Detection method and device for high-precision automatic optical equipment

Abstract

The invention discloses a detection method and device for high-precision automatic optical equipment, and the method comprises the steps: carrying out the multi-scale feature extraction of an obtained target image, so as to obtain a target detection result of a to-be-detected target.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a detection method and apparatus for high-precision automatic optical equipment. Background Art

[0002] Wafer is short for a circular piece of semiconductor crystal. It is a thin slice of cylindrical semiconductor crystal. Due to its round shape, it is called a wafer. During the wafer production process, it is tested to determine whether the produced wafers meet the requirements. For example, the color uniformity of the wafer surface is tested. If it does not meet the requirements, it will be reworked or scrapped. However, existing wafers are usually tested manually, and the test results are subjective, time-consuming and labor-intensive. Summary of the Invention

[0003] The embodiments of the present application provide a detection method and apparatus for high-precision automatic optical equipment.

[0004] In a first aspect, the present application provides a target detection method, comprising:

[0005] Acquire a target image of a target to be detected;

[0006] Perform multi-scale feature extraction on the target image to obtain first communication information and second channel information;

[0007] Based on the first channel information and the second channel information, a target detection result of the target to be detected is determined.

[0008] In a second aspect, the present application provides a target detection device, comprising:

[0009] An image acquisition module is used to acquire a target image of a target to be detected;

[0010] A feature extraction module, configured to perform multi-scale feature extraction on a target image to obtain first communication information and second channel information;

[0011] The target detection module is used to determine the target detection result of the target to be detected based on the first channel information and the second channel information.

[0012] In a third aspect, the present application further provides a computer device, comprising:

[0013] one or more processors;

[0014] Memory; and

[0015] One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement any target detection method in the first aspect.

[0016] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which is loaded by a processor to execute the steps in any one of the target detection methods in the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 Schematic diagram of a target detection system according to an embodiment of the present invention;

[0019] Figure 2 is a flow chart of an embodiment of a target detection method provided by an embodiment of the present invention;

[0020] Figure 3 is a flowchart of a specific embodiment of multi-scale feature extraction provided by an embodiment of the present invention;

[0021] Figure 4 This is a schematic diagram of a structure in which an image is cropped according to a 4*4 scale, as provided by an embodiment of the present invention;

[0022] Figure 5 This is a schematic diagram of a structure in which an image provided by an embodiment of the present invention is cropped according to an 8*8 scale;

[0023] Figure 6 is a principle block diagram of a target detection device provided by an embodiment of the present invention;

[0024] Figure 7 It is a schematic diagram of the structure of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0026] In the description of the present application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present application. In addition, the terms "first", "second", "third" and the like are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, features defined as "first", "second", "third", etc. may explicitly or implicitly include one or more of the said features. In the description of the present application, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0027] In this application, the word "exemplary" is used to mean "serving as an example, illustration, or illustration." Any embodiment described in this application as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments. The following description is given to enable any person skilled in the art to implement and use the present application. In the following description, details are listed for the purpose of explanation. It should be understood that one of ordinary skill in the art can recognize that the present application can be implemented without using these specific details. In other instances, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present application with unnecessary details. Therefore, the present application is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in this application.

[0028] It should be noted that since the method of the embodiment of the present application is executed in a computer device, the processing objects of each computer device exist in the form of data or information. For example, time is actually time information. It can be understood that if size, quantity, position, etc. are mentioned in subsequent embodiments, the corresponding data exist for the computer device to process. The details will not be repeated here.

[0029] The embodiments of the present application provide a detection method and apparatus for high-precision automatic optical equipment, which are described in detail below.

[0030] See also Figure 1 , Figure 1 This is a schematic diagram of a target detection system provided in an embodiment of the present application. The target detection system may include a computer device 100, in which a target detection device is integrated, such as Figure 1 Computer equipment in.

[0031] In the embodiment of the present application, the computer device 100 is mainly used to obtain a target image of a target to be detected; perform multi-scale feature extraction on the target image to obtain first communication information and second channel information; and determine a target detection result of the target to be detected based on the first channel information and the second channel information, which can improve the accuracy of the detection result and save time and effort.

[0032] In the embodiments of the present application, the computer device 100 may be an independent server, or a server network or server cluster composed of servers. For example, the computer device 100 described in the embodiments of the present application includes, but is not limited to, a computer, a network host, a single network server, a set of multiple network servers, or a cloud server composed of multiple servers. A cloud server is composed of a large number of computers or network servers based on cloud computing.

[0033] It is understood that the computer device 100 used in the embodiments of the present application can be a device that includes both receiving and transmitting hardware, that is, a device that has receiving and transmitting hardware capable of performing two-way communication over a two-way communication link. Such a device may include: a cellular or other communication device that has a single-line display, a multi-line display, or a cellular or other communication device without a multi-line display. The specific computer device 100 can be a desktop terminal or a mobile terminal. The computer device 100 can also be a mobile phone, a tablet computer, a laptop computer, etc.

[0034] Those skilled in the art will understand that Figure 1 The application environment shown in the figure is only one application scenario of the present application solution and does not constitute a limitation on the application scenario of the present application solution. Other application environments may also include Figure 1 More or fewer computer devices as shown in Figure 1 Only one computer device is shown in the figure. It can be understood that the target detection system can also include one or more other services, which are not limited here.

[0035] In addition, if Figure 1 As shown, the target detection system may further include a memory 200 for storing data, such as image data, such as target image, first channel image, second channel image, etc., such as channel information, such as first channel information, second channel information, etc.

[0036] It should be noted that Figure 1The scenario diagram of the target detection system shown is only an example. The target detection system and scenario described in the embodiment of the present application are intended to more clearly illustrate the technical solution of the embodiment of the present application, and do not constitute a limitation on the technical solution provided by the embodiment of the present application. Ordinary technicians in this field can know that with the evolution of the target detection system and the emergence of new business scenarios, the technical solution provided by the embodiment of the present application is also applicable to similar technical problems.

[0037] First, an embodiment of the present application provides a detection method for high-precision automatic optical equipment. The executor of the detection method for high-precision automatic optical equipment is a target detection device, which is applied to a computer device. The detection method for high-precision automatic optical equipment includes: obtaining a target image of the target to be detected; performing multi-scale feature extraction on the target image to obtain first communication information and second channel information; and determining a target detection result of the target to be detected based on the first channel information and the second channel information.

[0038] like Figure 2 FIG. 2 is a flow chart of an embodiment of a target detection method in an embodiment of the present application. The target detection method may include the following steps S201 to S203, which are specifically as follows:

[0039] S201: Acquire a target image of a target to be detected.

[0040] The target to be detected is a target object that needs to be detected. For example, the target to be detected is a wafer that needs to be detected for color uniformity. The target image is an image obtained by photographing the target to be detected by the imaging module. For example, when the target to be detected is a wafer, the target image is an image obtained by photographing the wafer by the imaging module.

[0041] Specifically, the target image can be an image of the target to be detected acquired by an imaging module configured by the computer device itself, or can be an image of the target to be detected acquired by an imaging module of another computer device through a network, Bluetooth, infrared, etc. For example, when the target detection method of the present application is applied to a smartphone, the smartphone can directly acquire the target image through its own imaging module. When the target detection method of the present application is applied to a server, the server can acquire the target image through the smartphone's imaging module and obtain the target image from the smartphone through a network, Bluetooth, infrared, etc.

[0042] S202 : Perform multi-scale feature extraction on the target image to obtain first communication information and second channel information.

[0043] The target image includes a first channel image and a second channel image. Multi-scale feature extraction refers to performing feature extraction on the first channel image and the second channel image respectively from multiple different image scales. For example, the multiple different image scales include image scales of 4*4 and 8*8. The multi-scale feature extraction includes performing feature extraction on the first channel image at image scales of 4*4 and 8*8, and performing feature extraction on the second channel image at image scales of 4*4 and 8*8.

[0044] The first channel information is the standard deviation of the first channel image obtained by extracting features of the first channel image from multiple different image scales. The first channel information describes the variation amplitude of pixel values ​​in the first channel image and can represent the color uniformity of the first channel image. For example, if the first channel image is the A channel image in the LAB image, the first channel information is the standard deviation of the A channel image.

[0045] The second channel information is the standard deviation of the second channel image obtained by extracting features of the second channel image from multiple different image scales. The second channel information describes the variation amplitude of the pixel values ​​in the second channel image and can represent the color uniformity of the second channel image. For example, if the second channel image is the B channel image in the LAB image, the second channel information is the standard deviation of the B channel image.

[0046] This embodiment performs multi-scale feature extraction on the target image, and then performs target detection based on the first channel information and the second channel information obtained by the multi-scale feature extraction. The target image can be detected by combining features of different scales of the target image to improve the accuracy of the detection result.

[0047] In a specific implementation, Figure 3 As shown, in the above step S202, multi-scale feature extraction is performed on the target image to obtain the first communication information and the second channel information, which may include the following steps S301 to S303, specifically as follows:

[0048] S301 : Preprocess the target image to obtain a first channel image and a second channel image.

[0049] Preprocessing includes image conversion and channel separation. Image conversion of the target image can convert the target image from one color space to another color space. For example, if the target image is an RGB image, image conversion of the target image can convert the target image from RGB to an HSV image or a LAB image. The first channel image and the second channel image are images of two different channels separated from the target image after image conversion. For example, if the target image after image conversion is an HSV image, the first channel image is the H channel image, and the second channel image is the S channel image; for another example, if the target image after image conversion is a LAB image, the first channel image is the A channel image, and the second channel image is the B channel image.

[0050] In a specific implementation, preprocessing the target image in step 301 to obtain the first channel image and the second channel image may include the following steps:

[0051] (1) Perform alignment processing on the target image to obtain the aligned target image.

[0052] Performing alignment processing on the target image can unify the image brightness expression of the target image and improve the accuracy of the detection result. In a specific embodiment, performing alignment processing on the target image to obtain the aligned target image can include the following steps:

[0053] (a) Performing image conversion on the target image to obtain a second converted image; wherein the second converted image includes a third channel image, a fourth channel image, and a fifth channel image.

[0054] Performing image conversion on the target image can convert the target image from one color space to another color space. In one specific embodiment, the target image is an RGB image, and performing image conversion on the target image can convert the RGB image into an HSV image. The second converted image is a three-channel image, which includes a third channel image, a fourth channel image, and a fifth channel image. For example, when the second converted image is an HSV image, the second converted image includes a V channel image (third channel image), an H channel image (fourth channel image), and an S channel image (fifth channel image).

[0055] (b) Determining brightness information of the target image based on the third channel image, and determining a brightness alignment factor based on the brightness information and preset target brightness information.

[0056] In a specific embodiment, the brightness information is the pixel average value of each pixel in the third channel image, and the step of determining the brightness information of the target image based on the third channel image includes: determining the pixel average value of the third channel image based on the pixel value of each pixel in the third channel image; and determining the pixel average value as the brightness information of the target image.

[0057] The target brightness information is a preset target brightness value. The target brightness information can be set according to user needs. For example, the target brightness information can be set to 120. The process of determining the brightness alignment factor can be expressed as: bright_factor = target_gray / current_gray, where bright_factor represents the brightness alignment factor, target_gray represents the target brightness information, and current_gray represents the brightness information of the target image.

[0058] (c) Correcting the third channel image based on the brightness alignment factor to obtain a corrected third channel image.

[0059] When correcting the third channel image based on the brightness alignment factor, the pixel value of each pixel in the third channel image may be multiplied by the brightness alignment factor, and then the pixel values ​​greater than 255 in the third channel image may be set to 255, thereby obtaining a corrected third channel image.

[0060] (d) Determine a corrected second converted image based on the corrected third channel image, fourth channel image, and fifth channel image.

[0061] The corrected second converted image and the second converted image are images in the same color space. Both the corrected second converted image and the second converted image contain a third channel image, a fourth channel image, and a fifth channel image. The difference between the corrected second converted image and the second converted image is that the third channel image in the second converted image is the original third channel image, while the third channel image in the corrected second converted image is the corrected third channel image. For example, if the second converted image is an HSV image, the corrected second converted image has the same H channel image and S channel image as those in the second converted image, and the V channel image in the corrected second converted image is the V channel image corrected by the brightness alignment factor.

[0062] (e) Performing image transformation on the corrected second transformed image to obtain the aligned target image.

[0063] Performing image conversion on the corrected second conversion image can convert the corrected second conversion image from one color space to another color space. For example, when the target image is an RGB image, the corrected second conversion image is an HSV image. Performing image conversion on the corrected second conversion image can convert the corrected second conversion image from an HSV image to an RGB image, so that the aligned target image and the target image are both RGB images.

[0064] (2) Performing image conversion on the aligned target image to obtain a first converted image.

[0065] Performing image conversion on the aligned target image can convert the aligned target image from one color space to another color space. For example, if the aligned target image is an RGB image, performing image conversion on the aligned target image can convert the aligned target image from the RGB color space to the LAB color space, so that the first converted image can be a LAB image.

[0066] (3) Perform channel separation on the first conversion image to obtain a first channel image and a second channel image.

[0067] The first conversion image includes multiple channel images. By performing channel separation on the first conversion image, a first channel image and a second channel image of different channels can be obtained. For example, the first conversion image is a LAB image. By performing channel separation on the first conversion image, an A channel image and a B channel image can be obtained.

[0068] S302 : Perform multi-scale feature extraction on the first channel image to obtain first channel information.

[0069] Performing multi-scale feature extraction on the first channel image refers to performing feature extraction on the first channel image at least at two scales, and the first channel information is image information obtained by performing multi-scale feature extraction on the first channel image. In a specific implementation, performing multi-scale feature extraction on the first channel image in step S302 to obtain the first channel information may include the following steps:

[0070] (1) Perform multi-scale cropping on the first channel image to obtain multiple first images and multiple second images.

[0071] The plurality of first images are image blocks obtained by cropping the first channel image according to a first scale, and the plurality of second images are image blocks obtained by cropping the second channel image according to a second scale, wherein the first scale and the second scale are different scales. For example, referring to Figure 4 As shown, the first scale is 4*4, and the first channel image is evenly cut into 4*4 parts, and 16 first images P1_n can be obtained, where n=1...16; refer to Figure 5 As shown, the second scale is 8*8, and the first channel image is evenly cropped into 8*8 parts, so that 64 second images P1_n can be obtained, where n=1...64.

[0072] Of course, this embodiment can also crop the first channel image according to other scales in addition to the first scale and the second scale to obtain other images in addition to multiple first images and multiple second images. For example, in addition to cropping the first channel image according to 4*4 and 8*8, the first channel image can also be cropped according to scales such as 16*16 and 32*32.

[0073] (2) For any first image among the plurality of first images, determine first pixel information of the first image; the first pixel information is a pixel mean value of pixel points in the first image.

[0074] The first pixel information is the pixel mean of the pixel points in the first image. After obtaining multiple first images, for any first image among the multiple first images, the pixel values ​​of multiple pixel points in the first image are averaged to obtain the first pixel information corresponding to the first image.

[0075] (3) For any second image among the plurality of second images, second pixel information of the second image is determined; the second pixel information is a pixel mean value of pixel points in the second image.

[0076] The second pixel information is the pixel mean of the pixel points in the second image. After obtaining multiple second images, for any second image among the multiple second images, the pixel values ​​of multiple pixel points in the second image are averaged to obtain the second pixel information corresponding to the second image.

[0077] (4) Determine first channel information based on the first pixel information and the second pixel information.

[0078] Specifically, determining the first channel information based on the first pixel information and the second pixel information may include: determining the first image information based on the first pixel information; determining the second image information based on the first pixel information, the first image information and the number of first images; determining the third image information based on the second pixel information; determining the fourth image information based on the second pixel information, the third image information and the number of second images; and performing weighted processing on the second image information and the fourth image information based on the first weight information corresponding to the second image information and the second weight information corresponding to the fourth image information to obtain the first channel information.

[0079] The first image information is the mean of the first pixel information of the plurality of first images, and the second image information is the standard deviation of the target image determined based on the first pixel information, the first image information, and the number of first images. The calculation process of the second image information can be expressed as follows: Wherein, std_P1 represents the second image information, P1_i represents the first pixel information corresponding to the i-th first image, n1 represents the number of first images, and μ1 represents the first image information.

[0080] The third image information is the mean of the second pixel information of the plurality of second images. The fourth image information is the standard deviation of the target image determined based on the second pixel information, the third image information, and the number of second images. The calculation process of the fourth image information can be expressed as follows: Wherein, std_P2 represents the fourth image information, P2_i represents the second pixel information corresponding to the i-th second image, n2 represents the number of second images, and μ2 represents the third image information.

[0081] Further, the calculation process of the first channel information can be expressed as: img_a_std = α_1*std_P1+β_1*std_P2, img_a_std represents the first channel information, std_P1 represents the second image information, std_P2 represents the fourth image information, α_1 represents the first weight information, β_1 represents the second weight information, α_1 and β_1 can be set as needed, for example, α_1 is set to 0.5, and β_1 is set to 0.5.

[0082] It should be noted that the above embodiment only lists the specific process of performing two-scale feature extraction on the first channel image to obtain the first channel information. In this embodiment, more scale feature extraction (such as three-scale feature extraction, four-scale feature extraction, M-scale feature extraction, etc.) can also be performed on the first channel image. When M scale feature extractions are performed on the first channel image, M image information of the first channel image can be obtained. When determining the first channel information, the M image information of the first channel image is weighted.

[0083] S303: Perform multi-scale feature extraction on the second channel image to obtain second channel information.

[0084] Performing multi-scale feature extraction on the second channel image refers to performing feature extraction on the second channel image at least at two scales, and the second channel information is image information obtained by performing multi-scale feature extraction on the second channel image. In a specific implementation, performing multi-scale feature extraction on the second channel image in step S303 to obtain the second channel information may include the following steps:

[0085] (1) Perform multi-scale cropping on the second channel image to obtain multiple third images and multiple fourth images.

[0086] The plurality of third images are image blocks obtained by cropping the second channel image according to the third scale, and the plurality of fourth images are image blocks obtained by cropping the second channel image according to the fourth scale, wherein the third scale and the fourth scale are different scales. Figure 4 As shown, the third scale is 4*4, and the second channel image is evenly cut into 4*4 parts, and 16 third images P1_n can be obtained, where n=1...16; refer to Figure 5 As shown, the fourth scale is 8*8, and the second channel image is evenly cropped into 8*8 parts, so that 64 fourth images P1_n can be obtained, where n=1...64.

[0087] It should be noted that the third scale can be the same as or different from the first or second scale; the fourth scale can be the same as or different from the first or second scale. For example, the first and third scales are both 4*4, and the second and fourth scales are both 8*8.

[0088] Similar to the aforementioned multi-scale cropping of the first channel image, this embodiment can also crop the second channel image at scales other than the third and fourth scales to obtain images other than multiple third and fourth images. For example, in addition to cropping the second channel image at 4*4 and 8*8 scales, the second channel image can also be cropped at scales such as 16*16 and 32*32.

[0089] (2) For any third image among the plurality of third images, determine third pixel information of the third image; the third pixel information is a pixel mean value of pixel points in the third image.

[0090] The third pixel information is the pixel mean of the pixel points in the third image. After obtaining multiple third images, for any third image among the multiple third images, the pixel values ​​of multiple pixel points in the third image are averaged to obtain the third pixel information corresponding to the third image.

[0091] (3) For any fourth image among the plurality of fourth images, determine fourth pixel information of the fourth image; the fourth pixel information is a pixel mean value of pixel points in the fourth image.

[0092] The fourth pixel information is the pixel mean of the pixel points in the fourth image. After obtaining multiple fourth images, for any fourth image among the multiple fourth images, the pixel values ​​of multiple pixel points in the fourth image are averaged to obtain the fourth pixel information corresponding to the fourth image.

[0093] (4) Determine the second channel information based on the third pixel information and the fourth pixel information.

[0094] Specifically, determining the second channel information based on the third pixel information and the fourth pixel information may include: determining the fifth image information based on the third pixel information; determining the sixth image information based on the third pixel information, the fifth image information and the number of third images; determining the seventh image information based on the fourth pixel information; determining the eighth image information based on the fourth pixel information, the seventh image information and the number of fourth images; and performing weighted processing on the sixth image information and the eighth image information based on the fifth weight information corresponding to the sixth image information and the sixth weight information corresponding to the eighth image information to obtain the second channel information.

[0095] The fifth image information is the mean of the third pixel information of the plurality of third images, and the sixth image information is the standard deviation of the target image determined based on the third pixel information, the fifth image information, and the number of third images. The calculation process of the sixth image information can be expressed as follows: Wherein, std_P3 represents the sixth image information, P3_i represents the third pixel information corresponding to the i-th third image, n3 represents the number of third images, and μ3 represents the fifth image information.

[0096] The seventh image information is the average of the fourth pixel information of the plurality of fourth images. The eighth image information is the standard deviation of the target image determined based on the fourth pixel information, the seventh image information, and the number of fourth images. The calculation formula for the eighth image information is: Wherein, std_P4 represents the eighth image information, P4_i represents the fourth pixel information corresponding to the i-th fourth image, n4 represents the number of fourth images, and μ4 represents the seventh image information.

[0097] The calculation process of the second channel information can be expressed as: img_b_std = α_2*std_P3+β_2*std_P4, img_b_std represents the second channel information, std_P3 represents the sixth image information, std_P4 represents the eighth image information, α_2 represents the fifth weight information, β_2 represents the sixth weight information, α_2 and β_2 can be set as needed, for example, α_2 is set to 0.5, and β_2 is set to 0.5.

[0098] It should be noted that the above embodiment only lists the specific process of performing two scale feature extractions on the second channel image to obtain the second channel information. In this embodiment, more scale feature extractions (such as three scale feature extractions, four scale feature extractions, and N scale feature extractions) can also be performed on the second channel image. When N scale feature extractions are performed on the second channel image, N image information of the second channel image can be obtained. When determining the second channel information, weighted processing is performed on the N image information of the second channel image.

[0099] S203 : Determine a target detection result of the target to be detected based on the first channel information and the second channel information.

[0100] The target detection result is used to measure the color uniformity of the target to be detected. For example, when the target to be detected is a wafer, the target detection result is used to measure the color uniformity of the wafer surface. The first communication information and the second channel information are obtained by performing multi-scale feature extraction on the target image, and then the target detection result of the target to be detected is determined based on the first channel information and the second channel information. Compared with the existing manual evaluation of the color uniformity of the target to be detected, the accuracy of the detection result can be improved, and time and effort can be saved.

[0101] In a specific implementation, determining the target detection result of the target to be detected based on the first channel information and the second channel information in step S203 may include the following step S401, specifically as follows:

[0102] S401 , performing weighted processing on the first channel information and the second channel information based on the third weight information corresponding to the first channel information and the fourth weight information corresponding to the second channel information to obtain a target detection result of the target to be detected.

[0103] The third weight information and the fourth weight information can be set according to user needs. For example, the third weight information is set to 0.7 and the fourth weight information is set to 0.3. After obtaining the third weight information and the fourth weight information, the first channel information and the second channel information are weighted and summed based on the third weight information and the fourth weight information to obtain the target detection result of the target to be detected. The calculation process of the target detection result can be expressed as: uniform_value = α_3*img_a_std + β_3*img_b_std, uniform_value represents the target detection result, img_a_std represents the first channel information, img_b_std represents the second channel information, α_3 represents the third weight information, and β_3 represents the fourth weight information.

[0104] In one specific embodiment, after determining the target detection result for the target to be detected based on the first channel information and the second channel information in step S203, the following steps may be performed: comparing the target detection result with a preset reference range; and outputting an alarm message when the target detection result is not within the reference range. The reference range can be set based on user needs. Different targets to be detected can have the same reference range, or different targets to be detected can have different reference ranges.

[0105] The alarm information includes but is not limited to output in one or more forms such as graphics, text, voice and color, for example, outputting a voice prompt of "the color uniformity of the wafer does not meet the requirements".

[0106] In summary, the target detection method provided by this embodiment obtains a target image of the target to be detected, performs multi-scale feature extraction on the target image, obtains first communication information and second channel information, and determines the target detection result of the target to be detected based on the first channel information and the second channel information. In this scheme, the first communication information and the second channel information are obtained by performing multi-scale feature extraction on the target image, and then the target detection result of the target to be detected is determined based on the first channel information and the second channel information. Compared with the existing method of manually evaluating the color uniformity of the target to be detected, the accuracy of the detection result can be improved, and time and labor can be saved. Furthermore, the second image information is determined based on multiple first images, and the fourth image information is determined based on multiple second images, the sixth image information is determined based on multiple third images, and the eighth image information is determined based on multiple fourth images, and then the first channel information is determined based on the second image information and the fourth image information, and the second channel information is determined based on the sixth image information and the eighth image information. The accuracy of the detection result can be further improved.

[0107] In order to better implement the target detection method in the embodiment of the present application, based on the target detection method, the embodiment of the present application also provides a target detection device, such as Figure 6 As shown, the target detection device 600 includes:

[0108] An image acquisition module 610 is used to acquire a target image of a target to be detected;

[0109] A feature extraction module 620 is configured to perform multi-scale feature extraction on the target image to obtain first communication information and second channel information;

[0110] The target detection module 630 is configured to determine a target detection result of the target to be detected based on the first channel information and the second channel information.

[0111] In an embodiment of the present application, multi-scale feature extraction is performed on the target image to obtain first communication information and second channel information, and then the target detection result of the target to be detected is determined based on the first channel information and the second channel information. Compared with the existing manual evaluation of the color uniformity of the target to be detected, the accuracy of the detection result can be improved, and time and effort can be saved.

[0112] In some embodiments of the present application, the feature extraction module 620 performs multi-scale feature extraction on the target image to obtain first communication information and second channel information, including:

[0113] Preprocess the target image to obtain a first channel image and a second channel image;

[0114] Performing multi-scale feature extraction on the first channel image to obtain first channel information;

[0115] Multi-scale feature extraction is performed on the second channel image to obtain second channel information.

[0116] In some embodiments of the present application, the feature extraction module 620 performs multi-scale feature extraction on the first channel image to obtain first channel information, including:

[0117] Performing multi-scale cropping on the first channel image to obtain multiple first images and multiple second images;

[0118] For any first image among the plurality of first images, determining first pixel information of the first image; the first pixel information is a pixel average value of pixels in the first image;

[0119] For any second image among the plurality of second images, determining second pixel information of the second image; the second pixel information is a pixel mean value of pixels in the second image;

[0120] First channel information is determined based on the first pixel information and the second pixel information.

[0121] In some embodiments of the present application, the feature extraction module 620 determines the first channel information based on the first pixel information and the second pixel information, including:

[0122] Determine first image information based on the first pixel information; the first image information is an average of first pixel information of the plurality of first images;

[0123] determining second image information based on the first pixel information, the first image information, and the number of first images;

[0124] Determining third image information based on the second pixel information; the third image information is an average of the second pixel information of the plurality of second images;

[0125] determining fourth image information based on the second pixel information, the third image information, and the number of second images;

[0126] The second image information and the fourth image information are weighted based on the first weight information corresponding to the second image information and the second weight information corresponding to the fourth image information to obtain first channel information.

[0127] In some embodiments of the present application, the feature extraction module 620 preprocesses the target image to obtain a first channel image and a second channel image, including:

[0128] Performing alignment processing on the target image to obtain an aligned target image;

[0129] Performing image conversion on the aligned target image to obtain a first converted image;

[0130] Channel separation is performed on the first conversion image to obtain a first channel image and a second channel image.

[0131] In some embodiments of the present application, the feature extraction module 620 performs alignment processing on the target image to obtain an aligned target image, including:

[0132] Performing image conversion on the target image to obtain a second converted image; wherein the second converted image includes a third channel image, a fourth channel image, and a fifth channel image;

[0133] Determining brightness information of a target image based on a third channel image, and determining a brightness alignment factor based on the brightness information and preset target brightness information; the brightness information is a pixel mean value of pixels in the third channel image;

[0134] Correcting the third channel image based on the brightness alignment factor to obtain a corrected third channel image;

[0135] Determine a corrected second converted image based on the corrected third channel image, fourth channel image, and fifth channel image;

[0136] Perform image transformation on the corrected second transformed image to obtain an aligned target image.

[0137] In some embodiments of the present application, the target detection module 630 determines the target detection result of the target to be detected based on the first channel information and the second channel information, including:

[0138] The first channel information and the second channel information are weighted based on the third weight information corresponding to the first channel information and the fourth weight information corresponding to the second channel information to obtain a target detection result of the target to be detected.

[0139] The present application also provides a computer device that integrates any one of the target detection devices provided in the present application. The computer device includes:

[0140] one or more processors;

[0141] Memory; and

[0142] One or more applications, wherein the one or more applications are stored in the memory and configured to execute, by the processor, the steps of the target detection method in any of the above target detection method embodiments.

[0143] The present application also provides a computer device that integrates any target detection device provided in the present application. Figure 7 , which shows a schematic diagram of the structure of the computer device involved in the embodiment of the present application, specifically:

[0144] The computer device may include one or more processing core processors 801, one or more computer readable storage media memories 802, a power supply 803, an input unit 804 and other components. Those skilled in the art will understand that Figure 7 The computer device structure shown in the figure does not constitute a limitation on the computer device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange components differently.

[0145] Processor 801 is the control center of the computer device. It connects the various components of the entire computer device using various interfaces and lines. By running or executing software programs and / or modules stored in memory 802 and accessing data stored in memory 802, it performs various functions of the computer device and processes data, thereby providing overall monitoring of the computer device. Optionally, processor 801 may include one or more processing cores. Preferably, processor 801 may integrate an application processor and a modem processor, wherein the application processor primarily processes the operating system, user interface, and application programs, while the modem processor primarily handles wireless communications. It is understood that the modem processor may not be integrated into processor 801.

[0146] The memory 802 can be used to store software programs and modules. The processor 801 executes various functional applications and data processing by running the software programs and modules stored in the memory 802. The memory 802 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 802 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage device. Accordingly, the memory 802 may also include a memory controller to provide the processor 801 with access to the memory 802.

[0147] The computer device also includes a power supply 803 for supplying power to various components. Preferably, the power supply 803 can be logically connected to the processor 801 via a power management system, thereby enabling the power management system to manage charging, discharging, and power consumption. The power supply 803 can also include one or more DC or AC power supplies, a recharging system, a power failure detection circuit, a power converter or inverter, a power status indicator, and other arbitrary components.

[0148] The computer device may further include an input unit 804, which may be configured to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.

[0149] Although not shown, the computer device may further include a display unit, etc., which will not be described in detail here. Specifically, in this embodiment, the processor 801 in the computer device will load the executable files corresponding to one or more application processes into the memory 802 according to the following instructions, and the processor 801 will run the application stored in the memory 802 to implement various functions as follows:

[0150] Acquire a target image of a target to be detected;

[0151] Perform multi-scale feature extraction on the target image to obtain first communication information and second channel information;

[0152] Based on the first channel information and the second channel information, a target detection result of the target to be detected is determined.

[0153] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be accomplished by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0154] To this end, an embodiment of the present application provides a computer-readable storage medium, which may include: a read-only memory (ROM), a random access memory (RAM), a disk, or an optical disk. A computer program is stored thereon, and the computer program is loaded by a processor to execute the steps of any target detection method provided in the embodiment of the present application. For example, the computer program loaded by the processor may execute the following steps:

[0155] Acquire a target image of a target to be detected;

[0156] Perform multi-scale feature extraction on the target image to obtain first communication information and second channel information;

[0157] Based on the first channel information and the second channel information, a target detection result of the target to be detected is determined.

[0158] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, please refer to the detailed description of other embodiments above and will not be repeated here.

[0159] In specific implementation, the above units or structures can be implemented as independent entities, or can be arbitrarily combined to implement as the same or several entities. The specific implementation of the above units or structures can refer to the previous method embodiments and will not be repeated here.

[0160] The specific implementation of the above operations can be found in the previous embodiments and will not be repeated here.

[0161] The above is a detailed introduction to a detection method and device for high-precision automatic optical equipment provided in the embodiments of the present application. Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, based on the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A target detection method, characterized in that: include: Acquire a target image of a target to be detected; Performing multi-scale feature extraction on the target image to obtain first communication information and second channel information; Based on the first channel information and the second channel information, a target detection result of the target to be detected is determined.

2. The method according to claim 1, characterized in that The performing multi-scale feature extraction on the target image to obtain first communication information and second channel information includes: Preprocessing the target image to obtain a first channel image and a second channel image; Performing multi-scale feature extraction on the first channel image to obtain first channel information; Multi-scale feature extraction is performed on the second channel image to obtain second channel information.

3. The method according to claim 2, characterized in that The performing multi-scale feature extraction on the first channel image to obtain first channel information includes: Performing multi-scale cropping on the first channel image to obtain a plurality of first images and a plurality of second images; For any first image among the plurality of first images, determining first pixel information of the first image; the first pixel information is a pixel mean value of pixels in the first image; For any second image among the plurality of second images, determining second pixel information of the second image, wherein the second pixel information is a pixel mean value of pixels in the second image; First channel information is determined based on the first pixel information and the second pixel information.

4. The method according to claim 3, characterized in that The determining first channel information based on the first pixel information and the second pixel information includes: Determine first image information based on the first pixel information; the first image information is an average of first pixel information of multiple first images; determining second image information based on the first pixel information, the first image information, and the number of the first images; Determine third image information based on the second pixel information; the third image information is an average of the second pixel information of the plurality of second images; determining fourth image information based on the second pixel information, the third image information, and the number of the second images; The second image information and the fourth image information are weighted based on the first weight information corresponding to the second image information and the second weight information corresponding to the fourth image information to obtain first channel information.

5. The method according to claim 2, characterized in that The preprocessing of the target image to obtain a first channel image and a second channel image includes: performing alignment processing on the target image to obtain an aligned target image; Performing image conversion on the aligned target image to obtain a first converted image; Channel separation is performed on the first converted image to obtain a first channel image and a second channel image.

6. The method according to claim 5, characterized in that The performing alignment processing on the target image to obtain the aligned target image includes: Performing image conversion on the target image to obtain a second converted image; wherein the second converted image includes a third channel image, a fourth channel image, and a fifth channel image; Determining brightness information of the target image based on the third channel image, and determining an alignment factor based on the brightness information and preset target brightness information; the brightness information is a pixel mean value of pixels in the third channel image; Correcting the third channel image based on the alignment factor to obtain a corrected third channel image; Determine a corrected second converted image based on the corrected third channel image, the fourth channel image, and the fifth channel image; Performing image conversion on the corrected second conversion image to obtain an aligned target image.

7. The method according to claim 1, characterized in that The determining, based on the first channel information and the second channel information, a target detection result of the target to be detected, includes: The first channel information and the second channel information are weighted based on the third weight information corresponding to the first channel information and the fourth weight information corresponding to the second channel information to obtain a target detection result of the target to be detected.

8. A target detection device, characterized in that: include: An image acquisition module is used to acquire a target image of a target to be detected; A feature extraction module, configured to perform multi-scale feature extraction on the target image to obtain first communication information and second channel information; A target detection module, configured to determine a target detection result of the target to be detected based on the first channel information and the second channel information; Preferably, the feature extraction module performs multi-scale feature extraction on the target image to obtain first communication information and second channel information, including: Preprocessing the target image to obtain a first channel image and a second channel image; Performing multi-scale feature extraction on the first channel image to obtain first channel information; performing multi-scale feature extraction on the second channel image to obtain second channel information; Preferably, the feature extraction module performs multi-scale feature extraction on the first channel image to obtain first channel information, including: Performing multi-scale cropping on the first channel image to obtain a plurality of first images and a plurality of second images; For any first image among the plurality of first images, determining first pixel information of the first image; the first pixel information is a pixel mean value of pixels in the first image; For any second image among the plurality of second images, determining second pixel information of the second image, wherein the second pixel information is a pixel mean value of pixels in the second image; determining first channel information based on the first pixel information and the second pixel information; Preferably, the feature extraction module determines the first channel information based on the first pixel information and the second pixel information, including: Determine first image information based on the first pixel information; the first image information is an average of first pixel information of multiple first images; determining second image information based on the first pixel information, the first image information, and the number of the first images; Determine third image information based on the second pixel information; the third image information is an average of the second pixel information of the plurality of second images; determining fourth image information based on the second pixel information, the third image information, and the number of the second images; performing weighted processing on the second image information and the fourth image information based on the first weight information corresponding to the second image information and the second weight information corresponding to the fourth image information to obtain first channel information; Preferably, the feature extraction module preprocesses the target image to obtain a first channel image and a second channel image, including: performing alignment processing on the target image to obtain an aligned target image; Performing image conversion on the aligned target image to obtain a first converted image; performing channel separation on the first converted image to obtain a first channel image and a second channel image; Preferably, the feature extraction module performs alignment processing on the target image to obtain an aligned target image, including: Performing image conversion on the target image to obtain a second converted image; wherein the second converted image includes a third channel image, a fourth channel image, and a fifth channel image; Determining brightness information of the target image based on the third channel image, and determining a brightness alignment factor based on the brightness information and preset target brightness information; the brightness information is a pixel mean value of pixels in the third channel image; Correcting the third channel image based on the brightness alignment factor to obtain a corrected third channel image; Determine a corrected second converted image based on the corrected third channel image, the fourth channel image, and the fifth channel image; performing image conversion on the corrected second converted image to obtain an aligned target image; Preferably, the target detection module determines the target detection result of the target to be detected based on the first channel information and the second channel information, including: The first channel information and the second channel information are weighted based on the third weight information corresponding to the first channel information and the fourth weight information corresponding to the second channel information to obtain a target detection result of the target to be detected.

9. A computer device, characterized in that: The computer device comprises: one or more processors; Memory; and One or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the processor to implement the target detection method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that A computer program is stored thereon, and the computer program is loaded by a processor to execute the steps in the target detection method according to any one of claims 1 to 7.

Images