A size detection method and device based on infrared rays

Through infrared detection method, the analysis of frame all-domain angioptic images and particle images is used, combined with spontaneous and environmental radiation, the reflection cadric thermal amplitude and jump value are calculated, which solves the low accuracy of size detection when the surface of the automobile display screen frame is twisted. Problem, achieving higher detection accuracy and result correction.

CN117589105BActive Publication Date: 2024-11-22DONGGUAN JUWEI ELECTRONIC TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202311496810.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-11-10
Publication Date
2024-11-22
Estimated Expiration
2043-11-10

AI Technical Summary

Technical Problem

In the prior art, when the surface of the automotive display frame is twisted, the dimensional detection accuracy is low, making it difficult to effectively correct the detection result.

Method used

Using infrared-based dimension detection method, the frame's full-domain anisotropic image is acquired, converted into a particle image, the corner sparse data and saturated adjacency sequence are determined, and the spontaneous radiation and environmental radiation components are combined to hash fusion, and the reflection hierarchical thermal amplitude is calculated. Quantity and jump value determine the frame distorted state.

Benefits of technology

It improves the accuracy of dimension detection when the surface of the automotive display frame is twisted, can intuitively express the degree of unevenness, enhances the accuracy of non-flat state recognition, and improves the detection accuracy by correcting the detection results by jump value.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117589105B_ABST
    Figure CN117589105B_ABST
Patent Text Reader

Abstract

The present application provides an infrared-based size detection method and device. By starting the infrared size detection of the frame of an automobile display screen, a frame global anisotropic image of the frame of the automobile display screen is obtained, a frame global granular image is determined, and then a frame global saturated adjacency sequence is obtained. A radiation threshold factor is determined according to a global spontaneous radiation component and a global environmental radiation component. The frame global saturated adjacency sequence and the radiation threshold factor are hashed and fused. The frame global mirror data and the frame global diffuse data of the frame global reflection gradient thermal amplitude are determined. The frame global mirror data and the frame global diffuse data are jump-connected to obtain a frame global jump value. When the frame global jump value is greater than a preset frame global stretch threshold, the frame of the automobile display screen is determined to be in a distorted state and uploaded to an infrared detection information library of the frame of the automobile display screen. This can improve the accuracy of size detection when the surface of the frame of the automobile display screen is distorted.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of infrared technology, and more specifically, to a size detection method and device based on infrared. Background Art

[0002] Infrared is a type of electromagnetic radiation that is below the visible spectrum, has a lower frequency and a longer wavelength. Infrared is located between visible light and microwaves in the spectrum. Unlike visible light, the human eye cannot directly perceive infrared radiation, but it can be detected by many thermal sensing devices and infrared sensors. Infrared radiation is generated by the heat of an object. Infrared imagers can capture infrared radiation emitted from the surface of an object and convert it into a thermal map to show the temperature distribution of the object. Infrared has a wide range of applications in science, engineering, and daily life, especially playing an important role in applications related to heat and temperature.

[0003] Infrared-based dimension detection is widely used in the manufacturing industry. It has important uses in the automotive manufacturing industry, such as quality control, cost control, and product consistency. Among them, it plays a key role in the dimension detection and surface distortion monitoring of the frame of the automotive display screen. The infrared sensor is used to measure the size or dimension change of the automotive display frame to ensure that the size of the automotive display frame meets the specification, which helps manufacturers produce high-quality automotive display frames to meet market demand. In the prior art, the detection device transmits infrared rays to the frame of the automotive display screen, and calculates the size according to the intensity and time delay of the reflected infrared radiation. However, when the frame surface of the automotive display screen is distorted, the accuracy of the dimension detection will be reduced due to the different degrees of distortion of the frame surface of the automotive display screen. Therefore, it is necessary to correct the dimension detection progress according to the distortion of the frame surface of the automotive display screen. However, how to correct the dimension detection result of the frame of the automotive display screen to improve the accuracy of the dimension detection when the surface of the frame of the automotive display screen is distorted is a problem faced by the industry. Summary of the invention

[0004] The present application provides an infrared-based dimension detection method and device to solve the technical problem of low dimension detection accuracy when the surface of a car display screen frame is distorted.

[0005] In order to solve the above technical problems, this application adopts the following technical solutions:

[0006] In a first aspect, the present application provides a size detection method based on infrared rays, comprising the following steps:

[0007] Start infrared size detection of the frame of the automobile display screen and obtain the full-frame anisotropic image of the frame of the automobile display screen;

[0008] Converting the frame global anisotropic image into a frame global granular image, determining frame global corner point sparse data from the frame global granular image, and further determining a frame global saturated adjacent sequence from the frame global corner point sparse data;

[0009] Acquire the global spontaneous radiation component and the global environmental radiation component of the global granular image of the framework, determine the radiation threshold factor according to the global spontaneous radiation component and the global environmental radiation component, and then hash and fuse the global saturated adjacent sequence of the framework and the radiation threshold factor to obtain a global hash fusion brightness value set of the framework;

[0010] Determine the frame global reflection gradient thermal amplitude according to the frame global hash fusion brightness value set, determine the frame global mirror data and the frame global diffuse data of the frame global reflection gradient thermal amplitude, perform transition connection on the frame global mirror data and the frame global diffuse data, and obtain the frame global transition value;

[0011] When the frame global jump value is greater than a preset frame global stretch threshold, the automobile display screen frame is determined to be in a distorted state and uploaded to the automobile display screen frame infrared detection information library.

[0012] In some embodiments, converting the frame global anisotropic image into a frame global granular image specifically includes:

[0013] Obtaining anisotropic pixel point sets of the global anisotropic image of the framework;

[0014] Determining anisotropic brightness gradient characteristic values ​​of the global anisotropic image of the framework;

[0015] Determining the granularity intensity value of each brightness gradient according to the anisotropic brightness gradient characteristic value;

[0016] A framework global granularity image is determined according to the anisotropic pixel point set and the granularity intensity value.

[0017] In some embodiments, determining the frame global corner point sparse data from the frame global granular image specifically includes:

[0018] Determining the granularity value of each pixel in the global granularity image of the framework;

[0019] Obtaining a global granularity factor of the global granularity image of the framework;

[0020] Determine a pixel point granularity correction value of the framework global granularity image according to the granularity value of each pixel point and the global granularity factor;

[0021] The framework global corner point sparse data is determined according to the pixel point granularity correction value of the framework global granularity image.

[0022] In some embodiments, the pixel point granularity correction value of the frame global granularity image may be determined by the following formula:

[0023]

[0024] Where A represents the pixel granularity correction value of the global granularity image of the framework, L represents the sum of the granularity values ​​of the pixels, δ represents the global granularity factor of the global granularity image of the framework, Indicates the uniformity of the granularity value of the pixel points in the global granularity image of the frame, q i Represents the granularity value of the i-th pixel in the global granularity image of the framework, i = 1, 2, 3, ..., n.

[0025] In some embodiments, performing hash fusion on the framework global saturation adjacent sequence and the radiation threshold factor to obtain a framework global hash fusion brightness value set specifically includes:

[0026] Determining the adjacency hash feature of the global saturated adjacency sequence of the framework;

[0027] Obtaining the distortion fusion factor of the global saturated adjacent sequence of the framework;

[0028] A framework global hash fusion brightness value set is determined according to the adjacent hash feature quantity of the framework global saturation adjacent sequence and the distortion fusion factor of the framework global saturation adjacent sequence.

[0029] In some embodiments, the frame global reflectance gradient thermal amplitude includes a frame global reflectance first gradient fusion brightness set and a frame global reflectance second gradient fusion brightness set.

[0030] In some embodiments, when the frame global jump value is less than or equal to a preset frame global stretch threshold, the automobile display frame is determined to be in a non-distorted state and uploaded to the automobile display frame infrared detection information library.

[0031] In a second aspect, the present application provides an infrared-based size detection device, which includes an infrared detection unit, and the infrared detection unit includes:

[0032] A frame global anisotropic image determination module is used to start infrared size detection of the frame of the automobile display screen and obtain a frame global anisotropic image of the frame of the automobile display screen;

[0033] A framework global saturated adjacent sequence determination module is used to convert the framework global anisotropic image into a framework global granular image, determine framework global corner point sparse data from the framework global granular image, and further determine the framework global saturated adjacent sequence through the framework global corner point sparse data;

[0034] A framework global hash fusion brightness value set determination module is used to obtain the global spontaneous radiation component and the global environmental radiation component of the framework global granular image, determine the radiation threshold factor according to the global spontaneous radiation component and the global environmental radiation component, and then hash and fuse the framework global saturated adjacent sequence and the radiation threshold factor to obtain the framework global hash fusion brightness value set;

[0035] a frame global transition value determination module, used to determine the frame global reflection gradient thermal amplitude according to the frame global hash fusion brightness value set, determine the frame global mirror data and the frame global diffuse data of the frame global reflection gradient thermal amplitude, perform transition connection on the frame global mirror data and the frame global diffuse data, and obtain the frame global transition value;

[0036] The automobile display screen frame state determination module is used to determine the automobile display screen frame as being in a distorted state when the frame global jump value is greater than a preset frame global stretch threshold, and upload it to the automobile display screen frame infrared detection information library.

[0037] In a third aspect, the present application provides a computer device, comprising a memory and a processor, wherein the memory stores codes, and the processor is configured to obtain the codes and execute the above-mentioned infrared-based size detection method.

[0038] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned infrared-based size detection method is implemented.

[0039] The technical solution provided by the embodiments disclosed in this application has the following beneficial effects:

[0040] In the infrared-based size detection method and device provided in the present application, firstly, infrared size detection of the automobile display screen frame is started, a frame global anisotropic image of the automobile display screen frame is obtained, the frame global anisotropic image is converted into a frame global granular image, the frame global corner point sparse data is determined from the frame global granular image, and then the frame global saturated adjacency sequence is determined through the frame global corner point sparse data, the global spontaneous radiation component and the global environmental radiation component of the frame global granular image are obtained, the radiation threshold factor is determined according to the global spontaneous radiation component and the global environmental radiation component, and then the The global saturation adjacency sequence of the frame and the radiation threshold factor are hashed and fused to obtain a global hash fusion brightness value set of the frame, the global reflection gradient thermal amplitude of the frame is determined according to the global hash fusion brightness value set of the frame, the global mirror data and the global diffuse data of the global reflection gradient thermal amplitude of the frame are determined, the global mirror data and the global diffuse data of the frame are jump-connected to obtain a global jump value of the frame, and when the global jump value of the frame is greater than a preset global stretch threshold of the frame, the frame of the automobile display screen is determined to be in a distorted state and uploaded to the infrared detection information library of the frame of the automobile display screen.

[0041] The scheme replaces the values ​​in the frame global corner point sparse data with the frame global saturation threshold to obtain a frame global saturation adjacency sequence, which can more intuitively show the unevenness of the frame surface of the automobile display screen. By obtaining the global spontaneous radiation component and the global environmental radiation component of the frame global granular image, the recognition accuracy when the frame surface is in an uneven state in the frame size detection of the automobile display screen can be effectively analyzed, which is conducive to improving the accuracy of size detection. According to the difference degree of fusion brightness between the frame global reflection first-order fusion brightness set and the frame global reflection second-order fusion brightness set and the set seven-layer diffusion degree data, the frame global jump value is obtained, which means that in the frame global hash fusion brightness value set, when the diffusion degree data of different levels undergo a step change, the degree of distortion of the frame surface of the automobile display screen can be better calculated, thereby improving the accuracy of size detection when the frame surface of the automobile display screen is distorted. By comparing the frame global jump value greater than the preset frame global stretch threshold, the result is corrected, thereby improving the accuracy of size detection when the frame surface of the automobile display screen is distorted. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 is an exemplary flow chart of an infrared-based size detection method according to some embodiments of the present application;

[0043] Figure 2 This is a schematic diagram of an exemplary process for determining a radiation threshold factor in some embodiments of the present application;

[0044] Figure 3 is a schematic diagram of exemplary hardware and / or software of an infrared detection unit according to some embodiments of the present application;

[0045] Figure 4 It is a structural schematic diagram of a computer device for implementing an infrared-based size detection method according to some embodiments of the present application. DETAILED DESCRIPTION

[0046] The core of the present application is to start the infrared size detection of the automobile display frame, obtain the frame global anisotropic image of the automobile display frame, and then determine the frame global granular image, and then obtain the frame global saturated adjacency sequence, determine the radiation threshold factor according to the global spontaneous radiation component and the global environmental radiation component, hash and fuse the frame global saturated adjacency sequence and the radiation threshold factor, determine the frame global mirror data and the frame global diffuse data of the frame global reflection gradient thermal amplitude, perform jump connection on the frame global mirror data and the frame global diffuse data, and obtain the frame global jump value. When the frame global jump value is greater than the preset frame global stretch threshold, the automobile display frame is judged to be in a distorted state and uploaded to the automobile display frame infrared detection information library, which can improve the accuracy of size detection when the surface of the automobile display frame is distorted.

[0047] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods. Figure 1 , which is an exemplary flow chart of an infrared-based size detection method according to some embodiments of the present application. The infrared-based size detection method 100 mainly includes the following steps:

[0048] In step 101, infrared size detection of the frame of the automobile display screen is started to obtain a full-domain anisotropic image of the frame of the automobile display screen.

[0049] In specific implementation, the infrared size detection of the automobile display frame is started, and the frame global anisotropic image of the automobile display frame is obtained from the automobile display frame infrared size detection database. The frame global anisotropic image can be obtained by shooting the surface of the automobile display frame to be inspected by an infrared imager according to the detection environment.

[0050] It should be noted that in the present application, the frame global anisotropic image of the automobile display frame is composed of the isotropic image of the flat area on the surface of the automobile display frame and the anisotropic image of the twisted area on the surface of the automobile display frame. The infrared radiation direction of the flat area on the surface of the automobile display frame is in the same direction, that is, the obtained infrared image is an isotropic image, and the infrared radiation directions of the anisotropic images of the twisted area on the surface of the automobile display frame are in different directions, that is, the obtained infrared image is anisotropic image, that is, the infrared radiation image of the surface of the automobile display frame is the frame global anisotropic image, and the frame global anisotropic image represents an image of the sum of the infrared radiation properties of the surface of the automobile display frame in different directions, and includes temperature distribution data, pixel brightness values ​​and brightness level data on the flat area and twisted area on the surface of the automobile display frame.

[0051] In step 102, the frame global anisotropic image is converted into a frame global granular image, frame global corner point sparse data is determined from the frame global granular image, and then the frame global saturated adjacent sequence is determined through the frame global corner point sparse data.

[0052] In some embodiments, converting the frame global anisotropic image into a frame global granular image may be achieved by using the following steps:

[0053] Obtaining anisotropic pixel point sets of the global anisotropic image of the framework;

[0054] Determining anisotropic brightness gradient characteristic values ​​of the global anisotropic image of the framework;

[0055] Determining the granularity intensity value of each brightness gradient according to the anisotropic brightness gradient characteristic value;

[0056] A framework global granularity image is determined according to the anisotropic pixel point set and the granularity intensity value.

[0057] In a specific implementation, the pixel values ​​of all pixels of the framework global anisotropic image are read to obtain anisotropic pixel set, a brightness gradient is set according to the brightness level of the infrared image, the framework global anisotropic image is divided to obtain brightness values ​​of different gradients, i.e., anisotropic brightness gradient feature values. For example, the brightness level range of the infrared image is [0, 255), and a brightness level is divided every 5 unit lengths. The brightness level range of the infrared image is divided into multiple brightness gradients, i.e., the brightness level [0, 5) is the first brightness gradient, and the brightness level [5, 10) is the second brightness gradient, until all the divisions of the brightness level range are completed to obtain all brightness gradients, and the brightness values ​​of the pixels in the framework global anisotropic image are classified according to the brightness gradient. If the brightness of the pixel in the framework global anisotropic image is 3, then the pixel belongs to the first gradient in the brightness level [0, 5), and if the brightness of the pixel in the framework global anisotropic image is 8, then the pixel belongs to the second gradient in the brightness level [5, 10), until all the pixels in the framework global anisotropic image are classified. The median brightness of the pixels in each brightness gradient is used as the corresponding anisotropic brightness gradient eigenvalue; the granularity intensity value of each brightness gradient is determined according to the anisotropic brightness gradient eigenvalue. In some embodiments, the average brightness value of the pixels in each gradient is calculated to obtain a gradient uniform brightness value, and the difference between each anisotropic brightness gradient eigenvalue and the corresponding gradient uniform brightness value is calculated and then an arithmetic square root operation is performed to obtain a granularity intensity value of each brightness gradient; and the framework global granularity image is determined according to the anisotropic pixel point set and the granularity intensity value. In some embodiments, the pixel points whose pixel brightness value of each brightness gradient in the anisotropic pixel point set of the framework global anisotropic image is less than the granularity intensity value of the corresponding brightness gradient are removed, and the granularity intensity value of the corresponding brightness gradient is used as the pixel brightness value in the brightness gradient, and the pixel points whose pixel brightness value of each brightness gradient in the anisotropic pixel point set of the framework global anisotropic image is equal to or greater than the granularity intensity value of the corresponding brightness gradient are retained, and the comparison of the pixel brightness values ​​in all brightness gradients is completed to obtain the framework global granularity image.

[0058] It should be noted that in this step, the anisotropic pixel set represents the pixel brightness value of the pixel points in the distorted area of ​​the frame surface of the automobile display screen in the global anisotropic image, the anisotropic brightness gradient eigenvalue is the median brightness of the pixel points in each gradient after the pixel points in the frame global anisotropic image are divided into gradients, the granularity intensity value of each brightness gradient represents the degree of brightness discreteness of the pixel points in each brightness gradient, and the frame global granular image is an image obtained by comparing and replacing the brightness values ​​of the pixel points in each brightness gradient in the anisotropic pixel set of the frame global anisotropic image, and the infrared radiation pixel points in the distorted area of ​​the image have lower brightness values ​​and are composed of relatively scattered brightness particles, and the brightness particles refer to the existence of sporadic brighter points on the pixel points compared with other brightness.

[0059] In some embodiments, determining the frame global corner point sparse data from the frame global granularity image may be implemented by the following steps:

[0060] Determining the granularity value of each pixel in the global granularity image of the framework;

[0061] Obtaining a global granularity factor of the global granularity image of the framework;

[0062] Determine a pixel point granularity correction value of the framework global granularity image according to the granularity value of each pixel point and the global granularity factor;

[0063] The framework global corner point sparse data is determined according to the pixel point granularity correction value of the framework global granularity image.

[0064] In a specific implementation, the granularity value of each pixel in the framework global granularity image is determined, and the framework global granularity image is placed in a two-dimensional coordinate diagram, the horizontal axis of the coordinate diagram is the increasing brightness data, and the vertical axis is the increasing brightness level. The ratio of the brightness data corresponding to each pixel in the framework global granularity image to the brightness level is the granularity value of each pixel in the framework global granularity image, and the granularity value represents the contrast of brightness particles; the global granularity factor of the framework global granularity image is the ratio of the pixel brightness value measured by the framework global granularity image at a certain ambient temperature to the brightness of the framework itself. The difference between the brightness values ​​of the pixels is usually a number that changes with the ambient temperature and ranges from 0 to 1; the standard deviation of the granularity value of each pixel is multiplied by the global granularity factor of the frame global granularity image to obtain the pixel granularity correction value of the frame global granularity image; the granularity value of each pixel is added to the pixel granularity correction value of the frame global granularity image to obtain the corner brightness of the pixel, and the corner brightness value of each pixel is used as the frame global corner sparse data; wherein the pixel granularity correction value of the frame global granularity image can be determined by the following formula:

[0065]

[0066] Where A represents the pixel granularity correction value of the global granularity image of the framework, L represents the sum of the granularity values ​​of the pixels, δ represents the global granularity factor of the global granularity image of the framework, Indicates the uniformity of the granularity value of the pixel points in the global granularity image of the frame, q i Represents the granularity value of the i-th pixel in the global granularity image of the framework, i = 1, 2, 3, ..., n.

[0067] It should be noted that the sum of the granularity values ​​of the pixels in this step is the sum of the granularity values ​​of all pixels in the framework global granularity image, and the uniformity of the granularity values ​​of the pixels in the framework global granularity image is the average granularity values ​​of the pixels in the framework global granularity image.

[0068] In a specific implementation, the global saturated adjacent sequence of the framework is determined by the global corner point sparse data of the framework. In some embodiments, the brightness value range of the 8-bit grayscale image is equally divided into 5 global brightness levels, such as the brightness range [0, 51) as the first global brightness level, the brightness range [51, 102) as the second global brightness level, the brightness range [102, 153) as the third global brightness level, the brightness range [153, 204) as the fourth global brightness level, and the brightness range [204, 25 5) As the fifth level of global brightness, all values ​​in the framework global corner point sparse data are classified according to the global brightness level. For example, if the value in the framework global corner point sparse data is 52, it is classified as the second level of global brightness. The median of all brightness values ​​belonging to the third level of global brightness is used as the framework global saturation threshold. All values ​​in the framework global corner point sparse data are traversed, and the values ​​in the framework global corner point sparse data that are higher than the framework global saturation threshold are replaced with the framework global saturation threshold to obtain the framework global saturation adjacent sequence.

[0069] It should be noted that the infrared radiation of the flat part of the car display frame is uniform, and the brightness display of the corresponding position on the frame global anisotropic image is relatively uniform and bright, that is, the brightness value of the pixel point is high, and the infrared radiation of the non-flat part of the car display frame is scattered, so the brightness display of the non-flat area on the frame global anisotropic image is darker than that of the flat part. The actual brightness value is determined according to the degree of distortion causing the non-flatness. In this application, the frame global corner point sparse data can more accurately present the contrast between the flat part and the non-flat part of the surface of the car display frame. The frame global saturation adjacent sequence is composed of the brightness values ​​of the replaced pixels compared with the frame global saturation threshold. In the above steps, the frame global saturation threshold is used to divide the boundary between the flat part and the non-flat part in the frame global corner point sparse data of the car display frame. The frame global saturation adjacent sequence obtained by replacing the values ​​in the frame global corner point sparse data with the frame global saturation threshold can more intuitively show the unevenness of the surface of the car display frame.

[0070] In step 103, the global spontaneous radiation component and the global ambient radiation component of the global granular image of the framework are obtained, a radiation threshold factor is determined according to the global spontaneous radiation component and the global ambient radiation component, and then the global saturated adjacent sequence of the framework and the radiation threshold factor are hashed and fused to obtain a global hashed fusion brightness value set of the framework.

[0071] It should be noted that the global spontaneous radiation component of the frame global granular image described in this step refers to the thermal radiation spontaneously scattered outward by the surface of the car display frame itself, and the global spontaneous radiation component is composed of the average brightness values ​​of the heavy pixel points of the frame global granular image. The global environmental radiation component refers to the thermal radiation of the environment surrounding the surface of the car display frame, including thermal radiation from the light source, background environment, etc. The average brightness value of the pixel points in the area of ​​the frame global granular image that does not contain the car display frame is selected as the global environmental radiation component.

[0072] In some embodiments, reference Figure 2 As shown, this figure is a schematic diagram of the process of determining the radiation threshold factor in some embodiments of the present application. In this embodiment, the radiation threshold factor can be determined by the following steps:

[0073] In step 1031, a radiation characteristic value of the global granularity image of the framework is determined according to the global spontaneous radiation component;

[0074] In step 1032, a convergence characteristic value of the framework global granularity image is determined according to the global environmental radiation component;

[0075] In step 1033, the radiation aggregation degree of the global spontaneous radiation component and the global ambient radiation component is determined;

[0076] In step 1034, a radiation threshold factor is determined according to the radiation characteristic value of the framework global granularity image, the convergence characteristic value of the framework global granularity image, and the radiation aggregation degree of the global environment radiation component.

[0077] In specific implementation, the radiation threshold factor can be determined by the following formula:

[0078]

[0079] Among them, ψ v represents the vth radiation threshold factor, f(ω) represents the radiation characteristic value of the global granular image of the frame, represents the convergence eigenvalue of the global granularity image of the frame, y s (σ) represents the compact parameter of the brightness value of the Sth pixel in the global granular image of the framework, δ represents the global granularity factor of the global granular image of the framework, ω represents the global spontaneous radiation component, σ represents the brightness value of the pixel in the global granular image of the framework, represents the global environmental radiation component, The vth difference between the global ambient radiation component and the compact parameter of the pixel brightness value in the global granular image of the frame, s = 1, 2, ..., S.

[0080] In specific implementation, the radiation characteristic value of the framework global granular image is determined according to the global spontaneous radiation component. In some embodiments, the absolute value of the difference between the global spontaneous radiation component and the mean brightness of the pixels of the framework global granular image is used as the radiation characteristic quantity, and the convergence characteristic value of the framework global granular image is determined according to the global environmental radiation component. In some embodiments, the absolute value of the difference between the global environmental radiation component and the mean brightness of the pixels of the framework global granular image is used as the convergence characteristic quantity, and the radiation aggregation degree of the global spontaneous radiation component and the global environmental radiation component is determined. In some embodiments, the radiation aggregation degree of the global spontaneous radiation component and the global environmental radiation component is determined, and the radiation aggregation degree is obtained by subtracting the vth difference between the compact parameters of the global spontaneous radiation component, the global environmental radiation component and the brightness value of the pixels in the framework global granular image.

[0081] It should be noted that the radiation characteristic value of the framework global granularity image in this step represents the distance between the global spontaneous radiation component and the average brightness of the pixels of the framework global granularity image. The larger the absolute value, the farther the distance, that is, the larger the difference between the spontaneous radiation component and the average brightness of the pixels of the framework global granularity image. The convergence characteristic value of the framework global granularity image represents the distance between the global environmental radiation component and the average brightness of the pixels of the framework global granularity image. The larger the absolute value, the smaller the influence of the global environmental radiation component on the brightness of the pixels in the framework global granularity image. The radiation aggregation degree of the global environmental radiation component represents the degree of correlation between the compact parameter of the brightness value of the pixels in the framework global granularity image and the global spontaneous radiation component. The smaller the radiation aggregation degree of the global environmental radiation component, the higher the correlation. The compact parameter of the brightness value of the pixels in the framework global granularity image is the absolute value of the difference between the brightness value of the pixels in the global granularity image and its average value, which represents the degree of change of the brightness value of the pixels in the global granularity image.

[0082] In some embodiments, the framework global saturation adjacent sequence and the radiation threshold factor are hashed and fused to obtain a framework global hash fusion brightness value set, which can be implemented by the following steps:

[0083] Determining the adjacency hash feature of the global saturated adjacency sequence of the framework;

[0084] Obtaining the distortion fusion factor of the global saturated adjacent sequence of the framework;

[0085] A framework global hash fusion brightness value set is determined according to the adjacent hash feature quantity of the framework global saturation adjacent sequence and the distortion fusion factor of the framework global saturation adjacent sequence.

[0086] In specific implementation, the adjacent hash feature of the framework global saturated adjacent sequence is determined. In some embodiments, the standard deviation of the values ​​in the framework global saturated adjacent sequence is used as the adjacent hash feature of the framework global saturated adjacent sequence. After multiplying the distortion fusion factor of the framework global saturated adjacent sequence by the adjacent hash feature of the framework global saturated adjacent sequence, the obtained results are added to each value in the framework global saturated adjacent sequence to obtain new values, and then the new values ​​form a framework global hash fusion brightness value set.

[0087] It should be noted that the adjacency hash feature of the global saturated adjacency sequence of the framework in the present application represents the amount of contact between adjacent values ​​in the global saturated adjacency sequence of the framework, and the contact amount refers to the degree of discreteness of the values ​​in the global saturated adjacency sequence of the framework, and the degree of discreteness can be represented by the standard deviation of the data. The lower the degree of discreteness of the values ​​in the global saturated adjacency sequence of the framework, the fewer the distorted parts of the surface of the frame of the automobile display screen, and the more evenly distributed the flat parts. The distortion fusion factor of the global saturated adjacency sequence of the framework represents the degree of adjacency of the values ​​in the global saturated adjacency sequence of the framework after being affected by the ambient thermal radiation, which is higher than the actual frame surface brightness value, that is, distortion occurs, and the adjacency of the frame surface brightness value represents the absolute value of the difference in the radiation amount of the flat part and the radiation amount of the non-flat part at the connection of the flat part and the non-flat part of the frame surface of the automobile display screen after radiating heat. The distortion fusion factor is used for correction to remove the factor that reduces the accuracy of flatness recognition due to the sudden decrease in thermal radiation of the non-flat part. The value range of the distortion fusion factor is between 0 and 1.

[0088] It should be noted that by acquiring the global spontaneous radiation component and the global environmental radiation component of the frame global granular image, the recognition accuracy when the frame surface is uneven during the size detection of the automobile display frame can be effectively analyzed, which is beneficial to improving the accuracy of size detection.

[0089] In step 104, the frame global reflection gradient thermal amplitude is determined according to the frame global hash fusion brightness value set, the frame global mirror data and the frame global diffuse data of the frame global reflection gradient thermal amplitude are determined, and the frame global mirror data and the frame global diffuse data are transitionally connected to obtain a frame global transition value.

[0090] In specific implementation, the framework global reflection gradient thermal amplitude is determined according to the framework global hash fusion brightness value set. In some embodiments, based on a large amount of experimental data, the values ​​in the framework global hash fusion brightness value set are used as the grayscale values ​​of the pixels. For example, the pixel brightness grayscale value 98 is used as the framework global reflection gradient threshold, the values ​​in the framework global hash fusion brightness value set that are less than the framework global reflection gradient threshold are used as the framework global reflection first gradient fusion brightness set, the values ​​in the framework global hash fusion brightness value set that are greater than or equal to the framework global reflection gradient threshold are used as the framework global reflection second gradient fusion brightness set, and the framework global reflection first gradient fusion brightness set and the framework global reflection second gradient fusion brightness set constitute the framework global reflection gradient thermal amplitude.

[0091] It should be noted that in the present application, the frame global reflection gradient threshold is a dividing point of the grayscale value of a pixel. A value lower than the frame global reflection gradient threshold indicates that the surface of the car display frame corresponding to the frame global hash fusion brightness value set where the value is located is uneven, otherwise it is flat.

[0092] In specific implementation, the frame global mirror data and frame global diffuse data of the frame global reflection gradient thermal amplitude are determined. In some embodiments, the average value of all fused brightness values ​​in the first gradient fused brightness set of the frame global reflection is used as the frame global diffuse data, and the average value of all fused brightness values ​​in the second gradient fused brightness set of the frame global reflection is used as the frame global mirror data. The values ​​in the first gradient fused brightness set represent the first fused brightness values, and the values ​​in the second gradient fused brightness set represent the second fused brightness values. The fused brightness values ​​represent the brightness of the pixels in the thermal radiation map.

[0093] It should be noted that in the present application, the first-tier fused brightness set of the frame global reflection is used to reflect that when the surface of the frame of the automobile display screen is uneven or distorted, the thermal radiation on the surface of the frame shows diffuse reflection. The larger the frame global diffuse data, the greater the degree of distortion of the frame surface area corresponding to the frame global reflection tiered thermal amplitude of the frame surface of the automobile display screen. Obviously, the error in the size detection of the surface of the frame of the automobile display screen is also greater. The second-tier fused brightness set of the frame global reflection is used to reflect that when the surface of the frame of the automobile display screen is slightly uneven or flat, the surface thermal radiation phenomenon is close to or equal to mirror reflection. The larger the frame global mirror data, the flatter the frame surface area corresponding to the frame global reflection tiered thermal amplitude of the frame surface of the automobile display screen.

[0094] In some embodiments, performing a transition connection on the frame global mirror data and the frame global diffuse data to obtain a frame global transition value may be achieved by using the following steps:

[0095] Get the vth radiation threshold factor ψ v ;

[0096] Get the global mirroring data of the frame

[0097] Get the global diffuse data of the framework

[0098] Determine the amplitude variation of the frame's global reflection gradient heat amplitude according to all the heat radiation D in the frame's global reflection gradient heat amplitude

[0099] Determine the brightness value k of the qth second fusion brightness in the second-order fusion brightness set of the frame global reflection q ;

[0100] Determine the mirror smoothing data of the second-order fusion brightness set of the frame global reflection

[0101] According to the vth radiation threshold factor ψ v , the framework global mirroring data The framework global diffuse data All the heat radiation D in the frame's global reflected gradient heat amplitude determines the amplitude variation of the frame's global reflected gradient heat amplitude The brightness value k of the qth second fusion brightness in the second fusion brightness set of the frame global reflection q Mirror smoothing data of the second-order fusion brightness set of the frame global reflection Determine the global transition value of the framework, wherein the global transition value of the framework can be determined by the following formula:

[0102]

[0103] Among them, Ω represents the global transition value of the framework, is the global diffuse data of the gth frame in the dth layer diffuse degree, g = 1, 2, 3, ..., G, is the average value of global diffuse data in the diffuse degree of the dth layer, f j is the brightness value of the jth first fusion brightness in the first fusion brightness set of the frame global reflection, Indicates the brightness smoothness of the second fusion brightness in the second-order fusion brightness set of the frame global reflection, q = 1, 2, 3, ..., Q, Represents the first fused brightness deviation in the first-order fused brightness set of the frame global reflection, j=1, 2, 3,…, J.

[0104] It should be noted that in this application Indicates that in the global diffuse degree layer of the framework, starting from the lowest brightness value of the fusion brightness corresponding to the global diffuse degree layer of the framework, the brightness value range [0, 20) is used as the first layer of diffuse degree data, and the brightness value range [20, 40) is used as the second layer of diffuse degree data. There are seven layers of diffuse degree data in total, among which the brightness range in the seventh layer of diffuse degree data is [120, 127.5) and the gth global diffuse data in the dth layer of diffuse degree g=1,2,3,…,G,means It represents the global diffuse data average of the diffuse degree data of the dth layer in all layers of diffuse degree data. is the second fusion brightness value k of the second fusion brightness set of the frame global reflection q , q=1,2,3,…,Q and the second fusion brightness value The sum of the differences between the average values ​​of represents the brightness smoothness of the pixels in the second-order fusion brightness concentration of the frame global reflection. It is the first fusion brightness of the frame full reflection, the first fusion brightness mean The first fusion brightness value f of the first fusion brightness concentration of the frame global reflection j The sum of the squares of the differences, j = 1, 2, 3, ..., J, represents the deviation of the first fusion brightness in the first-order fusion brightness set of the frame global reflection.

[0105] It should be noted that, based on the difference in fusion brightness between the first-tier fusion brightness set of the frame global reflection and the second-tier fusion brightness set of the frame global reflection and the set seven-layer diffusion degree data, the frame global jump value is obtained, which means that in the frame global hash fusion brightness value set, when the diffusion degree data of different levels undergo a step change, the degree of distortion of the frame surface of the automobile display can be better calculated, thereby improving the accuracy of size detection when the frame surface of the automobile display is distorted.

[0106] In step 105, when the frame global jump value is greater than the preset frame global stretch threshold, the automobile display frame is determined to be in a distorted state and uploaded to the automobile display frame infrared detection information library.

[0107] In specific implementation, when the frame global jump value is greater than a preset frame global stretch threshold, the automobile display frame is determined to be in a distorted state, and at the same time, the infrared size detection result of the automobile display frame surface is corrected according to the frame global jump value and uploaded to the automobile display frame infrared detection information library; when the frame global jump value is less than or equal to the preset frame global stretch threshold, the automobile display frame is determined to be in a non-distorted state, and the infrared size measurement result of the automobile display frame surface is directly output and uploaded to the automobile display frame infrared detection information library.

[0108] It should be noted that the global stretch threshold of the frame in this step refers to the maximum tolerance of the surface distortion of the automobile display frame, indicating the degree of influence of the distortion of the surface of the automobile display frame on the size detection. The larger the global stretch threshold of the frame, the greater the influence on the size detection result.

[0109] In addition, in another aspect of the present application, in some embodiments, the present application provides an infrared-based size detection device, the device comprising an infrared detection unit, referring to Figure 3 , which is a schematic diagram of exemplary hardware and / or software of an infrared detection unit according to some embodiments of the present application, the infrared detection unit 300 includes: a framework global anisotropic image determination module 301, a framework global saturated adjacent sequence determination module 302, a framework global hash fusion brightness value set determination module 303, a framework global transition value determination module 304 and a vehicle display screen frame state determination module 305, which are respectively described as follows:

[0110] The frame global anisotropic image determination module 301 in the present application is mainly used to start infrared size detection of the frame of the automobile display screen and obtain the frame global anisotropic image of the frame of the automobile display screen;

[0111] A framework global saturated adjacent sequence determination module 302, in the present application, the framework global saturated adjacent sequence determination module 302 is mainly used to convert the framework global anisotropic image into a framework global granular image, determine the framework global corner point sparse data from the framework global granular image, and then determine the framework global saturated adjacent sequence through the framework global corner point sparse data;

[0112] The framework global hash fusion brightness value set determination module 303 in the present application is mainly used to obtain the global spontaneous radiation component and the global environmental radiation component of the framework global granular image, determine the radiation threshold factor according to the global spontaneous radiation component and the global environmental radiation component, and then hash and fuse the framework global saturated adjacent sequence and the radiation threshold factor to obtain the framework global hash fusion brightness value set;

[0113] The framework global transition value determination module 304 in the present application is mainly used to determine the framework global reflection gradient thermal amplitude according to the framework global hash fusion brightness value set, determine the framework global mirror data and the framework global diffuse data of the framework global reflection gradient thermal amplitude, perform transition connection on the framework global mirror data and the framework global diffuse data, and obtain the framework global transition value;

[0114] The automobile display screen frame state determination module 305 in the present application is mainly used to determine the automobile display screen frame as being in a distorted state when the frame global jump value is greater than a preset frame global stretch threshold, and upload it to the automobile display screen frame infrared detection information library.

[0115] In addition, the present application also provides a computer device, which includes a memory and a processor, wherein the memory stores codes, and the processor is configured to obtain the codes and execute the above-mentioned infrared-based size detection method.

[0116] In some embodiments, reference Figure 4 , which is a schematic diagram of the structure of a computer device for implementing an infrared-based size detection method according to some embodiments of the present application. The infrared-based size detection method in the above embodiment can be Figure 4 The computer device shown in the figure is implemented, and the computer device includes at least one processor 401, a communication bus 402, a memory 403 and at least one communication interface 404.

[0117] The processor 401 may be a general-purpose central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more processors for controlling the execution of the infrared-based size detection method in the present application.

[0118] Communication bus 402 may include a pathway for transmitting information between the above-mentioned components.

[0119] The memory 403 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, an optical disc storage (including a compressed optical disc, a laser disc, an optical disc, a digital versatile disc, a Blu-ray disc, etc.), a disk or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of an instruction or data structure and can be accessed by a computer, but is not limited thereto. The memory 403 may exist independently and be connected to the processor 401 via the communication bus 402. The memory 403 may also be integrated with the processor 401.

[0120] The memory 403 is used to store the program code for executing the solution of the present application, and the execution is controlled by the processor 401. The processor 401 is used to execute the program code stored in the memory 403. The program code may include one or more software modules. The infrared-based size detection method in the above embodiment can be implemented by the processor 401 and one or more software modules in the program code in the memory 403.

[0121] The communication interface 404 uses any transceiver or other device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0122] In a specific implementation, as an embodiment, a computer device may include multiple processors, each of which may be a single-CPU processor or a multi-CPU processor. The processor here may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0123] The above-mentioned computer device can be a general-purpose computer device or a special-purpose computer device. In a specific implementation, the computer device can be a desktop computer, a portable computer, a network server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device or an embedded device. The embodiment of the present application does not limit the type of computer device.

[0124] In addition, the present application also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned infrared-based size detection method is implemented.

[0125] In summary, in the infrared-based size detection method and device disclosed in the embodiment of the present application, firstly, the scheme replaces the numerical value in the frame global corner point sparse data with the frame global saturation threshold to obtain the frame global saturation adjacent sequence, which can more intuitively show the unevenness of the surface of the automobile display frame. By obtaining the global spontaneous radiation component and the global environmental radiation component of the frame global granular image, the recognition accuracy when the frame surface appears uneven in the automobile display frame size detection can be effectively analyzed, which is beneficial to improving the accuracy of size detection. According to the first-tier fusion brightness set of the frame global reflection and the frame global The difference in fusion brightness between the second-tier fusion brightness sets of domain reflection is combined with the set seven-layer diffusion degree data, and then the frame global jump value is obtained, which means that in the frame global hash fusion brightness value set, when the diffusion degree data of different levels undergo a step change, the degree of distortion of the frame surface of the automobile display can be better calculated, thereby improving the accuracy of size detection when the frame surface of the automobile display is distorted. The frame global jump value is greater than the preset frame global stretch threshold for comparison, and the result is corrected, thereby improving the accuracy of size detection when the frame surface of the automobile display is distorted.

[0126] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0127] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A size detection method based on infrared, It is characterized in that The steps include: Start infrared size detection of the frame of the automobile display screen and obtain the full-frame anisotropic image of the frame of the automobile display screen; Converting the frame global anisotropic image into a frame global granular image, determining frame global corner point sparse data from the frame global granular image, and further determining a frame global saturated adjacent sequence from the frame global corner point sparse data; Acquire the global spontaneous radiation component and the global environmental radiation component of the global granular image of the framework, determine the radiation threshold factor according to the global spontaneous radiation component and the global environmental radiation component, and then hash and fuse the global saturated adjacent sequence of the framework and the radiation threshold factor to obtain a global hash fusion brightness value set of the framework; Determine the frame global reflection gradient thermal amplitude according to the frame global hash fusion brightness value set, determine the frame global mirror data and the frame global diffuse data of the frame global reflection gradient thermal amplitude, perform transition connection on the frame global mirror data and the frame global diffuse data, and obtain the frame global transition value; When the frame global jump value is greater than a preset frame global stretch threshold, the automobile display screen frame is determined to be in a distorted state and uploaded to the automobile display screen frame infrared detection information library.

2. The method according to claim 1, It is characterized in that Converting the frame global anisotropic image into a frame global granular image specifically includes: Obtaining anisotropic pixel point sets of the global anisotropic image of the framework; Determining anisotropic brightness gradient characteristic values ​​of the global anisotropic image of the framework; Determining the granularity intensity value of each brightness gradient according to the anisotropic brightness gradient characteristic value; A framework global granularity image is determined according to the anisotropic pixel point set and the granularity intensity value.

3. The method according to claim 1, It is characterized in that Determining the sparse data of the global corner points of the framework from the global granular image of the framework specifically includes: Determining the granularity value of each pixel in the global granularity image of the framework; Obtaining a global granularity factor of the global granularity image of the framework; Determine a pixel point granularity correction value of the framework global granularity image according to the granularity value of each pixel point and the global granularity factor; The framework global corner point sparse data is determined according to the pixel point granularity correction value of the framework global granularity image.

4. The method according to claim 3, It is characterized in that The pixel point granularity correction value of the frame global granularity image is determined by the following formula: Where A represents the pixel granularity correction value of the global granularity image of the framework, L represents the sum of the granularity values ​​of the pixels, δ represents the global granularity factor of the global granularity image of the framework, Indicates the uniformity of the granularity value of the pixel points in the global granularity image of the frame, q i Represents the granularity value of the i-th pixel in the global granularity image of the framework, i = 1, 2, 3, ..., n.

5. The method according to claim 1, It is characterized in that The framework global saturation adjacent sequence and the radiation threshold factor are hashed and fused to obtain a framework global hash fusion brightness value set, which specifically includes: Determining the adjacency hash feature of the global saturated adjacency sequence of the framework; Obtaining the distortion fusion factor of the global saturated adjacent sequence of the framework; A framework global hash fusion brightness value set is determined according to the adjacent hash feature quantity of the framework global saturation adjacent sequence and the distortion fusion factor of the framework global saturation adjacent sequence.

6. The method according to claim 1, It is characterized in that The frame global reflection tiered thermal amplitude includes the frame global reflection first tier fusion brightness set and the frame global reflection second tier fusion brightness set.

7. The method according to claim 1, It is characterized in that The method also includes, when the frame global jump value is less than or equal to a preset frame global stretch threshold, determining the automobile display screen frame as a non-distorted state and uploading it to an automobile display screen frame infrared detection information library.

8. A size detection device based on infrared, It is characterized in that An infrared detection unit is included, and the infrared detection unit includes: A frame global anisotropic image determination module is used to start infrared size detection of the frame of the automobile display screen and obtain a frame global anisotropic image of the frame of the automobile display screen; A framework global saturated adjacent sequence determination module is used to convert the framework global anisotropic image into a framework global granular image, determine framework global corner point sparse data from the framework global granular image, and further determine the framework global saturated adjacent sequence through the framework global corner point sparse data; A framework global hash fusion brightness value set determination module is used to obtain the global spontaneous radiation component and the global environmental radiation component of the framework global granular image, determine the radiation threshold factor according to the global spontaneous radiation component and the global environmental radiation component, and then hash and fuse the framework global saturated adjacent sequence and the radiation threshold factor to obtain the framework global hash fusion brightness value set; a frame global transition value determination module, used to determine the frame global reflection gradient thermal amplitude according to the frame global hash fusion brightness value set, determine the frame global mirror data and the frame global diffuse data of the frame global reflection gradient thermal amplitude, perform transition connection on the frame global mirror data and the frame global diffuse data, and obtain the frame global transition value; The automobile display screen frame state determination module is used to determine the automobile display screen frame as being in a distorted state when the frame global jump value is greater than a preset frame global stretch threshold, and upload it to the automobile display screen frame infrared detection information library.

9. A computer device, It is characterized in that The computer device comprises a memory and a processor, wherein the memory stores codes, and the processor is configured to obtain the codes and execute the infrared-based size detection method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program. It is characterized in that When the computer program is executed by a processor, the infrared-based size detection method according to any one of claims 1 to 7 is implemented.