Automatic orthogonal method and system for optical detection of polaroid

Abstract

The invention provides an automatic orthogonal method and system for polaroid optical detection. The automatic orthogonal method comprises the following steps: determining a sampling area and a sampling point range; the rotation rate of the direct-current and alternating-current motor is set to be N degrees per time, the direct-current and alternating-current motor is started, the polariscope is driven to rotate, after each rotation, the gray-scale value Mijk sampled by the cameras is recorded until the rotation is 90 degrees, and i is the ith camera; j is the jth rotation of the kth rotation cycle; k is a rotation cycle when the direct current motor rotates 90 degrees; the gray-scale values obtained through all the rotation cycles of all the cameras are compared, the minimum value Mijmin of the gray-scale values and the angle theta imin corresponding to the Mijmin are obtained, and the angle theta imin serves as the orthogonal rotation angle; and rotating the orthogonal motor corresponding to each camera to a corresponding orthogonal rotation angle, and carrying out defect detection. By adopting the technical scheme of the invention, the effectiveness and reliability of defect detection are improved, and the production efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of polarizers, and in particular to an automatic orthogonal method and system for optical detection of polarizers. Background Art

[0002] In the manufacturing process of polarizers, there is a process of compounding pressure-sensitive adhesive and applying release film. In this process, the release film is between the polarizer and the camera. The birefringence effect of the release film produces glare, which causes the orthogonal state to change or even fail to be orthogonal. Especially when using release films with different orientation angles, the orthogonal lens angle must be recalibrated, otherwise it will result in over-inspection or no defects.

[0003] At present, some manufacturers use coating lamination, tearing off the release film, orthogonal detection, and then laminating the release film to achieve orthogonal; this method changes the lamination area from only lamination to: lamination, tearing off the film, detection, and re-lamination process. Three new processes have been added, which has brought great challenges to the stability of the equipment and the dust-free environment. It causes unstable process and reduced yield. Some manufacturers use an automatic orthogonal motor after lamination, which starts to rotate automatically after changing the roll, finds the set value, and stops when it reaches the set value. This method is a simple automatic orthogonal to set value method. This method does not take into account the importance of "orthogonal" to defect detection. The actual detection effect is very unstable and does not achieve the "orthogonal" state. Either the defect is overloaded or no or fewer defects are detected, and the defect detection effect is not good. Moreover, there are large differences between different film materials, and overload is easy to occur, requiring technicians to keep an eye on and manually correct at any time. In actual mass production, it is not practical. Summary of the invention

[0004] In view of the above technical problems, the present invention discloses an automatic orthogonal method and system for optical inspection of polarizers, which solves the orthogonal inspection problem caused by glare of release films.

[0005] To this end, the technical solution of the present invention is:

[0006] An automatic orthogonal method for optical detection of polarizers, wherein a camera located above the polarizer is provided in a detection area, a light source is provided on one side of the camera, a polarizer is provided at the front end of the camera, the polarizer is connected to an orthogonal motor, and the polarizer is driven to rotate by the orthogonal motor. The automatic orthogonal method for optical detection of polarizers comprises the following steps:

[0007] Step S1, determine the sampling area and sampling point range;

[0008] Step S2, set the rotation rate of the orthogonal motor to N° / time, start the orthogonal motor, drive the polarizer to rotate, and record the grayscale value M sampled by the camera after each rotation. ijk, until it rotates 90°, where i is the i-th camera; j is the j-th rotation of the k-th rotation cycle; k is the number of rotations of the orthogonal motor of 90°, which is called a rotation cycle; compare the grayscale values ​​obtained in each rotation cycle of each camera to obtain the minimum grayscale value M ijmin , and M ijmin The corresponding angle θ imin , the angle θ imin as the orthogonal rotation angle;

[0009] Step S3, rotating the orthogonal motors corresponding to the cameras to corresponding orthogonal rotation angles to perform defect detection.

[0010] Polarizer film materials of different batches and models have different orthogonal parameters. This technical solution avoids the error caused by directly setting the position of the polarizer according to the set angle for orthogonality, making subsequent defect detection more reliable and preventing defective products from flowing into subsequent processes.

[0011] As a further improvement of the present invention, step S2 further includes rotating the motor to M ijmin The corresponding angle θ imin , and then perform local iteration to obtain the angle corresponding to the lowest grayscale value as the orthogonal rotation angle.

[0012] As a further improvement of the present invention, the local iteration includes: rotating the orthogonal motor to θ imin , and iterate within the range of N / 2 at a rotation speed of 0.2° / time, where N is a natural number not less than 2; and record the lowest grayscale value sampled by the camera, and obtain the angle corresponding to the lowest grayscale value as the orthogonal rotation angle.

[0013] As a further improvement of the present invention, step S3 also includes adjusting the parameters of the camera: if the grayscale value meets the set requirements, the camera exposure is adjusted preferentially within the exposure allowable range; if the grayscale value is higher than the set value, then based on the initial setting value of the exposure time, the exposure is gradually reduced at a frequency of Li / times, so that the grayscale value (the average grayscale mean of the sampled original video signal) is adjusted to the set requirement range; if the grayscale value is lower than the set value, it is iterated in the opposite direction. Among them, Li refers to the fixed scale value of the exposure time, that is, Li / times is the step size of each adjustment of the exposure time. That is, in order to make the average grayscale mean of the camera's original video signal reach the set target, the set automatic increment step size is used for adjustment, and the grayscale value after each exposure time step adjustment is compared with the average sampled grayscale value of the servo rotation. If the sampled average grayscale value is reached, the adjustment is stopped.

[0014] As a further improvement of the present invention, if the exposure condition adjustment cannot meet the set requirements, the camera gain is adjusted, and on the basis of the initial setting value of the camera gain, the gain is gradually fine-tuned at a frequency of Mi / times, so that the grayscale value is adjusted to the set requirement range; if the grayscale value is lower than the set value, it is iterated in the opposite direction. Among them, Mi refers to the fixed scale value of the gain, that is, Mi / times is the step size of each gain adjustment. That is, in order to make the average grayscale mean of the camera's original video signal reach the set target, the set automatic increment step size is used for adjustment, and the grayscale value after each gain step adjustment is compared with the average sampled grayscale value of the servo rotation. If the sampled average grayscale value is reached, the adjustment is stopped.

[0015] As a further improvement of the present invention, if the camera gain adjustment still cannot meet the set requirements, the light source is adjusted; if the grayscale value is higher than the set value tolerance range, the light source is gradually reduced at a frequency of Pi / time based on the initial set value of the light source brightness; if the pixel mean is lower than the set value tolerance range, the light source is gradually increased at a frequency of Pi / time. Wherein, Pi refers to a fixed scale value of the light source brightness, that is, Pi / time is the step size of each adjustment of the light source brightness. In order to make the average grayscale value of the camera's original video signal reach the set target, the set automatic increment step size is used for adjustment. The grayscale value after each light source brightness step adjustment is compared with the average sampled grayscale value of the servo rotation. If the sampled average grayscale value is reached, the adjustment is stopped.

[0016] As a further improvement of the present invention, in step S2, in each 90° rotation cycle, if the angle of the last rotation at a rotation rate of N° / time exceeds 90°, it is rotated to 90° and the grayscale value sampled by the camera is recorded.

[0017] As a further improvement of the present invention, in step S1, the sampling area is a non-head area, a non-joint area and a non-edge area. Further, the edge area is an area within 20 mm from the edge, and the non-edge area is an area outside 20 mm from the edge.

[0018] As a further improvement of the present invention, in step S1, the sampling point range is based on the pixel points, 1 is sampled every 5 horizontally, and 300 pixels are left unsampled at the edge of the camera; 1 is sampled every 5 vertically, and 200 pixels are sampled vertically; the average value of each sampling point is taken as the grayscale value M sampled in the current state of each camera ijk .

[0019] The present invention also discloses an automatic orthogonal system for optical detection of polarizers. A detection area is provided with a camera located above the polarizer, a light source is provided on one side of the camera, a polarizer is provided at the front end of the camera, the polarizer is connected to an orthogonal motor, and the polarizer is driven to rotate by the orthogonal motor. The orthogonal motor, the light source, and the camera are connected to a control module, and the control module executes the automatic orthogonal method for optical detection of polarizers as described above.

[0020] Furthermore, the system includes a motor control unit and a light source control unit, wherein the motor control unit is electrically connected to the motor, the light source control unit is electrically connected to the light source, and the motor control unit, the light source control unit, and each camera are electrically connected to the control module.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] The technical solution of the present invention solves the problem that the orthogonal state of the film material often changes after the release film is applied, resulting in the inability to effectively detect. The orthogonal state of the film material changes when the material is replaced, which can also improve the effectiveness of defect detection, realize automated operation, improve reliability, and improve yield rate and production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a hardware configuration diagram of an automatic orthogonal system for optical inspection of polarizers according to an embodiment of the present invention.

[0024] Figure 2 Schematic diagram of the camera sampling point range according to an embodiment of the present invention.

[0025] Reference numerals include:

[0026] 1- camera, 2- polarizer, 3- orthogonal motor, 4- light source, 5- motor control unit, 6- light source control unit, 7- control module. DETAILED DESCRIPTION

[0027] The preferred embodiments of the present invention are described in further detail below.

[0028] An automatic orthogonal system for optical inspection of polarizers, such as Figure 1As shown, the detection area is provided with a camera 1 located above the polarizer, a light source 4 is provided on one side of the camera 1, a polarizer 2 is provided at the front end of the camera 1, the polarizer 2 is connected to the orthogonal motor 3, and the polarizer 2 is driven to rotate by the orthogonal motor 3. The orthogonal motor 3, the light source 4, and the camera 1 are connected to the control module 7, each orthogonal motor 3 is electrically connected to the motor control unit 5, each light source 4 is electrically connected to the light source control unit 6, and the motor control unit 5, the light source control unit 6, and each camera 1 are electrically connected to the control module 7 through a signal line. The control module 7 performs the automatic orthogonal method for optical detection of polarizers according to the following steps:

[0029] Step S1, determine the sampling area and sampling point range;

[0030] If there are dense defects in the sampling area, it will cause serious interference to the orthogonal determination and needs to be eliminated.

[0031] 1) Avoidance of the front matter:

[0032] The quality of the roll head is generally poor, and the orthogonal judgment cannot be performed at the roll head position. The roll head judgment must be added to the orthogonal judgment unit, and generally the section <5 meters is set not to perform orthogonal judgment.

[0033] 2) Avoiding connectors:

[0034] There is no orthogonal condition at the joint, so orthogonal judgment cannot be performed at the joint. Especially for some brightening film cut-and-joint rolls, there is a cut-and-joint joint every 1 meter. The 5cm around the joint needs to be avoided. An external sensor must be added to determine the joint, and when triggered, it is transmitted to the orthogonal judgment unit. No judgment action is performed in the joint area.

[0035] 3) Edge avoidance:

[0036] The left and right cameras of the camera group usually do not have full-width detection. If there is no polarizer at the corresponding position of the camera, orthogonal judgment cannot be performed. The edge areas of the left and right cameras (usually 20mm from the edge) must be set to not participate in the orthogonal judgment pixel calculation.

[0037] Regarding camera sampling, if the camera sampling points are too few, it will cause greater interference from defects; if the camera sampling is too much, it will cause the orthogonal determination time to be too long. The full sampling method will not only cause the algorithm time to be too long, but also the interference of defects on the value will be very obvious. In order to reduce the impact of defects on polarizers on orthogonal determination, a sampling method must be used.

[0038] The recommended sampling algorithm is: 1 pixel is sampled every 5 pixels horizontally, leaving 300 pixels at each edge of the camera; 1 pixel is sampled every 5 pixels vertically, leaving 200 pixels vertically. Figure 2 shown.

[0039] The camera samples once, and the average of each sampling point is taken as the sampling value of each camera at that time, denoted as Mijk, where i is the i-th camera; j is the j-th rotation of the k-th rotation cycle; k is that every 90° rotation of the servo motor is called a rotation cycle.

[0040] Step S2, set the rotation rate of the orthogonal motor to N° / time. If the last iteration exceeds 90 degrees, the last rotation speed is automatically adjusted to meet the 90-degree requirement of the iteration cycle. For example, the initial position angle of the servo motor is 30° and the rotation speed is 4°; the 22nd rotation is 118°; if it rotates another 4°, it exceeds this rotation cycle. The last rotation only needs to be 2° to reach 120°.

[0041] Start the orthogonal motor to drive the polarizer to rotate. After each rotation, record the grayscale value M sampled by the camera. ijk , until it rotates 90°. Generally, four rotation cycles are performed to rotate 360°; if efficiency is considered, 1 to 2 rotation cycles are sufficient.

[0042] Compare the grayscale values ​​obtained in each rotation cycle of each camera to obtain the minimum grayscale value M ijmin , and M ijmin The corresponding angle θ imin , the angle θ imin As the orthogonal rotation angle; specifically, the following steps can be followed:

[0043] The average value of the same iteration data of each cycle unit of the orthogonal motor is calculated; for example: for the first camera, the average value of four rotation cycles is calculated, such as M13 = (M130 + M131 + M132 + M133) / 4, where the 3 in M13 represents the jth rotation in the same iteration of the four cycle units; the 0, 1, 2, and 3 at the end represent four rotation cycles respectively. The angle of the same rotation of the same cycle unit is consistent, and the rotation angle is θi + j*N.

[0044] Find the minimum value of Mij and the corresponding angle. If there are two Mij, take the first one.

[0045] The above steps determine the approximate angle, and further local iterations can be performed, including:

[0046] The DC motor rotates to the minimum value M determined in the previous step. ijmin The corresponding angle θ imin Perform local iteration.

[0047] Assume that the initial angle is θi, then the minimum value M ijmin The corresponding angle θ imin=θi+j*N; and iterate within the range of N / 2 at a rotation speed of 0.2° / time. Calculate the lowest value and record the angle θi1.

[0048] The angle θi1 finally fed back by each rotary motor is the optimal orthogonal rotation angle of the corresponding camera.

[0049] Step S3, rotating the orthogonal motors corresponding to the cameras to corresponding orthogonal rotation angles to perform defect detection.

[0050] Step S4: Due to the characteristics of the film material, the camera rotates to an orthogonal position, which is not equal to the set grayscale value, so the camera needs to be adjusted.

[0051] a) Exposure adjustment:

[0052] If the grayscale value meets the set value, the camera exposure is adjusted preferentially within the exposure allowable range.

[0053] If the grayscale value is higher than the set value, then based on the initial exposure time setting value, the exposure is gradually reduced at a frequency of Li (preset) / times, and adjusted according to the pixel mean feedback (the Li value is automatically reduced when approaching the target to avoid exceeding it). If the grayscale value is lower than the set value, it iterate in the opposite direction.

[0054] b) Gain adjustment:

[0055] If the exposure conditions still do not meet the requirements, adjust the camera gain. Based on the initial setting value of the camera gain, gradually fine-tune the gain at a frequency of Mi (preset) / times, and adjust it according to the pixel mean feedback (self-reduce the Mi value when approaching the target to avoid exceeding it).

[0056] 1) Light source adjustment:

[0057] If the camera adjustment still cannot meet the set value, the light source is controlled to adjust. If the pixel mean is higher than the set value tolerance range, the light source is gradually reduced at a frequency of Pi (pre-specified) / times based on the initial set value of the light source brightness; if the pixel mean is lower than the set value tolerance range, the light source is gradually increased at a frequency of Pi (pre-set) / times; when solving the target, the Pi value is automatically reduced and the light source value is fine-tuned.

[0058] After passing through the camera and light source, the orthogonal condition is established and the detection conditions are met, and the orthogonal detection can be carried out normally.

[0059] Finally, the above steps are followed to automatically and regularly perform orthogonal determination and conduct mass production inspection of polarizer films. During the whole process, manual or automatic orthogonal determination can be selected.

[0060] The following describes it in conjunction with specific embodiments.

[0061] Taking wide-width lines as an example, in order to adapt to market demand, the production line needs to frequently switch product names for production. The optical differences between film materials of different product names are very large. According to the traditional orthogonal motor, it will automatically rotate to the empirical set value and then stop. This often leads to overload of the on-site detection AOI after switching between film materials of different product names. Manual correction is required, otherwise the AOI detection cannot be in the "orthogonal" state, resulting in poor detection of orthogonal defects. For example, the "invisible orthogonal bright spot" defect has caused many customer complaints because of the poor orthogonal detection effect when switching the first roll of the product name.

[0062] Before adjustment in Table 1, the existing technology is used, that is, the orthogonal motor is rotated according to the set orthogonal angle to perform defect detection. During the adjustment, the technical solution steps S1 to S3 of this embodiment are used, that is, before the orthogonal motor automatically rotates to the set value, the film material is first subjected to sampling, extreme value search, extreme value judgment and other algorithms to confirm the orthogonal position, and then rotated to the orthogonal position; after the adjustment, the entire technical solution of this embodiment is used, including the camera adjustment content in step S4, that is, before the orthogonal motor automatically rotates to the set value, the film material is first subjected to sampling, extreme value search, extreme value judgment and other algorithms to confirm the orthogonal position, and then rotated to the orthogonal position and adjusted to the set value through the camera or light source to avoid the influence of optical differences of different film materials on the orthogonal detection state. According to the system abnormality log data statistics, the frequency of AOI overload abnormality caused by wide-width line product name switching has been reduced from 2.31% to 0.11%, and the consolidation period has been reduced to 0%.

[0063] Table 1 Statistics of AOI overload caused by product name switching of wide-width lines

[0064]

[0065] It can be seen that after adopting the technical solution of this embodiment, the qualification rate of subsequent processes is improved and the reliability of detection is improved.

[0066] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.

Claims

1. An automatic orthogonal method for optical inspection of polarizers, wherein a camera located above the polarizer is provided in the inspection area, a light source is provided on one side of the camera, a polarizer is provided at the front end of the camera, the polarizer is connected to an orthogonal motor, and the polarizer is driven to rotate by the orthogonal motor, characterized in that: The automatic orthogonal method for optical detection of polarizers comprises the steps of: Step S1, determine the sampling area and sampling point range; Step S2, set the rotation rate of the orthogonal motor to N° / time, start the orthogonal motor, drive the polarizer to rotate, and record the grayscale value M sampled by the camera after each rotation. ijk , until it rotates 90°, where i is the i-th camera; j is the j-th rotation of the k-th rotation cycle; k is the number of rotations of the orthogonal motor of 90°, which is called a rotation cycle; compare the grayscale values ​​obtained in each rotation cycle of each camera to obtain the minimum grayscale value M ijmin , and M ijmin The corresponding angle θ imin , the angle θ imin as the orthogonal rotation angle; Step S3, rotating the orthogonal motors corresponding to the cameras to corresponding orthogonal rotation angles to perform defect detection.

2. The automatic orthogonal method for optical inspection of polarizers according to claim 1, characterized in that: Step S2 also includes rotating the motor to M ijmin The corresponding angle θ imin , and then perform local iteration to obtain the angle corresponding to the lowest grayscale value as the orthogonal rotation angle.

3. The automatic orthogonal method for optical inspection of polarizers according to claim 2, characterized in that: The local iteration includes: rotating the orthogonal motor to θ imin , and iterate within the range of N / 2 at a rotation speed of 0.2° / time, where N is a natural number not less than 2; and record the lowest grayscale value sampled by the camera, and obtain the angle corresponding to the lowest grayscale value as the orthogonal rotation angle.

4. The automatic orthogonal method for optical inspection of polarizers according to claim 1, characterized in that: Step S3 also includes adjusting the parameters of the camera: if the grayscale value meets the set requirements, the camera exposure is adjusted preferentially within the exposure allowable range; if the grayscale value is higher than the set value, the exposure is gradually reduced at a frequency of Li / times based on the initial set value of the exposure time, so that the grayscale value is adjusted within the set requirement range; If the grayscale value is lower than the set value, iterate in the opposite direction.

5. The automatic orthogonal method for optical inspection of polarizers according to claim 4, characterized in that: If the exposure condition adjustment cannot meet the set requirements, adjust the camera gain. Based on the initial setting value of the camera gain, gradually fine-tune the gain at a frequency of Mi / times to adjust the grayscale value to within the set requirement range; If the grayscale value is lower than the set value, iterate in the opposite direction.

6. The automatic orthogonal method for optical inspection of polarizers according to claim 5, characterized in that: If the camera gain adjustment still cannot meet the set requirements, adjust the light source; if the grayscale value is higher than the set value tolerance range, then based on the initial setting value of the light source brightness, gradually reduce the light source at a frequency of Pi / times; If the pixel mean is lower than the set value tolerance range, the light source is gradually increased at a frequency of Pi / times.

7. The automatic orthogonal method for optical inspection of polarizers according to claim 1, characterized in that: In step S2, in each 90° rotation cycle, if the angle of the last rotation exceeds 90° at a rotation rate of N° / time, the camera is rotated to 90° and the grayscale value sampled by the camera is recorded.

8. The automatic orthogonal method for optical inspection of polarizers according to claim 1, characterized in that: In step S1, the sampling area is a non-head area, a non-joint area and a non-edge area.

9. The automatic orthogonal method for optical inspection of polarizers according to claim 1, characterized in that: In step S1, the sampling point range is based on the pixel points, and 1 is taken every 5 horizontal points, and 300 pixels are left at the edge of the camera without sampling; The vertical sampling is done every 5 pixels, and 200 pixels are sampled vertically; the average value of each sampling point is taken as the grayscale value M sampled at the current state of each camera. ijk .

10. An automatic orthogonal system for optical inspection of polarizers, characterized in that: The detection area is provided with a camera located above the polarizer, a light source is provided on one side of the camera, and a polarizer is provided at the front end of the camera. The orthogonal motor drives the polarizer to rotate, and the orthogonal motor, light source, camera and control module are connected. The control module executes the automatic orthogonal method for optical inspection of polarizers as described in any one of claims 1 to 9.

Images