Defect detection method of near-eye display
Patent Information
- Application Number
- CN202380024509.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-01-31
- Publication Date
- 2024-11-22
AI Technical Summary
Existing imaging systems are difficult to effectively detect defects in virtual images in near-eye displays, resulting in poor image quality.
By dividing the pixel matrix into multiple sub-pixel units, each sub-pixel unit contains MxM pixels, the pixels of the sub-pixel units are controlled to emit light at the same time to generate a sub-test image, the imaging device is used to capture and superimpose the sub-test images to form a test image, and the pixel brightness is compared with the area Brightness, set a brightness threshold to identify defective pixels and classify image defect levels based on visible range.
Accurately identify and classify defective pixels in virtual images, improve image quality, and achieve effective detection and processing of near-eye display defects.
Smart Images

Figure CN119013718A_ABST
Abstract
Description
Defect Detection Methods for Near-Eye Displays Technical Field
[0001] The present invention relates to the technical field of near-eye displays, and more particularly, to a defect detection method for near-eye displays. Background Art
[0002] Near-eye displays generally refer to augmented reality (AR), virtual reality (VR), and head-up / head-mount displays. A key metric for evaluating image quality is visual artifacts, or image defects. In the field of near-eye displays, image defects are particularly serious. Generally speaking, defects in the virtual image rendered in a near-eye display originate from defective pixels in the image generator (e.g., Micro LED / OLED). These defects can appear as uneven brightness at points, lines, or areas.
[0003] Existing technologies use traditional imaging systems (e.g., cameras) to capture and analyze images. However, due to crosstalk between pixels in virtual images or low image resolution in near-eye displays, defects are difficult to detect using traditional measurement methods. In reality, defects still exist, resulting in poor image quality. In other words, detecting defects in near-eye displays using existing imaging systems remains challenging.
[0004] Therefore, it is necessary to develop more advanced test modes to effectively detect defects in virtual images in near-eye displays, so as to detect, identify and classify defects in virtual images, and then perform subsequent processing and defect resolution.
[0005] Summary of the Invention
[0006] In view of this, the object of the present invention is to provide a defect detection method for near-eye displays, which can overcome the defects of the prior art, achieve the purpose of detecting defects in virtual images and identifying and classifying them, and effectively improve the image quality of virtual images.
[0007] To achieve the above object, the present invention discloses a defect detection method for a near-eye display, which is characterized by comprising the following steps:
[0008] Dividing a pixel matrix into a plurality of sub-pixel units, wherein each of the sub-pixel units comprises MxM pixels, where M is a natural number greater than 2;
[0009] sequentially controlling the pixels located at the same position among the M×M pixels of the sub-pixel units to emit light simultaneously to generate M×M sub-test images in an image generator;
[0010] sequentially capturing the sub-test images generated by the image generator using an imaging device;
[0011] superimposing the sub-test images to generate a test image corresponding to a virtual image, wherein each pixel in the test image has a pixel brightness;
[0012] comparing the pixel brightness of at least one pixel in the test image with a region pixel brightness of a region of pixels in the test image;
[0013] When the pixel brightness of the at least one pixel is less than a brightness threshold of the pixel brightness in the region, setting the at least one pixel as a defective pixel; and
[0014] When a visible range of the test image corresponding to the virtual image includes the defective pixel, the virtual image is classified as a defective image.
[0015] The brightness threshold is between 50% and 70% of the brightness of pixels in the area.
[0016] After the step of comparing the pixel brightness of at least one pixel in the test image with the region pixel brightness of the region pixels in the test image, the method further comprises the following steps:
[0017] When the number of pixels in a unit area of the test image whose pixel brightness is less than the brightness threshold of the pixel brightness in the area is greater than a set number threshold, the unit area is set as a defect area.
[0018] The visible range includes a first visible range, a second visible range, a third visible range, and a fourth visible range, which are arranged in ascending order. When the visible range of the test image corresponding to the virtual image includes the defective pixel, the step of classifying the virtual image as a defective image further includes the following steps:
[0019] When the first visible range of the test image includes the defective pixel or the second visible range includes the defective area, classifying the virtual image as a fifth-level defective image;
[0020] When the second visible range of the test image includes the defective pixel or the third visible range includes the defective area, classifying the virtual image as a fourth-level defective image;
[0021] When the defective pixel is included in the third visible range of the test image or the defective area is included in the fourth visible range, classifying the virtual image as a third-level defective image;
[0022] When the defective pixel is included in the fourth visible range of the test image, classifying the virtual image as a second-level defective image; and
[0023] When the fourth visible range of the test image does not include the defective pixel, the virtual image is classified as a first-level defective image.
[0024] The shape of the visual range is circular.
[0025] The shape of the visible range is an ellipse.
[0026] Before the step of dividing the pixel matrix into a plurality of sub-pixel units, the method further includes the following steps:
[0027] The pixel matrix and the virtual image are corrected and aligned.
[0028] In summary, the defect detection method for near-eye displays of the present invention has the following beneficial effects: it can accurately find defective pixels in a virtual image and identify and classify them, so as to facilitate subsequent defect processing, thereby ultimately achieving the purpose of improving the image quality of the virtual image. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] FIG1 shows a flowchart of a defect detection method for a near-eye display according to a specific embodiment of the present invention.
[0030] FIG. 2A shows a schematic diagram of a pixel matrix of a near-eye display according to the present invention.
[0031] FIG. 2B is a schematic diagram showing the principle of the defect detection method for a near-eye display according to the present invention.
[0032] FIG. 3 is a schematic diagram showing a defective pixel of a defect detection method for a near-eye display according to the present invention.
[0033] FIG. 4 shows a flowchart of a defect classification method for a near-eye display according to a specific embodiment of the present invention.
[0034] FIG5 is a schematic diagram showing a test image and a visible range according to a specific embodiment of the present invention.
[0035] FIG6 shows a schematic diagram of a test image and a visible range according to a specific embodiment of the present invention. DETAILED DESCRIPTION
[0036] To make the advantages, spirit, and features of the present invention more easily and clearly understood, the following detailed description and discussion will be given using examples and with reference to the accompanying drawings. It should be noted that these examples are merely representative of the present invention, and the specific methods, devices, conditions, materials, etc. exemplified therein are not intended to limit the present invention or the corresponding embodiments.
[0037] The terms used in the various embodiments disclosed in the present invention are only used for the purpose of describing specific embodiments and are not intended to limit the various embodiments disclosed in the present invention. The singular form used in the specification also includes the plural form, unless the context clearly indicates otherwise. Unless otherwise specified, all terms used in this specification (including technical terms and scientific terms) have the same meanings as those generally understood by those skilled in the art in the field to which the various embodiments disclosed in the present invention belong. The above terms (such as those defined in generally used dictionaries) will be interpreted as having the same meanings as in the context of the same technical field, and will not be interpreted as having idealized meanings or overly formal meanings unless the term is clearly defined in the various embodiments disclosed in the present invention.
[0038] Please refer to Figures 1, 2A, and 2B. Figure 1 shows a flowchart of the steps of a near-eye display defect detection method according to a specific embodiment of the present invention. Figure 2A shows a schematic diagram of a pixel matrix 1 of a near-eye display of the present invention. Figure 2B shows a schematic diagram of the principle of the near-eye display defect detection method of the present invention. As shown in FIG1 , the defect detection method for a near-eye display of the present invention includes the following steps: step S1: dividing a pixel matrix into a plurality of sub-pixel units, wherein each sub-pixel unit includes M×M pixels, where M is a natural number greater than 2; step S2: sequentially controlling pixels located at the same position among the M×M pixels of the sub-pixel units to emit light simultaneously to generate M×M sub-test images in an image generator; step S3: sequentially capturing the M×M sub-test images displayed by the image generator using an imaging device; step S4: superimposing and integrating the sub-test images in a program to form a complete test image corresponding to a virtual image, wherein each pixel in the test image has a pixel brightness; step S5: comparing the pixel brightness of pixels in the test image with the regional pixel brightness of pixels in a region of the test image; step S6: setting the pixel as a defective pixel when the pixel brightness of the pixel is less than a brightness threshold of the regional pixel brightness; and step S7: classifying the virtual image as a defective image when a visible range of the test image corresponding to the virtual image includes the defective pixel.
[0039] As shown in Figures 1 and 2A , a near-eye display may include a pixel matrix 1 for displaying an image or a virtual image. The pixel matrix 1 includes a plurality of pixels 12. In this embodiment, the pixel matrix 1 is 9x9 in size. In practical applications, the size of the pixel matrix and the number of pixels may be determined based on the resolution of the near-eye display. The shape of the pixel matrix is not limited to the square shown in Figure 2A but may also be rectangular or other shapes.
[0040] In this specific embodiment, the image generator can generate a virtual image for display by the pixel matrix 1. The size of the virtual image can correspond to the size of the pixel matrix 1. In practical applications, the virtual image can be a color image, but is not limited to this. It can also be an all-white, all-gray, or other solid color image. Furthermore, the positions of the pixels 12 corresponding to the virtual image each contain image data (e.g., optical properties such as brightness and chromaticity). The image generator can drive the pixels 12 of the pixel matrix 1 according to the image data to render the virtual image from the near-eye display.
[0041] Furthermore, in this embodiment, the near-eye display defect detection method may further include the following steps: calibrating and aligning the pixel matrix 1 and the virtual image. In practice, the pixel matrix 1 and the virtual image may be aligned using a near-eye display image calibrator or adjustment mechanism so that the position of each pixel 12 in the pixel matrix 1 corresponds to the position of the virtual image.
[0042] As shown in FIG2B , in step S1, the pixel matrix is divided into a plurality of sub-pixel units 11. In this embodiment, the pixel matrix is divided into four sub-pixel units 11, that is, each sub-pixel unit 11 includes 3×3 pixels 12. In actual applications, the number of sub-pixel units and pixels in the sub-pixel units can also be determined according to the size of the pixel matrix.
[0043] In step S2, the image generator controls the pixels 12 located at the same position among the pixels 12 of the sub-pixel unit 11 to emit light simultaneously. As shown in Figure 2B, the image generator first simultaneously controls the pixel 12 in the upper left corner of all sub-pixel units 11 to emit light, and controls the other pixels not to emit light. In step S3, the imaging system can capture / capture the image presented by the current pixel matrix 1 to generate a sub-test image 2A. The imaging device can be a light measuring device (LMD), and the light measuring device can include a colorimeter / camera and an NED / lens. When the imaging system generates the sub-test image, it also records the pixel brightness value of each illuminated pixel 12 in the sub-test image.
[0044] Furthermore, after the imaging system captures the sub-test image 2A in which the pixel 12 in the upper left corner of all sub-pixel units 11 is illuminated, the image generator then simultaneously controls the pixel 12 in the middle of the first row of all sub-pixel units 11 to illuminate, and controls the other pixels not to illuminate, and the imaging system then captures its sub-test image 2B. In this specific embodiment, the sub-pixel unit 11 includes 9 pixels, so the image generator sequentially controls the pixels 12 located at the same position in the sub-pixel unit 11 to illuminate simultaneously. When the image generator simultaneously controls the pixel 12 in the lower right corner of all sub-pixel units 11 to illuminate and the imaging system then captures its sub-test image 2I, that is, when all pixels 12 in the sub-pixel unit 11 are sequentially lit, 9 sub-test images will be generated.
[0045] In this embodiment, because each sub-pixel unit 11 comprises a 3x3 sub-pixel matrix, the spacing between each illuminated pixel 12 in each sub-test image is three times the spacing between adjacent pixels. Therefore, when the pixels 12 in each sub-pixel unit 11 illuminate, all illuminated pixels maintain the maximum distance between them, and the brightness of each pixel 12 is not affected by interference or crosstalk from the brightness of other pixels. This allows the imaging system to record the true brightness value of each illuminated pixel 12 when capturing its sub-test image.
[0046] In step S4, after the imaging system generates nine sub-test images, it may superimpose all of the sub-test images to generate a test image 3 corresponding to the virtual image. In practical applications, the imaging system may sequentially display the sub-test images to render the test image 3 corresponding to the virtual image. Because each sub-test image is an image of the light emitted by pixels 12 at different and non-overlapping locations within the sub-pixel unit 11, after the imaging system displays all of the sub-test images, all of the pixels 12 in the pixel matrix 1 are illuminated, forming a complete test image 3. Furthermore, the brightness of each pixel 12 in the test image 3 includes a true luminance value.
[0047] Please refer to Figures 1 and 3 . Figure 3 shows a schematic diagram of a defective pixel in the defect detection method for a near-eye display of the present invention. The image presented in Figure 3 is a test image 3 generated by superimposing all sub-test images on the aforementioned imaging system to generate a corresponding virtual image. In step S5, after test image 3 is generated, the pixel brightness of pixel 12 in test image 3 is compared with the regional pixel brightness of a region of pixels in the test image. In practical applications, the near-eye display may further include a comparison and analysis chip. As shown in Figure 3 , when the comparison and analysis chip detects pixel 12', regional pixel 13' may be a 3x3 pixel adjacent to or adjacent to pixel 12', and the regional pixel brightness may be the average pixel brightness value of the 3x3 pixels. Since the brightness of each pixel 12 in test image 3 is a true luminous brightness value, the defect detection method for a near-eye display of the present invention detects defective pixels by identifying the brightness level between pixels and adjacent pixels. It is worth noting that the number of pixels in the regional pixel can also be determined according to needs or design.
[0048] In step S6, when the pixel brightness of a pixel is less than a brightness threshold of the regional pixel brightness, the pixel is set as a defective pixel. In practical applications, the brightness threshold can be set between 50% and 70% of the regional pixel brightness. As shown in FIG3 , when the brightness of pixel 12' is greater than 70% of the regional pixel brightness of regional pixel 13', pixel 12' can be set and defined as a defective pixel, or pixel 12' can be referred to as a darker point. In a specific embodiment, when the brightness of pixel 12' is greater than 50% of the regional pixel brightness of regional pixel 13' and less than 70% of the regional pixel brightness of regional pixel 13', pixel 12' can also be set and defined as a defective pixel. Furthermore, when the brightness of pixel 12" is less than 50% of the regional pixel brightness of regional pixel 13", pixel 12" can be set and defined as a defective pixel, or pixel 12" can be referred to as a dead point. Pixels 12' and 12" are point defective pixels.
[0049] In addition to detecting whether a single pixel is a defective pixel, the defect detection method of the near-eye display of the present invention can also detect multiple pixels. As shown in Figure 1, in this specific embodiment, after executing step S5, step S61 can also be executed: when the number of pixel brightnesses of multiple pixels in a unit area in the test image that is less than the brightness threshold of the regional pixel brightness is greater than the set number threshold, the unit area is set as a defective area. In actual applications, the set number threshold can be the number of defective pixels in the unit area exceeding a certain value, or the number of defective pixels exceeding 30% of the total number of pixels in the unit area, but is not limited to this. Assuming that the unit area is a 10x10 matrix, and the unit area contains 70 pixels that are all identified as defective pixels, the unit area is also defined as a defective area. In addition, when two adjacent or connected pixels or three pixels are all identified as defective pixels (as shown in areas B and B' in Figure 3), these defective pixels can be defined as linear defects.
[0050] Therefore, the defect detection method for near-eye displays of the present invention can establish a judgment standard for defective pixels through a test image corresponding to a virtual image, so as to effectively detect the severity and distribution range of the defective pixels, thereby forming an indicator for judging the quality of the virtual image.
[0051] In step S7, when a visible range of the test image corresponding to the virtual image contains a defective pixel, the virtual image is classified as a defective image. The present invention can also identify and define the location of the defective pixel, thereby determining whether the virtual image requires subsequent processing and optimization. In practical applications, the visible range can be the area covered by the field of view (FOV), and the near-eye display comparison analysis chip can determine and classify the virtual image as a defective image based on whether the visible range contains a defective pixel.
[0052] Please refer to Figures 4 and 5 together. Figure 4 shows a step flow chart of a defect classification method for a near-eye display according to a specific embodiment of the present invention. Figure 5 shows a schematic diagram of a test image and a visible range according to a specific embodiment of the present invention. As shown in Figure 5, in this specific embodiment, the visible range includes a first visible range A1, a second visible range A2, a third visible range A3, and a fourth visible range A4, which are arranged in order from small to large. Furthermore, the first visible range A1, the second visible range A2, the third visible range A3, and the fourth visible range A4 are circular and are located at the center of the test image 3 of the corresponding virtual image. As shown in Figures 1 and 4, step S7 in Figure 1 can be further as follows: Step S71: Determine whether the first visible range A1 of the test image 3 contains a point defect pixel or whether the second visible range A2 contains a defective area. If the judgment result is yes, the classified virtual image is classified as a fifth-level defect image (step S711). If the judgment result is no, step S72 is executed: whether the second visible range A2 of the test image 3 contains a point-defective pixel or whether the third visible range A2 contains a defective area. If the judgment result is yes, the classified virtual image is classified as a fourth-level defect image (step S721). If the judgment result is no, step S73 is executed: whether the third visible range A3 of the test image 3 contains a point-defective pixel or whether the fourth visible range A2 contains a defective area. If the judgment result is yes, the classified virtual image is classified as a third-level defect image (step S731). If the judgment result is no, step S74 is executed: whether the fourth visible range A4 of the test image 3 contains a point-defective pixel. If the judgment result is yes, the classified virtual image is classified as a second-level defect image (step S741). If the judgment result is no, the classified virtual image is classified as a first-level defect image (step S75).
[0053] In practical applications, the first visible range A1 is the area covered by a field of view angle of F15, the second visible range A2 is the area covered by a field of view angle of F30, the third visible range A3 is the area covered by a field of view angle of F45, and the fourth visible range A4 is the area covered by a field of view angle of F60. Furthermore, the higher the image level of the defective image, the higher the quality of the corresponding virtual image, with a level 1 defective image having the highest image level and a level 5 defective image having the lowest image level. Furthermore, when the image level of the virtual image is lower than that of the level 4 defective image, the near-eye display may issue a warning signal, indicating that the virtual image needs to be processed and optimized.
[0054] Furthermore, in this embodiment, when a line defect pixel is included between the first visible range A1 and the second visible range A2 of the test image 3, the virtual image can be classified as a fifth-level defect image. When a line defect pixel is included between the second visible range A2 and the third visible range A3 of the test image 3, the virtual image can be classified as a fourth-level defect image. When a line defect pixel is included between the third visible range A3 and the fourth visible range A4 of the test image 3, the virtual image can be classified as a third-level defect image.
[0055] The shape of the visible range can be other shapes besides the aforementioned circular shape. Please refer to Figure 6, which shows a schematic diagram of a test image 3' and visible range A' according to a specific embodiment of the present invention. As shown in Figure 6, the shape of visible range A' can also be an ellipse to accommodate the human field of view when using a near-eye display.
[0056] In summary, the defect detection method for near-eye displays of the present invention has the following beneficial effects: it formulates a judgment standard for defective pixels and can accurately find the defective pixels in the virtual image and identify and classify them, so as to effectively detect the severity and distribution range of the defective pixels, and is conducive to subsequent defect processing, thereby ultimately achieving the purpose of improving the image quality of the virtual image.
[0057] The above detailed description of the preferred embodiments is intended to more clearly describe the features and spirit of the present invention, but is not intended to limit the scope of the present invention by the preferred embodiments disclosed above. On the contrary, its purpose is to cover various changes and equivalent arrangements within the scope of the patent application to be applied for by the present invention. Although the present invention has been disclosed as above in the form of embodiments, it is not intended to limit the present invention. Anyone skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be determined by the scope of the appended patent application.
Claims
1. A method for detecting defects in near-eye displays, characterized in that The following steps are involved: Dividing a pixel matrix into a plurality of sub-pixel units, wherein each of the sub-pixel units comprises MxM pixels, where M is a natural number greater than 2; sequentially controlling the pixels located at the same position among the M×M pixels of the sub-pixel units to emit light simultaneously to generate M×M sub-test images in an image generator; sequentially capturing the sub-test images generated by the image generator using an imaging device; superimposing the sub-test images to generate a test image corresponding to a virtual image, wherein each pixel in the test image has a pixel brightness; comparing the pixel brightness of at least one pixel in the test image with a region pixel brightness of a region of pixels in the test image; When the pixel brightness of the at least one pixel is less than a brightness threshold of the pixel brightness in the region, setting the at least one pixel as a defective pixel; as well as When a visible range of the test image corresponding to the virtual image includes the defective pixel, the virtual image is classified as a defective image.
2. The defect detection method for a near-eye display according to claim 1, wherein: The brightness threshold is between 50% and 70% of the brightness of the pixels in the area.
3. The defect detection method for a near-eye display according to claim 1, wherein: After the step of comparing the pixel brightness of at least one pixel in the test image with the region pixel brightness of the region pixels in the test image, the method further includes the following steps: When the number of pixels in a unit area of the test image whose pixel brightness is less than the brightness threshold of the pixel brightness in the area is greater than a set number threshold, the unit area is set as a defect area.
4. The defect detection method for a near-eye display according to claim 3, wherein: The visible range includes a first visible range, a second visible range, a third visible range, and a fourth visible range, which are arranged in ascending order. When the visible range of the test image corresponding to the virtual image includes the defective pixel, the step of classifying the virtual image as a defective image further includes the following steps: When the first visible range of the test image includes the defective pixel or the second visible range includes the defective area, classifying the virtual image as a fifth-level defective image; When the second visible range of the test image includes the defective pixel or the third visible range includes the defective area, classifying the virtual image as a fourth-level defective image; When the defective pixel is included in the third visible range of the test image or the defective area is included in the fourth visible range, classifying the virtual image as a third-level defective image; When the defective pixel is included in the fourth visible range of the test image, classifying the virtual image as a second-level defective image; and When the fourth visible range of the test image does not include the defective pixel, the virtual image is classified as a first-level defective image.
5. The defect detection method for a near-eye display according to claim 1, wherein: The shape of the visual range is circular.
6. The defect detection method for a near-eye display according to claim 1, wherein: The shape of the visual range is elliptical.
7. The defect detection method for a near-eye display according to claim 1, wherein: Before the step of dividing the pixel matrix into a plurality of sub-pixel units, the method further includes the following steps: The pixel matrix and the virtual image are corrected and aligned.
