TWI733062B - Optical image automatic capture method for compound detection conditions - Google Patents

Abstract

The present invention aims to disclose a method for automatically capturing optical images under multiple detection conditions, which uses the mutual matching of image capturing devices and light sources to perform optical image capturing operations of the object under test, so as to effectively obtain the optical image under different wavelengths of light. The inspection results of good/bad products, thereby improving the inspection efficiency.

Description

Optical image automatic capture method for compound detection conditions

The present invention relates to a detection method, and particularly refers to a method for automatically capturing optical images through multiple detection conditions.

For products completed on the production line, quality control was carried out by manual inspection in the early stage. Due to the limitations of human visual perception, there will be subjective cognitive differences. For example, it is impossible to judge in detail the defects in fast-moving, micron to nanometer-level fine size products, or the external testing environment or the special conditions of the product itself. The limitation of the visible region of the spectrum makes it impossible to accurately display the defects in the product in the human eye. Later, it was developed to use manual equipment to perform quality control operations. However, due to negligence such as the physiological state of the inspector or improper operation of the instrument, the problem of missed inspection often occurs.

Nowadays, thanks to the advancement of science and technology, the era of machine vision inspection methods replacing human visual inspection has arrived, and industries in various fields have begun to use automatic inspection methods to replace manual inspections. On the one hand, the automatic detection method can not only reduce a lot of manpower, but also overcome most of the detection conditions, improve detection accuracy and detection speed, and improve the lack of conventional manual detection. It has a good detection effect. Quite objective and versatile.

As for the inspection of flat products, such as printed circuit board (PCB), display panel (Display Panel) or glass (Glass), there are those who use machine vision, non-contact optical technology for inspection in the industry. . Take the printed circuit board as an example. In this field, automatic optical inspection is a common representative method in the manufacturing process. The conventional method is to illuminate the printed circuit board with light sources with different directions or different light intensities to generate color gradations in the image. The difference is then captured by the image, and then analyzed and compared with the captured image of the printed circuit board through the schema and parameters of the qualified product in the data database.

In the previous stage, in the inspection process of the printed circuit board, after the light emitted from different intensities or different directions is irradiated, only the color gradation difference performance of the printed circuit board under the light can be obtained, and it cannot be obtained under other optical conditions. Defect identification information that can be displayed. For example, applying reflected light can detect the scratches of the printed circuit board, but cannot detect whether there are holes; applying penetrating light, it can detect the holes of the printed circuit board, but cannot detect whether it has scratches .

In the same way, light sources in different directions are used to illuminate the printed circuit board respectively, and then image capture is performed one after another. Although the image analysis obtained is sufficient to effectively identify the information of the defect, it takes a lot of time. For example, first apply reflected light, then scan the printed circuit board to detect scratches, then apply penetrating light, and then scan to detect the hole in the printed circuit board, which makes the time required for scanning and inspection to be at least twice Of time.

For this reason, in view of the above-derived problems, the inventor started to develop solutions based on the idea of improving the invention. After many years of thinking, the optical image automatic capture method with the compound detection conditions of the present invention was produced. Serve the public and promote the development of the industry.

An object of the present invention is to provide a method for automatically capturing optical images with multiple detection conditions, which uses image capturing device and light source adaptation to capture the image of the object under test, and the information presented by each color layer in the image data can be effectively identified The flaws of the object to be tested further improve the accuracy and efficiency of the test results.

An object of the present invention is to provide a method for automatically capturing optical images under multiple detection conditions, which uses the interaction of the function of the photosensitive element of the imaging device to obtain the various color layers of the object to be tested by irradiating light of different wavelengths. Image information can be used to clearly determine whether the object to be tested is flawed.

In order to achieve the aforementioned objectives and effects, the present invention discloses an automatic optical image capturing method with multiple detection conditions, which includes: providing a detection platform; providing an image capturing device to lock an image area relative to the detection platform; Provide at least one light source, including at least two wavebands of light irradiating at least one side of the image area; an object to be tested enters the image area and is irradiated by the light rays, and the image capturing device is used to capture the image to obtain the light including the wavebands Light and one of the image data of the object under test, and the image data and the color separation image information of the flawless part of the object under test, or the same as the color separation image information of the flawless part of the object under test Compare, and obtain a test result.

In an embodiment of the present invention, it is also disclosed that the light source and the image capturing device are arranged on the same side of the detection platform.

In an embodiment of the present invention, it also discloses that the image data includes image information of a red layer and a green layer, respectively, and that the image information of the red layer and the green layer and the flawless part of the object under test are separately The color separation image information, or the color separation image information of the flawless object that is the same as the object under test, is compared to obtain the detection result.

In an embodiment of the present invention, it also discloses that the light of these wavebands is a red waveband and a green waveband.

In an embodiment of the present invention, it is also disclosed that the image data includes image information of a red layer and a blue layer, respectively, and that the image information of the red layer and the blue layer and the object to be measured are different. The color separation image information of the defective part or the color separation image information of the non-defective object to be tested is compared to obtain the detection result.

In an embodiment of the present invention, it is also disclosed that the light of these wavebands is a red waveband and a blue waveband.

In an embodiment of the present invention, it is also disclosed that the image data includes image information of a green layer and a blue layer, respectively, and the image information of the green layer and the blue layer and the object to be measured are different. The color separation image information of the defective part or the color separation image information of the non-defective object to be tested is compared to obtain the detection result.

In an embodiment of the present invention, it is also disclosed that the light of these wavebands is a green waveband and a blue waveband.

In an embodiment of the present invention, it is also disclosed that the image data includes image information of a red layer, a green layer, and a blue layer, and the images of the red layer, the green layer, and the blue layer are separately The information is compared with the color separation image information of the flawless part of the test object, or the color separation image information of the flawless test object that is the same as the test object, to obtain the detection result.

In an embodiment of the present invention, it is also disclosed that the light of these wavebands is a red waveband, a green waveband, and a blue waveband.

In an embodiment of the present invention, it is also disclosed that the image capturing device is a photographing mechanism or a photographing mechanism.

In an embodiment of the present invention, it is also disclosed that the light source is arranged on one side of the inspection platform, and the image capturing device is arranged on the other side of the inspection platform.

In order to enable your reviewer to have a further understanding and understanding of the features of the present invention and the effects achieved, only the examples and detailed descriptions are provided as follows:

In the prior art, the conventional automatic optical inspection method for the object under test cannot simultaneously capture images under different optical conditions and compare the images. Therefore, the present invention proposes an optical image automatic capture method with multiple detection conditions to solve the traditional problems to be improved to improve the accuracy of the detection results and the detection efficiency, and is described in detail below.

Here is a description of a method for automatically capturing optical images with multiple inspection conditions of the present invention. Please refer to the first figure, which is a flowchart of the method for automatically capturing optical images with multiple inspection conditions of the present invention. As shown in the figure, the steps of the present invention include:

Step S10: Provide a detection platform;

Step S12: Provide an image capturing device to lock an image area relative to the detection platform;

Step S14: Provide at least one light source, including at least two wavelengths of light illuminating at least one side of the image area;

Step S16: an object to be measured enters the image area and is irradiated by the light rays, and the image capturing device is used to capture the image to obtain one of the light rays of the wavelength bands and the object to be measured; and

Step S18: Compare the image data with the color separation image information of the flawless part of the test object, or the color separation image information of the flawless test object that is the same as the test object, to obtain a detection result.

The following description is to achieve the component composition of the system of the method of the invention. Please refer to Figure 2A and Figure 2B, which are schematic diagrams 1 and 2 of the optical image automatic capturing system with compound detection conditions of the present invention. As shown in the figure, the optical image automatic capturing system 1 of the present invention with compound detection conditions includes: an image capturing device 10 and at least one light source 12. The image capturing device 10 is arranged on one side of a detection platform 9 and opposite to the detection platform 9 and lock an image area 100. The light source 12 and the imaging device 10 can be installed on the same side of the detection platform 9 (as shown in Figure 2A), or the light source 12 can be installed on one side of the detection platform 9 and the imaging device 10 can be installed on the other side of the detection platform 9 (As shown in the second figure B), it includes at least two wavelengths of light L1 and L2 illuminating at least one side of the image area 100. The image capturing device 10 captures images in the image area 100 to obtain image data 102.

The aforementioned image capturing device 10 is a photographing mechanism or a photographing mechanism.

The above-mentioned light source 12, as shown in Figure 2A, can be composed of lamps with multiple light wavebands, and can emit light L1 and L2 of two wavebands (colors) or more at one time according to the illumination needs of the inspector. One of the test objects 8 in the image area 100. It is also possible to arrange multiple light sources 12 on the other side of the detection platform 9 as shown in the second figure B, and each light source 12 respectively emits a wavelength of light L1, L2, and L3 to illuminate the test object 8; The light of the wave band can be the visible light wave band that the human eye can receive, or the invisible light wave band that can be read, analyzed, and compared by the detection software.

The above-mentioned detection software is not limited to any software, as long as it can analyze the various color layers of the image data 102, such as: Adobe Photoshop CC.

The above-mentioned DUT 8 can be a printed circuit board, a display panel or glass.

Referring again to the first and second A, the following describes the flow of the optical image automatic capturing method of the present invention with multiple detection conditions. When the automatic optical detection operation is performed on the object 8 to be tested, step S10 to step S18 will be executed. First, the detection platform 9, the image capturing device 10, and the light source 12 have been positioned at the working positions described in the second figure A and the text above. When the test object 8 enters the detection platform 9 via a conveyor belt (not shown) or a robotic arm (not shown), the light L1, L2 emitted by the light source 12 illuminates the test object 8 in the image area 100, and the image is taken The device 10 captures an image of the object 8 in the image area 100 whose image changes caused by the light rays L1 and L2. Then the image capturing device 10 captures the image data 102, and then the image data 102 and the color separation image information of the flawless part of the test object 8 or the color separation image information of the flawless test object that is the same as the test object 8 Compare. Afterwards, the test result is obtained, and it is known whether the test object 8 tested by the appropriate talent is a good product or a defective product.

Please refer to the third, fourth A, and fourth B diagrams, which are the image data reference diagram, the color separation image information reference diagram 1 and the color separation image of the optical image automatic capture method of the composite detection condition of the present invention For information, refer to Figure 2. Specifically, if the light source 12 emits a light L1 in the green wavelength band and a light L2 in the blue wavelength band, as shown in the third figure, it is the image change appearance of the object 8 to be irradiated by the light source 12, which is also The image data 102 obtained by the image capturing device 10 captured in the image area 100. Continuing as shown in the fourth A and the fourth B, the image data 102 is processed by color separation by the detection software to generate image information with a green layer (fourth A), and image information with a blue layer (fourth Picture B). Since the light L1 in the green wavelength band and the light L2 in the blue wavelength band are used to illuminate the object 8, the photosensitive element (not shown) of the imaging device 10 has a sensitivity difference for different lights L1 and L2. That is, when the light L1 of the green wavelength band irradiates the test object 8, the green part of the color of the image data 102 has a richer performance; similarly, when the light L2 of the blue wavelength band irradiates the test object 8, the color of the image data 102 The blue part has a richer performance. Therefore, after the image data 102 is processed by color separation, the image information of each color layer has obvious differences, which can be the same as the color separation image information of the flawless part of the test object 8 or the same as the flawless image information of the test object 8. Identify and compare the color separation image information of the measured object.

Continuing the previous paragraph, as shown in Figure 4A, the image information displays the green band layer of the image data 102 under low-brightness gray-scale optical conditions, so that it is possible to clearly determine whether there are defects in the brighter areas. For example, the defect marked by the red arrow in the bold circle in the figure 4A can be used to clearly distinguish the defect from the other parts of the DUT 8. However, the figure 4A is thick The flaws marked by the blue arrow in the circle frame are not easy to identify because they are located in the dark area. Similarly, as shown in Fig. 4B, the image information displays the blue band layer of the image data 102 with high-brightness gray-scale optical conditions, so that it can be clearly judged whether there is a darker area in the green band layer. defective. For example, the defect marked by the blue arrow in the bold circle in Figure 4B can clearly distinguish that the defect is different from the other parts of the test object 8. However, in Figure 4B The defect marked by the red arrow in the thick circle frame is because it is located in the bright area. When the brightness is increased, it is contaminated by light and it is difficult to distinguish clearly. Therefore, the optical image automatic capture method of the present invention can simultaneously visualize different types of defects/defects of the object to be tested, which facilitates the inspection personnel and the inspection software to identify the pros and cons of the object to be tested.

Among them, in order to respond to the color appearance of the object 8 itself, the light waveband of the light source 12 can be changed accordingly. For example, the light source 12 can emit light L1 in the green band and light L2 in the red band, L2 in the red band, and light L3 in the blue band, or light L1 in the green band, L2 in the red band, and light in the blue band. L3. In addition, the intensities of the three light rays of the green wave band L1, the red wave band light L2, and the blue wave band light L3 can be arranged and combined in different, partially identical or identical ways. According to the wavelength and intensity of the light emitted by the light source 12, the corresponding color layer can be separated, so that the inspection personnel and the inspection software can analyze and compare the defects.

In summary, the optical image automatic capture method of the present invention under compound detection conditions obtains the image information of each color layer through the interaction of the photosensitive element of the image capturing device and the light source emitting light of different wavelengths to illuminate the object under test. , Can clearly and efficiently judge whether there are defects in the test object. Herein, the automatic optical image capturing method of the present invention has effectively improved the defect problem caused by the traditional automatic optical inspection of the object under test.

1: Optical image automatic capture system 8: Object to be tested 9: Detection platform 10: Image capture device 100: Image area 102: Image data 12: Light source L1: Light L2: Light L3: Light S10: Step S12: Step S14: Step S16: Step S18: Step

The first figure: it is the flow chart of the optical image automatic capturing method under the compound detection conditions of the present invention; the second figure A: it is the schematic diagram 1 of the optical image automatic capturing system under the compound detection conditions of the present invention; the second B Figure: It is the second schematic diagram of the optical image automatic capturing system of the compound detection condition of the present invention; Figure 3: It is the image data reference diagram of the optical image automatic capturing method of the compound detection condition of the present invention; Figure 4A : It is the color separation image information of the optical image automatic capture method of the compound detection condition of the present invention, refer to Figure 1; and Figure B: It is the color separation image of the optical image automatic capture method of the compound detection condition of the present invention For information, refer to Figure 2.

S10: steps

S12: steps

S14: Step

S16: steps

S18: steps

Claims (12)

A method for automatically capturing optical images with multiple detection conditions, comprising: providing a detection platform; providing an image capturing device for locking an image area relative to the detection platform; providing at least one light source, including at least two of which are simultaneously irradiated Wave bands of light irradiate at least one side of the image area; an object to be measured enters the image area and is illuminated by the light at the same time, and the image capturing device is used to capture the image, and at the same time, the light containing the bands and the object to be measured are obtained One of the image data; and the color separation image information of the different color layers corresponding to the light of some bands in the image data and the flawless part of the test object, or the same as the flawless part of the test object to be tested The color separation image information of the object is compared, and a detection result is obtained. As described in the first item of the scope of patent application, the optical image automatic capturing method with compound detection conditions, wherein the light source and the image capturing device are arranged on the same side of the detection platform. For example, the optical image automatic capture method with compound detection conditions described in the first item of the scope of patent application, in which the different color layers corresponding to the light of some bands and the color separation image information of the flawless part of the object under test are combined, Or in the step of comparing with the color separation image information of the flawless test object that is the same as the test object, the image data contains the image information of a red layer and a green layer after the color separation process. The image information of the red layer and the green layer is compared with the color separation image information of the flawless part of the test object, or with the color separation image information of the flawless test object that is the same as the test object to obtain The test result. The optical image automatic capturing method with compound detection conditions as described in item 1 or 3 of the scope of patent application, wherein the light of these wavelength bands is a red wavelength band and a green wavelength band. For example, the optical image automatic capture method with compound detection conditions described in the first item of the scope of patent application, in which the different color layers corresponding to the light of some bands and the color separation image information of the flawless part of the object under test are combined, Or with the same flawless treatment as the DUT In the step of comparing the color separation image information of the measured object, the image data contains the image information of a red layer and a blue layer after the color separation process, and the image information of the red layer and the blue layer are separately processed Compare with the color separation image information of the flawless part of the test object, or with the color separation image information of the flawless test object that is the same as the test object, to obtain the detection result. In the optical image automatic capturing method with compound detection conditions as described in item 1 or 5 of the scope of patent application, the light in these bands is a red band and a blue band. For example, the optical image automatic capture method with compound detection conditions described in the first item of the scope of patent application, in which the different color layers corresponding to the light of some bands and the color separation image information of the flawless part of the object under test are combined, Or in the step of comparing the color separation image information of the non-defective test object that is the same as the test object, the image data contains the image information of a green layer and a blue layer after the color separation process. Compare the image information of the green layer and the blue layer with the color separation image information of the flawless part of the test object, or with the color separation image information of the flawless test object that is the same as the test object, And get the test result. The optical image automatic capturing method with compound detection conditions as described in item 1 or 7 of the scope of patent application, wherein the light of these wavelength bands is a green wavelength band and a blue wavelength band. For example, the optical image automatic capture method with compound detection conditions described in the first item of the scope of patent application, in which the different color layers corresponding to the light of some bands and the color separation image information of the flawless part of the object under test are combined, Or in the step of comparing with the color separation image information of the non-defective test object that is the same as the test object, the image data after color separation processing contains images of a red layer, a green layer and a blue layer Information, and separately the image information of the red layer, the green layer, and the blue layer and the color separation image information of the flawless part of the test object, or the same as the flawless test object of the test object. The color separation image information is compared to obtain the detection result. In the optical image automatic capturing method with compound detection conditions as described in item 1 or 9 of the scope of patent application, the light of these bands is a red band, a green band, and a blue band. As described in item 1 of the scope of patent application, the method for automatically capturing optical images with multiple detection conditions, wherein the image capturing device is a photographing mechanism or a photographing mechanism. As described in the first item of the scope of patent application, the method for automatically capturing optical images with multiple detection conditions, wherein the light source is disposed on one side of the detection platform, and the image capturing device is disposed on the other side of the detection platform.

Images