device are not limited unless otherwise specified, as long as the device can perform the above functions.

Those skilled in the art can understand that the accompanying drawing is only a schematic diagram of a preferred embodiment, and the above-mentioned serial numbers of the embodiments of the present invention are only for description, and do not represent the advantages or disadvantages of the embodiments.

The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection of the present invention. within the range.

Claims (3)

1. A method for automatically detecting the hardness of a tile, characterized in that it comprises the following steps:

using four strip light sources at different positions to perform illumination treatment on the ceramic tile groove respectively, and obtaining an image obtained after polishing the groove through an industrial camera and an industrial control computer;

according to an image obtained by an industrial camera, performing noise reduction processing on the image by using wavelet noise reduction, and automatically obtaining an optimal threshold value in a mode of calculating information entropy;

regularization and approximation processing are carried out on the identified groove image, and the regularization and approximation processing specifically comprises the following steps:

obtaining and identifying coordinates A (x) of four boundary points of the irregular figure_a,y_a),B(x_b,y_b),C(x_c,y_c),D(x_d,y_d) Redrawing the graph line through the four boundary points to obtain a regular tile dent, and selecting

To

As the length l of the groove of the strip of tiles₁；

Installing a laser distance sensor on the industrial camera, continuously adjusting the distance from the industrial camera to the surface of the ceramic tile through the laser distance sensor, and fixing the distance from the industrial camera to the ceramic tile as d;

constructing a corresponding relation table of the positions of the pixel points on the image and the actual distances, designing a pixel corresponding table according to the size of the ceramic tile,

the detected length l of the groove₁Converted into a true length L₁Respectively turning on other strip light sources to obtain the lengths L of the other 3 grooves₂、L₃、L₄Calculating the average of four lengths

As an evaluation value of the tile hardness;

according to the corresponding relation table of the length L of the groove and the hardness, the hardness grades of the tiles produced by the production line are divided, the groove of the tiles with any hardness is polished by the grooving sheet made of standard diamond materials to obtain the correspondence between the length and the hardness, and the hardness grades are divided.

2. A method for the automated detection of the hardness of ceramic tiles according to claim 1, characterized in that the devices applied in said method comprise: the device comprises a groove polishing device, a cleaning device, a strip-shaped light source, an industrial camera, an industrial control computer and three position sensors;

the industrial control computer receives and analyzes signals of the industrial camera sensor; the cleaning device is arranged in the middle of the conveyor belt;

polishing two grooves on the ceramic tile by using a diamond groove grinding wheel polisher with the hardest strength, and cleaning the polished ceramic tile by using a ceramic tile cleaning device;

four strip-shaped light sources are respectively positioned on the periphery of the ceramic tile and are vertically arranged relative to the ceramic tile;

and analyzing and calculating the information through an industrial camera and an industrial control computer, and accurately reacting the hardness of the ceramic tile in real time.

3. The method for automatically detecting the hardness of the ceramic tiles as claimed in claim 2,

the industrial camera and the surface position of the ceramic tile are kept fixed, and the industrial camera and the surface position of the ceramic tile are guaranteed to be fixed and unchanged by using the distance sensor.

Images