A method and system for detecting surface quality of precious metal products

Abstract

The present invention relates to the technical field of precious metal quality detection, and more specifically, to a method and system for detecting the surface quality of precious metal products, including: collecting a grayscale image of a precious metal product; obtaining a first area in the grayscale image where the surface consistency of the precious metal product needs to be detected and a second area where the surface defects of the precious metal product need to be detected; for the first area, dividing the first area into a preset number of detection areas. The present invention uses an image processing method to detect whether the surface quality of the precious metal is qualified, which greatly avoids the on-site experience of the staff and the low efficiency, is easily affected by the on-site environment, and is prone to missed detection and false detection. And through the method of judging whether the surface consistency of the precious metal product and the surface defects of the precious metal product are qualified by the present invention, it is finally determined whether the surface quality of the precious metal is qualified, which improves the accuracy of the detection and improves the work efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of precious metal quality detection, and more specifically, to a method and system for detecting the surface quality of precious metal products. Background Art

[0002] Precious metals refer to metals with small reserves in the earth's crust and are difficult to mine and extract. They are more expensive than general metals, hence the name. In the production process of precious metal products, due to the influence of raw materials, equipment, technology and other factors, different types of defects such as scratches, roller marks, inclusions, pits, black spots, poor consistency, etc. appear on the surface. These defects not only seriously affect the appearance of the product, but also reduce the product quality, bring adverse effects to the development of the enterprise, and cause property losses to customers. It is of great significance to detect the surface quality of precious metal products and find defects in time, so as to provide a basis for process adjustment, equipment improvement and product quality assurance.

[0003] In the related technology, for example, the Chinese patent document with publication number CN214771895U discloses a precious metal detection and inspection workbench, which is provided with a rotating disk and a cylinder. The cylinder drives the movable plate to slide, so that the movable plate drives the gear to rotate through the rack on its surface, thereby driving the rotating disk and the fixture to rotate, so that the precious metal workpiece can be flipped during detection, which is convenient for the staff to carry out detection and improves the detection efficiency.

[0004] At present, most of the surface quality inspections of precious metal products are done manually, which relies on the on-site experience of personnel and is inefficient. In addition, the staff are easily affected by the on-site environment, which can easily lead to missed inspections and false inspections, resulting in low accuracy in the surface quality inspections of precious metal products. Summary of the invention

[0005] The present invention provides a method and system for detecting the surface quality of precious metal products, aiming to solve the problem in the related art that the surface quality of precious metals is detected manually, which relies on the on-site experience of personnel and is inefficient, and the personnel are easily affected by the on-site environment, which is prone to missed detection and false detection, resulting in low accuracy of the surface quality detection of precious metal products.

[0006] In a first aspect, the present invention provides a method for detecting the surface quality of precious metal products, comprising: acquiring a grayscale image of the precious metal product; obtaining a first area in the grayscale image where the surface consistency of the precious metal product needs to be detected and a second area where the surface defects of the precious metal product need to be detected; for the first area, dividing the first area into a preset number of detection areas, and obtaining the grayscale mean, grayscale root mean square difference and roughness value of each detection area based on a deep learning model, and calculating the difference coefficient of the first area , the calculation formula is: ; In the formula, Indicates The grayscale mean of the detection area is Indicates The grayscale mean of the detection area is Indicates The roughness value of the detection area, Indicates The roughness value of the detection area, Indicates The root mean square difference of grayscale in the detection area is is the upper limit of the grayscale root mean square error of the detection area, Represents the total number of detection areas, represents a standard normalization function; based on the trained defect detection model, the defect type and the area corresponding to the defect in the second area are obtained, and the defect degree of the second area is calculated, and the defect degree of the second area and the product of the total number of defect types in the second area and the total defect area are positively correlated; based on the size of the sum of the defect degree and the difference coefficient, determine whether the quality of the precious metal is qualified.

[0007] The effect is that the image processing method is used to detect whether the surface quality of precious metals is qualified, which greatly avoids the on-site experience and low efficiency of the staff, is easily affected by the on-site environment, and is prone to missed detection and false detection.

[0008] Furthermore, the defect degree of the second region is calculated using the following formula: ; In the formula, represents the defect degree of the second region, M represents the total number of defect types in the second region, represents the total defect area in the second region, Represents the standard normalization function.

[0009] The effect is that the above method can be used to calculate the defect degree of the second area, and the result of the defect degree of the second area can be normalized to facilitate subsequent calculations.

[0010] Furthermore, based on the size of the sum of the defect degree and the difference coefficient, determining whether the quality of the precious metal is qualified includes: in response to the sum of the defect degree and the difference coefficient being less than or equal to a first threshold, the quality of the precious metal is qualified; in response to the sum of the defect degree and the difference coefficient being greater than the first threshold, the quality of the precious metal is unqualified.

[0011] The effect is that through the above method, it is possible to quickly and accurately determine whether the quality of precious metals is qualified.

[0012] Furthermore, it also includes: judging whether the surface consistency of the precious metal product is qualified according to the size of the difference coefficient of the first area; judging whether the surface defects of the precious metal product are qualified according to the size of the defect degree of the second area; in response to the surface consistency of the precious metal product and the surface defects of the precious metal product being qualified, the quality of the precious metal is qualified.

[0013] The effect is that the surface consistency of precious metal products and the surface defects of precious metal products are judged separately, which improves the accuracy of judgment and can more accurately determine the quality of precious metals.

[0014] Furthermore, whether the surface consistency of the precious metal product is qualified is judged based on the size of the difference coefficient of the first area, including: in response to the difference coefficient of the first area being less than or equal to a second threshold, the surface consistency of the precious metal product is determined to be qualified.

[0015] Further, judging whether the surface defects of the precious metal product are qualified is performed based on the degree of defects in the second area, including: in response to the degree of defects in the second area being less than or equal to a third threshold, the surface defects of the precious metal product are determined to be qualified.

[0016] Further, judging whether the surface consistency of the precious metal product is qualified also includes: in response to the grayscale mean, grayscale root mean square difference and roughness value of the detection area meeting the preset conditions, determining that the surface consistency of the precious metal product is qualified, wherein the preset conditions are:

[0017] ;

[0018] In the formula, Indicates The grayscale mean of the detection area is is the first standard difference value, Indicates The roughness value of the detection area, represents the second standard difference value, Indicates The root mean square difference of grayscale in the detection area is is the upper limit of the grayscale root mean square difference of the detection area, and n represents the total number of detection areas.

[0019] The effect is that the above method is used to determine whether the surface consistency of the precious metal product is qualified, thereby improving the quality requirements of the inspection.

[0020] Furthermore, the empirical value of the first threshold is 0.5.

[0021] In a second aspect, the present invention also provides a precious metal product surface quality detection system, comprising a processor and a memory, wherein the memory stores a computer program, and the processor executes the computer program to implement any of the above-mentioned precious metal product surface quality detection methods.

[0022] Beneficial effects: The image processing method is used to detect whether the surface quality of precious metals is qualified, which greatly avoids the on-site experience of staff and low efficiency, and is easily affected by the on-site environment, and is prone to missed detection and false detection. And through the method of judging whether the surface consistency of precious metal products and the surface defects of precious metal products of the present invention are qualified, it is finally determined whether the surface quality of precious metals is qualified, which improves the accuracy of detection and improves work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The following detailed description is read with reference to the accompanying drawings, which illustrate several embodiments of the present invention in an exemplary and non-limiting manner, and in which like or corresponding reference numerals represent like or corresponding parts, wherein:

[0024] Figure 1 is a schematic diagram schematically showing the structure of a detection device according to an embodiment of the present invention;

[0025] Figure 2 FIG. 4 is a flow chart schematically illustrating a method for determining whether the surface quality of a precious metal is acceptable according to an embodiment of the present invention.

[0026] Reference numerals

[0027] 11. Base platform; 12. Support; 13. Workbench bracket; 14. Lens assembly; 15. Rotating platform; 16. LED fill light. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0029] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0030] like Figure 1As shown, the surface quality inspection device for precious metal products includes a base platform 11, a support member 12 and a workbench bracket 13. A lens assembly 14 that can be rotated is provided on the workbench bracket 13 for capturing images of the side of the precious metal product. The support member 12 is vertically fixed on the base platform 11, and the workbench bracket 13 is movably connected to the support member 12. Specifically, the workbench bracket 13 is detachably mounted on the support member 12 by bolts, and the shape of the support member 12 is circular, so that the workbench bracket 13 can be rotated or moved up and down on the support member 12. The position of the lens assembly 14 is adjusted to facilitate capturing images of precious metal products.

[0031] A rotating platform 15 is provided on the base platform 11, and the rotating platform 15 is embedded in the base platform 11, so that the rotating platform 15 device can rotate clockwise or counterclockwise, and a plurality of LED fill lights 16 are embedded in the rotating platform 15, and the plurality of LED fill lights 16 are evenly distributed on the upper edge of the rotating platform 15. In addition, an LED fill light 16 is also provided on the side of the lens assembly 14 close to the rotating platform 15. It should be noted that a polarizing filter for reducing or eliminating reflection is also provided on the lens assembly 14. Through the above structure, the problem of uneven illumination and severe reflection causing unclear photos when acquiring images is solved. The above device is used to detect the surface quality of precious metal products, and the specific steps are as follows.

[0032] like Figure 2 As shown, S101: collecting a grayscale image of a precious metal product.

[0033] In one embodiment, the precious metal product to be inspected is placed on the rotating platform 15, and the brightness of the LED fill light 16 on the rotating platform 15 and the lens assembly 14 is adjusted, and then the magnification of the lens assembly 14 is adjusted until the image of the precious metal product is clearly displayed on the display.

[0034] In one embodiment, after placing the precious metal product to be inspected on the rotating platform 15, the rotating platform 15 is rotated clockwise or counterclockwise to a preset angle, and then the lens assembly 14 is used to photograph the precious metal product at the preset angle. The above steps are performed along the same rotation direction to photograph the precious metal product. images, of which , is a preset angle. In this embodiment, is 15 degrees. In other embodiments, It can be 20 degrees or 30 degrees, etc. Finally, the above The images are spliced ​​into a complete image, which is used as the image of the precious metal product. The image of the precious metal product is gray-scaled to obtain a gray-scale image of the precious metal product.

[0035] S102: Performing region detection on the grayscale image.

[0036] In one embodiment, after obtaining the grayscale image of the precious metal product, the grayscale image is divided into two areas. Specifically, the grayscale image is divided into two areas, including: dividing the area in the grayscale image where the surface consistency of the precious metal product needs to be detected into a first area, and dividing the area where the surface defects of the precious metal product need to be detected into a second area.

[0037] S103: Calculate the coefficient of difference of the first area.

[0038] In one embodiment, for the first area, the first area is divided into a preset number of detection areas. Specifically, for the division of the first area, the first area can be divided into equal parts or divided into the first area according to a window of a preset size to obtain a plurality of detection areas. In this embodiment, the preset window size is In other embodiments, the preset window size may be or wait.

[0039] In one embodiment, after the detection area is obtained, the grayscale mean, grayscale root mean square difference and roughness value of each detection area are obtained based on the deep learning model, and the difference coefficient of the first area is calculated according to the grayscale mean, grayscale root mean square difference and roughness value of each detection area.

[0040] In one embodiment, the difference coefficient of the first region is calculated using the following formula: ; In the formula, Indicates The grayscale mean of the detection area is Indicates The grayscale root mean square difference of the detection area is Indicates The grayscale mean of the detection area is Indicates The roughness value of the detection area, Indicates The root mean square difference of grayscale in the detection area is is the upper limit of the grayscale root mean square error of the detection area, Represents the total number of detection areas, represents the standard normalization function. Among them, the difference coefficient of the first region The smaller the difference, the higher the surface consistency of the precious metal product. The larger the value, the lower the surface consistency of the precious metal product.

[0041] S104: Calculate the defect degree of the second area.

[0042] In one embodiment, based on the trained defect detection model, the defect type and the area corresponding to the defect in the second area are obtained, and the defect degree of the second area is calculated. The defect degree of the second area and the product of the total number of defect types in the second area and the total defect area are positively correlated.

[0043] In one embodiment, the defect degree of the second region is calculated using the following formula:

[0044] ;

[0045] In the formula, represents the defect degree of the second region, M represents the total number of defect types in the second region, represents the total defect area in the second region, represents the standard normalized function. Among them, the defect degree of the second area The larger it is, the greater the surface defects of the precious metal product.

[0046] S105: Determine whether the quality of the precious metal is qualified.

[0047] In one embodiment, whether the quality of the precious metal is qualified is determined based on the size of the sum of the defect degree and the difference coefficient. Specifically, if the sum of the defect degree and the difference coefficient is less than or equal to the first threshold, the quality of the precious metal is qualified; if the sum of the defect degree and the difference coefficient is greater than the first threshold, the quality of the precious metal is unqualified. In this embodiment, the empirical value of the first threshold is 0.5. In other embodiments, the empirical value of the first threshold can be 0.55 or 0.6, etc. The empirical value of the first threshold can be determined according to actual conditions.

[0048] In the second embodiment, the following method can also be used to determine whether the quality of the precious metal is qualified. Specifically, the surface consistency of the precious metal product is judged to be qualified according to the size of the difference coefficient of the first area; the surface defects of the precious metal product are judged to be qualified according to the size of the defect degree of the second area; if the surface consistency and surface defects of the precious metal product are both qualified, the quality of the precious metal is qualified.

[0049] For judging whether the surface consistency of the precious metal product is qualified based on the size of the difference coefficient of the first area, it includes: if the difference coefficient of the first area is less than or equal to the second threshold, the surface consistency of the precious metal product is determined to be qualified; in this embodiment, the empirical value of the second threshold is 0.2, in other embodiments, the empirical value of the second threshold can be 0.25 or 0.3, etc., and the empirical value of the second threshold can be determined according to actual conditions.

[0050] For judging whether the surface defects of the precious metal product are qualified according to the size of the defect degree of the second area, it includes: if the defect degree of the second area is less than or equal to the third threshold value, the surface defects of the precious metal product are determined to be qualified. In this embodiment, the empirical value of the third threshold value is 0.2. In other embodiments, the empirical value of the third threshold value can be 0.3 or 0.35, etc. The empirical value of the third threshold value can be determined according to the actual situation. By the above method, the quality of the precious metal product can be determined to be qualified only when it is judged that the surface consistency of the precious metal product and the surface defects of the precious metal product are both qualified. This method can meet the situation where the quality requirements of precious metal products are high.

[0051] In the third embodiment, judging whether the surface consistency of the precious metal product is qualified according to the size of the difference coefficient of the first area also includes: comparing the grayscale mean, grayscale root mean square difference and roughness value of each detection area, and when all detection areas meet the preset conditions, it is determined that the surface consistency of the precious metal product is qualified. The preset conditions are:

[0052] .

[0053] In the formula, Indicates The grayscale mean of the detection area is is the first standard difference value, Indicates The roughness value of the detection area, represents the second standard difference value, Indicates The root mean square difference of grayscale in the detection area is is the upper limit of the grayscale root mean square error of the detection area, Indicates the total number of detection areas.

[0054] The empirical value of the first standard difference value is 0.2. In other embodiments, the empirical value of the first standard difference value may be 0.24 or 0.26, etc., and the empirical value of the first standard difference value may be determined according to actual conditions; the empirical value of the second standard difference value is 0.18. In other embodiments, the empirical value of the second standard difference value may be 0.2 or 0.32, etc., and the empirical value of the second standard difference value may be determined according to actual conditions. The empirical value of the upper limit of the grayscale root mean square difference is 30. In other embodiments, the empirical value of the upper limit of the grayscale root mean square difference may be 20 or 50, etc., and may be determined according to actual conditions.

[0055] In the third embodiment, the surface quality of the precious metal is determined to be qualified only when the surface consistency of the precious metal product is determined to be qualified according to the above method, and the surface defects of the precious metal product are also qualified. The use of this method can further improve the quality of surface inspection of precious metal products.

[0056] Through the above steps, the image processing method is used to detect whether the surface quality of the precious metal is qualified, which greatly avoids the on-site experience of the staff and the low efficiency, and is easily affected by the on-site environment, and is prone to missed detection and false detection. And through the method of judging whether the surface consistency of the precious metal product and the surface defects of the precious metal product are qualified by the present invention, it is finally determined whether the surface quality of the precious metal is qualified, which improves the accuracy of the detection and improves the work efficiency.

[0057] The present invention also provides a precious metal product surface quality detection system. The system includes a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a precious metal product surface quality detection method according to the first aspect of the present invention is implemented.

[0058] The system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and will not be described in detail herein.

[0059] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, apparatus or device. For example, a computer-readable storage medium may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random-Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high-bandwidth memory HBM (High-Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device. Any application or module described in the present invention may be implemented using computer-readable / executable instructions that may be stored or otherwise maintained by such a computer-readable medium.

[0060] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0061] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0062] The above-mentioned embodiments only express several implementation methods of the present invention, and the description is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent application. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

Claims

1. A method for detecting the surface quality of precious metal products, characterized in that: include: Collect grayscale images of precious metal products; Acquire a first area in the grayscale image where the surface consistency of the precious metal product needs to be detected and a second area where the surface defects of the precious metal product need to be detected; For the first area, the first area is divided into a preset number of detection areas, and the grayscale mean, grayscale root mean square difference and roughness value of each detection area are obtained based on the deep learning model, and the difference coefficient of the first area is calculated , the calculation formula is: ； In the formula, Indicates The grayscale mean of the detection area is Indicates The grayscale mean of the detection area is Indicates The roughness value of the detection area, Indicates The roughness value of the detection area, Indicates The grayscale root mean square difference of the detection area is is the upper limit of the grayscale root mean square error of the detection area, Represents the total number of detection areas, represents the standard normalization function; Based on the trained defect detection model, the defect type and the area corresponding to the defect in the second area are obtained, and the defect degree of the second area is calculated, and the defect degree of the second area and the product of the total number of defect types in the second area and the total defect area are positively correlated; Whether the quality of the precious metal is qualified is determined based on the size of the sum of the defect degree and the difference coefficient.

2. The method for detecting surface quality of precious metal products according to claim 1, characterized in that: The defect degree of the second area is calculated using the formula: ； In the formula, represents the defect degree of the second region, M represents the total number of defect types in the second region, represents the total defect area in the second region, Represents the standard normalization function.

3. The method for detecting surface quality of precious metal products according to claim 1, characterized in that: Based on the sum of the defect degree and the coefficient of difference, determine whether the quality of the precious metal is qualified, including: In response to the sum of the defect degree and the coefficient of difference being less than or equal to a first threshold, the quality of the precious metal is qualified; In response to the sum of the defect degree and the coefficient of variation being greater than a first threshold, the quality of the precious metal is unqualified.

4. The method for detecting surface quality of precious metal products according to claim 1, characterized in that: Also includes: According to the size of the difference coefficient of the first area, judging whether the surface consistency of the precious metal product is qualified; According to the degree of the defect in the second area, judging whether the surface defect of the precious metal product is qualified; In response to the surface consistency of the precious metal product and the surface defects of the precious metal product being qualified, the quality of the precious metal is qualified.

5. The method for detecting surface quality of precious metal products according to claim 4, characterized in that: According to the size of the difference coefficient of the first area, judging whether the surface consistency of the precious metal product is qualified includes: In response to the coefficient of difference of the first area being less than or equal to a second threshold, the surface consistency of the precious metal product is determined to be qualified.

6. The method for detecting surface quality of precious metal products according to claim 4, characterized in that: According to the degree of the defect in the second area, judging whether the surface defect of the precious metal product is qualified includes: In response to the defect level of the second area being less than or equal to a third threshold, the surface defects of the precious metal product are determined to be qualified.

7. The method for detecting surface quality of precious metal products according to claim 1, characterized in that: Judging whether the surface consistency of precious metal products is qualified also includes: In response to the grayscale mean value, grayscale root mean square difference and roughness value of the detection area meeting the preset conditions, it is determined that the surface consistency of the precious metal product is qualified, wherein the preset conditions are: ； In the formula, Indicates The grayscale mean of the detection area is is the first standard difference value, Indicates The roughness value of the detection area, represents the second standard difference value, Indicates The grayscale root mean square difference of the detection area is is the upper limit of the grayscale root mean square difference of the detection area, and n represents the total number of detection areas.

8. The method for detecting surface quality of precious metal products according to claim 3, characterized in that: The empirical value of the first threshold is 0.

5.

9. A precious metal product surface quality detection system, comprising a processor and a memory, characterized in that: The memory stores a computer program, and the processor executes the computer program to implement the method for detecting the surface quality of precious metal products as described in any one of claims 1-8.

Images