TWI728708B - Automated optical inspection system and metod for inspecting defect in contact lens edge the same - Google Patents

Abstract

An automated optical inspection (AOI) system and method for inspecting defect in contact lens edge the same are provided. The AOI system can divide the contact lens edge into a plurality of segments, and then perform similarity analysis s on these segments, and detect whether there are defect in the contact lens edge according to the results of the similarity analysis. Therefore, the AOI system does not need to use a qualified reference image to compare with the contact lens, the AOI system uses each part of the contact lens edge itself for self-inspection to improve the accuracy of detection.

Description

Automatic optical detection system and method for detecting edge defect of contact lens

The invention relates to a detection system, and more particularly to an automatic optical inspection (Automated Optical Inspection, AOI) system and a method for detecting edge defects of contact lenses.

Defects on the edge of contact lenses include deformation, nicks, cracks, and burrs. In order to improve the efficiency of manual visual inspection, the industry currently uses AOI systems for automated inspection. The AOI system can be divided into two parts: hardware and software. The hardware part uses a camera device with an appropriate light source to capture contact lens images, while the software part focuses on the development of algorithms, such as edge detection algorithms developed. The AOI system can detect whether there are defects on the edge of the contact lens.

However, the known edge detection algorithm usually compares the captured contact lens image with a qualified reference image to detect the flaws on the edge of the contact lens, but its disadvantage is that it is still difficult to target some subtle flaws. Was detected. In addition, uneven illumination will also affect the accuracy of detection. Therefore, how to design an AOI system and a method for detecting edge defects of contact lenses has become an important topic in this field.

In view of this, an embodiment of the present invention provides an AOI system, including a tray, a light source Module, image capture module and system host. The carrier plate is used to carry the contact lens, and the light source module is placed under the carrier plate and used to emit parallel light onto the contact lens. The image capturing module is placed on the top of the tray and corresponds to the position of the contact lens, and is used to obtain the contour image of the contact lens. The system host is coupled to the image capture module to divide the edge of the contact lens on the contour image into multiple segments, and then perform similarity analysis on the interaction of these segments, and detect the edge of the contact lens based on the results of the similarity analysis Whether there are blemishes.

In addition, embodiments of the present invention provide a method for detecting edge defects of contact lenses, which is implemented in an AOI system. The AOI system includes a tray, a light source module, an image capture module, and a system host. The method includes the following step. First, the contact lens is carried by the carrier plate, and the parallel light is emitted to the contact lens by the light source module. Secondly, the image capturing module is used to obtain the contour image of the contact lens. Then, the system host is used to divide the edge of the contact lens on the contour image into multiple segments, and then the similarity analysis is performed on the interaction of these segments, and according to the results of the similarity analysis, it is detected whether the edge of the contact lens is flawed.

In order to further understand the features and technical content of the present invention, please refer to the following detailed description and drawings about the present invention. However, the drawings provided are only for reference and description, and are not used to limit the present invention.

1: AOI system

10: loading plate

12: Light source module

14: Image capture module

16: System host

20: Contact lenses

30: buffer

S310~S350: Process steps

P1~P16: section

Fig. 1 is a schematic diagram of an AOI system provided by an embodiment of the present invention.

Fig. 2 is a schematic diagram of a contour image of a contact lens provided by an embodiment of the present invention.

FIG. 3 is a flowchart of the steps of a method for detecting edge defects of a contact lens provided by an embodiment of the present invention.

The following are specific specific examples to illustrate the implementation of the present invention, and those skilled in the art can understand the advantages and effects of the present invention from the content provided in this specification. The present invention can be implemented or applied through other different specific embodiments, and various details in this specification can also be based on different viewpoints and applications, and various modifications and changes can be made without departing from the concept of the present invention. In addition, the drawings of the present invention are merely schematic illustrations, and are not drawn according to actual dimensions, and are stated in advance. The following embodiments will further describe the related technical content of the present invention in detail, but the provided content is not intended to limit the protection scope of the present invention.

It should be understood that although terms such as "first", "second", and "third" may be used herein to describe various elements or signals, these elements or signals should not be limited by these terms. These terms are mainly used to distinguish one element from another, or one signal from another signal. In addition, the term "or" used in this article may include any one or a combination of more of the associated listed items depending on the actual situation.

First, please refer to FIG. 1, which is a schematic diagram of an AOI system provided by an embodiment of the present invention. As shown in Figure 1, the AOI system 1 includes a tray 10, a light source module 12, an image capture module 14 and a system host 16. The light source module 12 and the image capture module 14 can be implemented by pure hardware. , Or it can be realized through hardware and firmware or software, but the present invention is not limited by this. In this embodiment, the carrier plate 10 is used to carry the contact lens 20, and in practice, the carrier plate 10 can contain a buffer 30 so that the contact lens 20 is immersed in the buffer 30 for protection, but the present invention also Do not use this as a restriction. In addition, the light source module 12 is placed under the carrier plate 10 and used to emit parallel light to the contact lens 20. It should be noted that parallel light is also called directional light, which is a group of parallel light without attenuation. Therefore, the AOI system 1 will be able to avoid the effects of uneven illumination.

The image capturing module 14 is placed above the tray 10 and corresponds to the position of the contact lens 20 to obtain the contour image of the contact lens 20, as shown in FIG. 2. In practice, the image capturing module 14 may be composed of, for example, a charge coupled device (CCD) and a lens, but the present invention is not limited thereto. In addition, the image capturing module 14 is coupled to the system host 16 and transmits the obtained contour image to the system host 16. In this embodiment, the system host 16 may be composed of, for example, a personal computer and peripheral devices, but the present invention is not limited thereto. All in all, those skilled in the art should understand that the system host 16 includes an operating system (not shown in FIG. 1), and the operating system will carry an edge detection algorithm to instruct the system host 16 to use the contour image The edge of the contact lens is divided into multiple segments, and then the similarity analysis is performed on the interaction of these segments, and according to the result of the similarity analysis, whether the edge of the contact lens is flawed is detected.

For example, for the convenience of the following description, this embodiment only uses the system host 16 to divide the edge of the contact lens on the contour image of FIG. 2 into 16 sections, namely section P1 to section P16, but it is not used To limit the present invention. That is to say, each segment represents a small part of the edge of the contact lens, and in practice, the similarity analysis of the edge (segment) without a defect and the edge (segment) without a defect should give a high degree of similarity ( Similarity Measure), but the similarity analysis of the edges (sections) with flaws and the edges (sections) without flaws results in low similarity. In addition, because each defect has its own uniqueness, even if the edge (section) with a certain defect and the edge (section) with another defect are analyzed for similarity, a high degree of similarity will not be obtained. Therefore, in this embodiment, the result of the similarity analysis will be expressed by the similarity, and the system host 16 can compare the similarity of any two edges (sections) with a threshold to detect the edge of the contact lens. Whether there are blemishes. Please note that the present invention does not limit the specific value of the threshold value, and those skilled in the art should be able to design according to actual needs or applications.

As shown in Figure 2, because the section P1 has notch defects but the section P2 has no defects, the similarity between the section P1 and the section P2 should be lower than the threshold, and the system host 16 at this time has not yet been able to know that it is a section. Segment P1 has defects and segment P2 has no defects. Therefore, when the similarity between the section P1 and the section P2 is lower than the threshold value, the system host 16 detects that the edge of the contact lens is flawed, and can determine that at least one of the section P1 and the section P2 is a flawed invisible The edge of the glasses. Next, in order to further confirm the flawed It is the section P1 and/or the section P2, so the system host 16 can analyze the similarity between the section P1 and the section P3, and perform the similarity analysis on the section P2 and the section P3. However, since the section P3 is not flawed, when the similarity between the section P1 and the section P3 is also lower than the threshold, but the similarity between the section P2 and the section P3 is higher than the threshold, the system host 16 can The segment P1 is judged to be a defective contact lens edge.

Similarly, the system host 16 can perform similarity analysis on the section P1 and the section P4, perform similarity analysis on the section P2 and the section P4, and perform similarity analysis on the section P3 and the section P4. Therefore, in other embodiments, even if the sections P1 to P4 are defective in two sections, the system host 16 can also use these six combinations (that is, section P1 and section P2, section P1 and section P1). Section P3, section P1 and section P4, section P2 and section P3, section P2 and section P4, and section P3 and section P4) similarity, it is judged that section P1~section P4 are Which two sections are flawed. In a word, the present invention does not limit the sequence of taking out two sections from these sections for similarity analysis. In addition, if the section P1 to section P4 are defective in more than three sections, the system host 16 has to perform similarity analysis by interacting with other sections and section P1 to section P4. In other words, the more sections are divided, the more combinations of similarity analysis the sections can interact with, and based on the similarity of the more combinations, the system host 16 can effectively determine the more defective Section, which improves the accuracy of detection. Therefore, those skilled in the art should be able to determine the number of segments to be divided according to actual needs or applications.

In addition, as mentioned above, the AOI system 1 uses parallel light to avoid the impact of uneven illumination. However, in order to achieve the accuracy of overcoming the impact of illumination, the system host 16 uses multiple sections with similar illumination levels. Perform similarity analysis interactively. For example, section P1 to section P4 are four sections with similar light levels. Therefore, the system host 16 uses sections P1 to P4 with similar light levels to perform similarity analysis interactively, and according to the preceding six sections. Based on the similarity of each combination, the system host 16 can detect whether there is a defect in the section P1 to the section P4. Similarly, the sections P5~P8 are the other four sections with similar light levels. Therefore, the system host 16 is the section P5~ with similar light levels. Section P8 to interactively perform similarity analysis. In other words, the system host 16 can alternately take out two sections from section P5~section P8 for similarity analysis, that is, a total of six new combinations (that is, section P5 and section P6, section The similarity between section P5 and section P7, section P5 and section P8, section P6 and section P7, section P6 and section P8, and section P7 and section P8), and according to this new six Based on the similarity of each combination, the system host 16 can detect whether the section P5~section P8 are defective. Since the details of the judgment are as described in the foregoing content, it will not be repeated here.

Similarly, section P9~section P12 are the other four sections with similar illumination levels, and section P13~section P16 are the last four sections with similar illumination levels. Therefore, the system host 16 is based on The section P9~section P12 with similar light intensity is used for interactive similarity analysis, and the section P13~section P16 with similar light intensity is used for interactive similarity analysis. As shown in Figure 2, because the section P10 has deformation defects but the section P9 has no defects, the similarity between the section P9 and the section P10 should be lower than the threshold, and the system host 16 at this time has not yet been able to know that it is a section. Segment P10 has a defect and segment P9 has no defect. Therefore, when the similarity between the section P9 and the section P10 is lower than the threshold value, the system host 16 can also determine that at least one of the section P9 and the section P10 is a defective contact lens edge. Then, since the section P11 is also free of defects, when the similarity between the section P10 and the section P11 is also lower than the threshold, but the similarity between the section P9 and the section P11 is higher than the threshold, the system host 16 can The segment P10 is judged to be a defective contact lens edge. Since the operation details are as described in the previous content, I will not repeat them here.

Please note that this embodiment only indicates that one of the first three adjacent sections with similar light levels is flawed, and the two sections are interactively extracted from these three sections for similarity analysis. Based on the similarity of the three combinations, the system host 16 can detect which of the three sections is defective, but for multiple sections with similar light levels, and there is more than one of these multiple sections. When the location section is defective, those with ordinary knowledge in the technical field should understand that the system host 16 can extract two sections from these multiple sections interactively for similarity analysis. For the similarity of all combinations, it is detected which of the multiple sections are defective, so the details of the judgment of the modification or change will not be repeated here. As mentioned above, the present invention does not limit the sequence of taking out two sections from these multiple sections for similarity analysis, and the more sections are divided, the combination of similarity analysis that the sections can interact with The more, and based on the similarity of the more combinations, the system host 16 can effectively determine the more defective sections.

Finally, in order to further explain the operation process of the AOI system 1, the present invention further provides an implementation method of its operation method. Please refer to FIG. 3, which is a flow chart of the method for detecting edge defects of contact lenses provided by an embodiment of the present invention. It should be noted that the method in FIG. 3 can be executed in the AOI system 1 in FIG. 1, so please refer to FIG. 1 together to facilitate understanding, but the present invention does not limit the method in FIG. 3 to only be executed in the AOI system in FIG. System 1.

As shown in FIG. 3, in step S310, the carrier plate 10 is used to carry the contact lens 20, and in step S320, the light source module 12 is used to emit parallel light onto the contact lens 20. Next, in step S330, the image capturing module 14 is used to obtain the contour image of the contact lens 20, and in step S340, the system host 16 is used to divide the edge of the contact lens on the contour image into multiple segments, as shown in FIG. 2 Shown. Then, in step S350, similarity analysis is performed on these section interactions, and according to the result of the similarity analysis, it is detected whether there is a defect on the edge of the contact lens. Since the details are as described in the previous content, it will not be repeated here.

In summary, the embodiments of the present invention provide an AOI system and a method for detecting edge defects of contact lenses. The edge of the contact lens can be divided into multiple segments, and then similarity analysis is performed on the interaction of these segments, and based on the similarity The result of the sexual analysis is to detect whether there are defects on the edge of the contact lens. Therefore, the present invention does not need to use a qualified reference image to compare with the contact lens, but uses each part of the edge of the contact lens to perform self-detection. Its advantages are not only simplifying the complicated image processing, but also aiming at some of the more subtle The flaws can also be detected by the similarity of each part, which improves the accuracy of detection. In addition, the present invention not only uses parallel light to illuminate the contact lens, but also Similarity analysis is performed interactively on multiple sections with similar illumination levels to overcome the problem of uneven illumination affecting detection.

The content provided above is only the preferred and feasible embodiments of the present invention, and does not limit the scope of the patent application of the present invention. Therefore, all equivalent technical changes made by using the description and schematic content of the present invention are included in the application of the present invention. Within the scope of the patent.

S310~S350: Process steps

Claims (10)

An automatic optical inspection system includes: a carrier plate for carrying a contact lens; a light source module placed under the carrier plate for emitting parallel light to the contact lens; and an image capturing module Above the carrier plate and corresponding to the position of the contact lens to obtain a contour image of the contact lens; and a system host coupled to the image capturing module for the invisible image on the contour image The edge of the lens is divided into multiple segments, and then the similarity analysis is performed on the interaction of these segments, and according to the result of the similarity analysis, it is detected whether the edge of the contact lens is flawed. The automatic optical inspection system according to claim 1, wherein the result of the similarity analysis is expressed by a similarity, and when the similarity of a first section and a second section of the sections is lower than When a threshold value is reached, the system host detects that the edge of the contact lens has the defect, and determines that at least one of the first section and the second section is the edge of the contact lens with the defect. The automatic optical inspection system according to claim 2, wherein when the similarity between the first section and the third section is also lower than the threshold value, the second section and the third section have the same degree of similarity. When the similarity is higher than the threshold, the system host determines that the first section is the edge of the contact lens with the defect. The automatic optical inspection system according to claim 2, wherein when the similarity between the second section and the third section is also lower than the threshold value, the difference between the first section and the third section When the similarity is higher than the threshold value, the system host determines that the second section is the edge of the contact lens with the defect. The automatic optical inspection system according to claim 1, wherein the system host performs the similarity analysis interactively with the sections with similar light levels. A method for detecting edge defects of contact lenses, implemented in an automatic optical inspection In the system, the automatic optical inspection system includes a carrier plate, a light source module, an image capture module, and a system host. The method includes: using the carrier plate to carry the contact lens, and using the light source module to emit parallel Light onto the contact lens; use the image capturing module to obtain a contour image of the contact lens; and use the system host to divide the edge of the contact lens on the contour image into a plurality of sections, and then the regions Segment interaction performs similarity analysis, and according to the results of the similarity analysis, it is detected whether the edge of the contact lens has the defect. The method according to claim 6, wherein the result of the similarity analysis is expressed by a similarity, and when the similarity of a first section and a second section of the sections is lower than a threshold At this time, the system host detects that the edge of the contact lens has the defect, and determines that at least one of the first section and the second section is the edge of the contact lens with the defect. The method according to claim 7, wherein when the similarity between the first section and a third section is also lower than the threshold value, but the similarity between the second section and the third section is high When the threshold is reached, the system host determines that the first section is the edge of the contact lens with the defect. The method according to claim 7, wherein when the similarity between the second section and a third section is also lower than the threshold value, but the similarity between the first section and the third section is high At the threshold, the system host determines that the second section is the edge of the contact lens with the defect. The method according to claim 6, wherein the system host performs the similarity analysis interactively with the sections with similar light levels.

Images