CN103018261B - Appearance fault detecting system and appearance fault detecting method of RFID (Radio Frequency Identification Device) antenna - Google Patents

Abstract

The invention discloses an appearance defect detection method for an RFID antenna, comprising: a pre-calibration step of an imaging device; Antennas take images of them; collect the captured images and perform edge detection on them, thereby obtaining the number of antennas in the detected image and recording information such as their length, width, and center point coordinates; and selectively performing line detection according to requirements. Width detection, pattern disconnection/sticking detection, burr/print contamination detection and other items, so as to obtain the quality conclusion of the detected antenna. The invention also discloses a corresponding detection system. Through the present invention, the quality inspection of the RFID antenna can be performed in a compact and easy-to-operate manner, and the inspection results can be obtained with high efficiency and accuracy, and at the same time, it is convenient for real-time feedback and quality control, so it is especially suitable for the processing and manufacturing of industrialized RFID process.

Description

Appearance defect detection system and method thereof for an RFID antenna

technical field

The invention belongs to the field of RFID manufacturing, and more particularly relates to a system and method for detecting appearance defects of RFID antennas.

Background technique

Video recognition, also known as RFID technology, also known as electronic tags, is a communication technology that identifies specific targets and reads and writes relevant data through radio signals. Its basic components include tags, readers and antennas. The tags are composed of coupling elements and chips. Each tag corresponds to a unique electronic code and is attached to the object to mark the target object; the reader is used to read and write or write tags information, while the antenna is used to transmit radio frequency signals between the tag and the reader. Because RFID technology does not need to establish mechanical or optical contact between the identification system and a specific target, and has the advantages of high-speed identification, resistance to harsh environments, and strong confidentiality, it is widely used in logistics, warehouse management, medical and identification, food processing, and industry. The production data real-time monitoring and quality tracking of the manufacturing process have been widely used in many fields.

In RFID equipment, the antenna, as the main energy transmission functional layer, is an electronic circuit designed according to the functions required by radio frequency identification. It arranges conductive silver paste or conductive carbon paste on the surface by etching, hot stamping or conductive ink printing. PVC, PC or PET antenna substrate, and packaged with the surface layer, protective layer and bottom layer. The processing quality of the wiring pattern of the antenna directly affects whether the signal can be transmitted to the maximum extent between the tag and the reader, and even affects the normal operation of the RFID device.

However, the current high-precision antenna pattern detection equipment is blank at home and abroad. Most of the existing technologies are manually (human eyes directly observe the printing of the antenna substrate) or semi-manual (human eyes observe the printing of the antenna substrate through the camera screen). situation) to perform instruction detection, which is not only inefficient and costly, but also difficult to guarantee the accuracy and consistency of the detection conclusions, and greatly limits the development of the RFID antenna tag production industry. Correspondingly, there is a technical demand for further improvement in the quality detection method of RFID antennas in the related field.

Contents of the invention

In view of the above defects or technical requirements of the prior art, the object of the present invention is to provide a system and method for detecting appearance defects of RFID antennas, which can detect the line width and broken wires of RFID antennas in a compact and easy-to-operate manner. Quality inspection of key factors such as sticking/sticking and burrs/printing pollution can obtain the quality inspection results of RFID antennas with high efficiency and accuracy, and facilitate real-time feedback and quality control, so it is especially suitable for industrial RFID processing and manufacturing processes.

According to one aspect of the present invention, an appearance defect detection system of an RFID antenna is provided, wherein the appearance defect detection system includes a camera device, a three-degree-of-freedom mobile module, a strip light source, a backlight light source and a visual detection device, in:

The camera device is installed on a three-degree-of-freedom mobile module that can move back and forth along the three directions of the X-axis, Y-axis and Z-axis, and is located above the RFID antenna to be detected for photographing the antenna within its field of view Image;

The strip-shaped light source is symmetrically arranged on both sides of the camera device, and the backlight light source is arranged below the RFID antenna to be detected, so that it can be selectively turned on and matched with the camera device according to the difference of the RFID antenna substrate, so as to carry out the detection of the antenna. shooting operation;

The visual detection device includes a calibration unit, an image acquisition unit, an edge detection unit, a line width measurement unit, and a connected domain analysis unit, wherein the calibration unit is used to establish pixel coordinates for the image to be captured by the camera device, and convert the pixel coordinates into the same Coordinate values under the world coordinate system; the image acquisition unit is used to collect the antenna image taken by the camera; the edge detection unit is used to perform edge detection on the collected overall image, and obtain the number of RFID antennas in the image and their corresponding length, width, and center point coordinate value information; the line width measurement unit is used to set the interest area for each antenna in the collected image, and measure the average line width and standard antenna in the area respectively The line widths of the antennas are compared; the connected domain analysis unit is used to obtain separate images of each antenna from the collected images, and perform binarization processing and blob analysis to obtain detection results related to pattern disconnection and/or adhesion, or The image of each antenna is matched and subtracted with the standard antenna image, and then binarization processing and blob analysis are performed to obtain detection results related to burrs and/or printing contamination.

As a further preference, the visual inspection device further includes a display unit for displaying intermediate and final inspection results during the inspection of appearance defects.

As a further preference, the camera device is an industrial CCD camera.

According to another aspect of the present invention, a corresponding detection method is also provided, characterized in that the detection method comprises the following steps:

(a) Before using the camera device to capture the image of the RFID antenna to be detected, perform a pre-calibration step of the camera device;

(b) Make the camera device reach the required position by operating the three-degree-of-freedom mobile module, and take images of one or more RFID antennas to be detected through the cooperation between it and the bar light source or backlight light source;

(c) Collect the captured image and perform edge detection on it, then set the edge center point that meets the detection conditions, that is, the edge detection length is consistent with the standard antenna, as the center point of the antenna to be detected, and count the edge centers that meet the detection conditions Point the number to obtain the number of RFID antennas in the detection image, and record information about the length, width, and center point coordinates of each antenna;

(d) Selectively perform at least one of the following detection steps according to the requirements of the working conditions, and when any one of the detection steps does not meet the requirements, it is determined that the quality of the corresponding RFID antenna is unqualified:

(d1) Line width detection: set the area of interest for each antenna in the detection image, and measure the average line width in this area to compare with the line width of the standard antenna;

(d2) Disconnection/sticking detection: According to the length, width and center point coordinate value information of each RFID antenna obtained after edge detection, separate images of each antenna are obtained and binary processing and blob analysis are performed on them, Then compare the number of black and white connected domains with the corresponding number of black and white connected domains of the standard antenna, so as to determine whether the pattern disconnection and/or adhesion phenomenon occurs;

(d3) Burr/print contamination detection: According to the length, width and center point coordinate value information of each RFID antenna obtained after edge detection, separate images of each antenna are obtained respectively, and are matched with standard antenna images and subtracted Perform binarization and blob analysis, and then determine whether burrs and/or printing contamination have occurred based on the existence of black areas and their area information.

As a further preference, in step (b), an industrial CCD camera is used to capture images of one or more RFID antennas to be detected.

As a further preference, in step (d2), if the number of black and white connected domains of the antenna image to be detected is equal to the number of black and white connected domains corresponding to the standard antenna, it is considered that there is no pattern disconnection and adhesion; otherwise , it is determined that the RFID antenna corresponding to the single image does not meet the quality requirements.

As a further preference, in step (d2), if the number of white connected domains in the image of the antenna to be detected is more than the number of white connected domains corresponding to the standard antenna, or the number of black connected domains of the former is less than that of the latter If the number of corresponding black connected domains corresponds to the antenna, it is determined that the antenna has pattern adhesion; if the number of white connected domains in the image of the antenna to be detected is less than the number of white connected domains corresponding to the standard antenna, or the number of black connected domains of the former If there are more than the number of black connected domains corresponding to the latter, it is determined that the antenna has a pattern disconnection phenomenon.

As a further preference, in step (d3), if there is no black area, or if there is a black area but its actual area is within the tolerance range, it is determined that there is no burr or printing contamination; if there is a black area and the black area If the actual area of the area exceeds the set tolerance threshold, it is determined that there is a burr or printing contamination.

As a further preference, in step (d3), it can be further judged that the black area belongs to burr or printing contamination according to the actual area size of the black area.

Generally speaking, compared with the prior art, the appearance defect detection system and method for RFID antennas according to the present invention mainly have the following technical advantages:

1. By adopting the quality inspection method of first edge detection and positioning and then local defect analysis, it can adapt to the detection of various complex antenna patterns, and helps to improve the detection efficiency while ensuring the accuracy and consistency of the detection results;

2. By choosing to detect factors such as antenna line width, pattern disconnection/sticking, burr/printing pollution, etc., the order and focus of quality inspection can be adjusted according to different needs, which is convenient for process management, so it is especially suitable for industrialization. actual control;

3. The overall appearance defect detection system has a compact structure and is easy to operate. It can be flexibly used in the detection of RFID antennas of different substrate types to obtain high-resolution detection images and achieve high visual detection accuracy.

Description of drawings

Fig. 1 is the general structural representation of the appearance defect detection system for RFID antenna according to the present invention;

Fig. 2 is a schematic diagram of the composition of the appearance defect detection device shown in Fig. 1;

Fig. 3 is the process flow chart of the appearance defect detection method for RFID antenna according to the present invention;

Throughout the drawings, the same reference numerals are used to designate the same elements or structures, wherein:

1-camera device 2-three degrees of freedom mobile module 3-strip light source 4-backlight light source 5-visual inspection device 6-rack 7-antenna to be tested 501-calibration unit 502-image acquisition unit 503-edge detection unit 504 -Line width measurement unit 505-connected domain analysis unit 506-display unit

Detailed ways

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

FIG. 1 is a schematic diagram of the overall structure of an appearance defect detection system for RFID antennas according to the present invention. As shown in the figure, the appearance defect detection system for RFID antenna according to the present invention mainly includes a camera device 1 , a three-degree-of-freedom mobile module 2 , a strip light source 3 , a backlight light source 4 and a visual inspection device 5 . The imaging device 1 is, for example, in the form of an industrial camera, which is installed on the three-degree-of-freedom mobile module 2 and above the manufactured RFID antenna as the object to be detected, so that it can move freely in the directions of the X, Y and Z axes, And take multiple images of the antenna substrate within its field of view. In order to provide the imaging device 1 with the necessary light for capturing images, the front and rear sides of the imaging device 1 (or the left and right sides, depending on the installation form of the imaging device itself on the three-degree-of-freedom mobile module) are symmetrical A strip light source 3 is installed and can move with the camera device. In addition, a backlight light source 4 is also provided below the RFID antenna, which is mainly because the antenna substrate may be made of different materials: when the substrate is a transparent substrate, the backlight light source 4 can be used for illumination; and when the substrate is opaque or poorly transparent , then the strip light source 3 can be used instead for illumination.

When the camera device 1 cooperates with the strip light source 3 or the backlight source 4 to obtain images of one or more RFID antennas 7 within its field of view, the camera device 1 transmits the captured image to the visual inspection device through a data line or wirelessly. 5. The visual inspection device 5 is, for example, in the form of a computer, and includes a calibration unit 501 , an image acquisition unit 502 , an edge detection unit 503 , a line width measurement unit 504 and a connected domain analysis unit 505 . Wherein, the calibration unit 501 is used to establish pixel coordinates for the image to be captured by the camera 1 , and convert the pixel coordinates into coordinate values in the same world coordinate system. The image acquisition unit 502 is used to acquire images captured by the camera device 1 . The edge detection unit 503 performs edge detection on the image collected by the image acquisition unit 502 and sets the edge center point that meets the detection condition, that is, the edge detection length matches the standard antenna (that is, the antenna design template used as a comparison reference), as the edge center point to be detected. Detect the center point of the antenna, and count the number of edge center points that also meet the conditions, thereby obtaining the number of complete RFID antennas in the detection image; in addition, according to the coordinate values obtained by the conversion of the calibration unit 501, record the relevant length of each antenna respectively. , width and center point coordinates and other information.

Considering that the line width of the antenna is the main factor affecting the conductive characteristics of the antenna, if the line width is too narrow, the resistance value will be too small or even cause a short circuit, and if the line width is too wide, the resistance value will be too large and lead to an open circuit. Therefore, in the process of visual appearance defect detection First of all, it is necessary to perform quality inspection on the line width of each RFID antenna. The line width measurement unit 504 is used to set a region of interest (ROI, Region Of Interest) for each antenna in the detection image, and measure the average line width in this area for comparison with the line width of the standard antenna: if it is within the tolerance range, it is considered The line width measurement is qualified; otherwise, it is determined that the RFID antenna corresponding to the average line width of the area does not meet the quality requirements. In industrial practice, it is generally possible to select the location where the key or concentrated wiring of the antenna to be detected and its surroundings are selected as an area of interest, and for example, a threshold of no more than 5% can be set as acceptable in the RFID antenna manufacturing process Of course, these setting values can also be adjusted according to specific process conditions and customer needs. As for the average line width, it is generally understood in the field that after the user defines the stripe color and stripe direction of the antenna pattern, the average value of the width of all stripes in the specified area.

If the line width detection is qualified, continue to detect whether the antenna pattern is broken and/or stuck. The connected domain analysis unit 505 can extract the individual complete images of each antenna from the overall detection image according to the information about the length, width and center point coordinates of each antenna acquired by the edge detection unit 502 . By performing binarization and blob analysis on these single images, we can know the number information of its black and white connected domains, and compare it with the number of black and white connected domains corresponding to the standard antenna: if the black and white connected domains If the number is equal to that of the standard antenna, it is considered that there is no pattern broken line adhesion phenomenon; otherwise, it is determined that the RFID antenna corresponding to the single image does not meet the quality requirements. Specifically, it may also be determined according to the respective numbers of black and white connected domains whether pattern disconnection or adhesion occurs. For example, if the number of white connected domains in the single image is more than the number of white connected domains obtained after binarization and blob analysis of the standard antenna, or the number of black connected domains in the former is less than that corresponding to the latter If the number of black connected domains is smaller than the number of black connected domains, it can be considered that the antenna has pattern adhesion; for another example, if the number of white connected domains in the single image is less than the white connected domains obtained after binarization and blob analysis of the standard antenna number, or the number of black connected domains of the former is more than the number of black connected domains corresponding to the latter, it can be considered that the antenna has pattern disconnection.

If the phenomenon of pattern disconnection and adhesion is not detected, it may continue to detect whether the antenna has burrs and/or printing contamination and the like. The so-called burrs refer to the protrusions or depressions that appear on one or both sides of the lines in the antenna plane due to manufacturing reasons such as etching and printing. This phenomenon may cause short circuits or short circuits between antenna lines, or affect the antenna resistance. The electrical performance parameters of the value, its specific area or deviation threshold need to be determined according to the type, use, size and complexity of the RFID tag. The so-called printing pollution refers to the redundant pattern presented on the substrate in addition to the normal antenna pattern. In the present invention, the connected domain analysis unit 505 is also used to perform this process step. The connected domain analysis unit 505 extracts individual complete images of each antenna from the overall detection image according to the information about the length, width and center point coordinates of each antenna obtained by the edge detection unit 502 . For these individual images, it is first necessary to perform a matching operation with the standard antenna image, and obtain information such as the matching rotation angle; then, subtract the two images according to the matching information, and perform binarization on the obtained image And blob analysis, so that the difference between the image to be detected and the standard antenna image can be clearly shown: if there is no black area, or although there is a black area but its actual area is within the tolerance range, it is considered that there is no black area Burr or print contamination; if there is a black area and the actual area of the black area exceeds the tolerance range, record the center point corresponding to the black area and determine that the RFID antenna corresponding to the single image does not meet the quality requirements. More specifically, according to the actual size of the black area, it can be further judged that the area belongs to burrs or printing contamination.

In addition, for the convenience of actual operation and control, the visual detection device 5 may also include a display unit 506, which is used to display the intermediate and final detection results of the above-mentioned various detection steps, and can inform the operator of the specific detection code and content.

Next, the process of detecting appearance defects according to the present invention will be described in detail. As shown in Figure 2, according to the RFID antenna appearance defect detection method of the present invention comprises the following steps:

Firstly, before the RFID antenna within the field of view is photographed by the camera 1, a pre-calibration step of the camera is performed. The pre-calibration step specifically includes establishing pixel coordinates for the image to be captured by the camera device, and converting the pixel coordinates into coordinate values in the same world coordinate system. In addition, if the working parameters have changed compared with the parameters of the last working time, the pre-calibration step should also be performed again to correct the distortion and ensure the correct detection conclusion.

Then, by operating the three-degree-of-freedom moving module 2, the camera device 1 reaches the desired position, and cooperates with the strip light source 3 or the backlight light source 4 to capture images of one or more manufactured RFID antennas within its field of view. The camera device 1 transmits the captured overall image to the visual inspection device 5 and is collected by the image acquisition unit 502 .

Next, for the overall image collected by the image acquisition unit 502, the edge detection unit 503 performs edge detection on it through the edge detection operator, and sets the edge center point that meets the detection condition, that is, the edge detection length is consistent with the standard antenna, as the edge center point to be detected The center point of the antenna, and count the number of edge center points that also meet the conditions, thereby obtaining the complete RFID antenna number in the detection image; in addition, the coordinate values obtained by converting the calibration unit 501 can also be recorded respectively. , width and center point coordinates and other information.

After completing the statistics of the number of antennas to be detected and obtaining their corresponding position information, the line width measurement unit 504 sets a region of interest for each antenna in the detection image, and measures the average line width in the region of interest and the line width of the standard antenna For comparison: if it is within the tolerance range, the line width measurement is considered qualified; otherwise, it is determined that the RFID antenna corresponding to the average line width of the area does not meet the quality requirements.

In addition, the connected domain analysis unit 505 can be used to detect whether the antenna has pattern disconnection/sticking, and whether there is burr/printing pollution and other phenomena: first, according to the length, width and center point of each antenna acquired by the edge detection unit 502 Coordinate values and other information, and separate complete images of each antenna are extracted from the overall detection image. For these individual images, when it is necessary to detect pattern disconnection/sticking, perform binarization processing and blob analysis on each image, thereby obtaining the number information of its black and white connected domains, and corresponding to the standard antenna Compare the number of black and white connected domains of . When it is necessary to detect burrs/printing contamination, it is necessary to perform a matching operation on each image with the standard antenna image, and obtain matching rotation angle and other information; then, subtract the two images according to the matching information, and compare the obtained Binarization and blob analysis are performed on the image to reveal whether there are glitches or print contamination.

Finally, the display unit 506 can display the detection results of the above-mentioned various detection steps, and inform the operator of the specific detection code and content. In addition, it should be noted that the above-mentioned quality inspection steps related to line width, broken wires/sticking, and burrs/printing contamination can be adjusted in order, or only some of them can be selected for execution according to needs. If the above defects are detected in any link, the antenna is considered unqualified, and the inspection of the antenna is ended.

It is easy for those skilled in the art to understand that the above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present invention, All should be included within the protection scope of the present invention.

Claims (1)

1. A method for performing appearance defect detection to an RFID antenna, characterized in that the method comprises the following steps: (a) Before the image of the RFID antenna to be detected is taken by the camera device, the pre-calibration step of the camera device is performed. In this operation, the pixel coordinates of the image to be captured by the camera device are first established, and then the pixel coordinates are transformed into the same world Coordinate values in the coordinate system; (b) Make the camera reach the required position above the RFID antenna to be detected by operating the three-degree-of-freedom mobile module, and use the camera device and the strip light sources symmetrically arranged on both sides of it or arranged on the RFID antenna to be detected The cooperation between the backlight light sources below takes the overall image of multiple RFID antennas to be detected; (c) Collect the overall image taken, and perform edge detection on it, then set the center point of the edge that meets the detection conditions, that is, the edge detection length is consistent with the standard antenna, as the center point of the antenna to be detected, and count the edges that meet the detection conditions The number of center points obtains the RFID antenna quantity in the detection image, and simultaneously records information about length, width and center point coordinate values of each antenna in the overall image according to the coordinate values obtained by step (a) conversion; (d) Perform appearance defect detection of key indicators such as line width, pattern disconnection and adhesion, burrs and printing contamination in sequence. During this appearance defect detection process: (d1) First, the line width of each RFID antenna is inspected for quality. In this operation, each antenna in the image to be detected is set to a region of interest, and the region of interest is set as the location where the RFID antennas are concentrated, and then Measure the average line width in each region of interest and compare it with the line width of the standard antenna: when the comparison value does not exceed the threshold of 5% of the line width of the standard antenna, the line width measurement is considered qualified, and proceed to step (d2); Otherwise, it is determined that the corresponding RFID antenna does not meet the line width quality requirements, and the operation is terminated; (d2) According to the information about the length, width and center point coordinate value of each antenna recorded in step (c), extract the complete image of each RFID antenna from the overall image; perform binarization processing on these individual images respectively and blob analysis to obtain the number information of its black and white connected domains, and then compare it with the number of black and white connected domains corresponding to the standard antenna: when the number of black and white connected domains is equal to that of the standard antenna, it is determined that the There are image disconnection and adhesion defects, and continue to perform step (d3); and when the number of white connected domains is more than the standard antenna or the number of black connected domains is less than the standard antenna, it is determined that there is pattern adhesion in the corresponding RFID antenna, End the operation; in addition, when the number of white connected domains is less than the standard antenna or the number of black connected domains is more than the standard antenna, it is determined that the corresponding RFID antenna has a pattern disconnection phenomenon, and the operation is also terminated; (d3) According to the information about the length, width and center point coordinate value of each antenna recorded in step (c), the complete image of each RFID antenna is extracted from the overall image; these individual images are respectively executed with the standard antenna image The matching operation, and obtain the matching rotation angle information; subtract the two images according to the matching information, and perform binarization and blob analysis on the image obtained after the subtraction, and then according to the existence of the black area and its area information , so as to determine whether there is burr and/or printing contamination: when there is no black area or there is a black area but its actual area is within the tolerance range, it is determined that there is no burr and printing contamination; and when there is a black area and its actual area When it exceeds the tolerance range, record the center point corresponding to the black area and determine that the RFID antenna corresponding to the individual image is defective, and then continue to further determine whether it is a burr defect or a printing contamination defect based on the actual size of the black area.