Target measurement method and device, computer device and medium

Abstract

Embodiments of the present application disclose a target measurement method and device, computer equipment and a medium. In a specific embodiment, the method comprises: performing binaryzation segmentation on an image containing a target to obtain a binaryzation image containing a background region and a connected target region; iteratively performing an erosion operation based on a preset operator on the target region in the binaryzation image until a termination condition is met, the termination condition being that the number of target regions contained in the binaryzation image is not one; and determining a minimum pixel width of the target according to the number of times of performing the erosion operation and parameters of the preset operator. The embodiment greatly improves the measurement efficiency on the basis of ensuring the measurement accuracy of the minimum width of the target, and the measurement accuracy can be set.

Description

Technical Field

[0001] This invention relates to the field of measurement and testing technology. More specifically, it relates to a target measurement method and apparatus, computer equipment, and medium. Background Technology

[0002] Currently, many testing scenarios require the measurement of target dimensions. For example, in display panel testing, it is necessary to measure whether the width of the strip of optically clear adhesive (OCA) used for bonding the film layer meets the requirements. Since the coated strip of OCA has irregular edges, the measurement focuses on whether its minimum width meets the requirements. The inventors have found that existing measurement methods, such as minimum width measurement based on target edge detection, suffer from problems such as excessive preparatory work and low measurement efficiency. Summary of the Invention

[0003] The purpose of this invention is to provide a target measurement method and apparatus, computer equipment and medium to solve at least one of the problems existing in the prior art.

[0004] To achieve the above objectives, the present invention adopts the following technical solution:

[0005] The first aspect of the present invention provides a target measurement method, comprising:

[0006] Binarization segmentation is performed on the image containing the target to obtain a binary image containing a background region and a connected target region;

[0007] The target region in the binarized image is iteratively eroded using a preset operator until a termination condition is met, wherein the number of target regions contained in the binarized image is not one.

[0008] The minimum pixel width of the target is determined based on the number of times the erosion operation is executed and the parameters of the preset operator.

[0009] Optionally, the preset operator is a circular operator.

[0010] Optionally, the parameter of the preset operator is the radius of the circular operator.

[0011] Optionally, the minimum pixel width a of the target is determined as a∈[c×r×2-2r,c×r×2], where r is the radius of the circular operator and c is the number of times the erosion operation is executed.

[0012] Optionally, the radius r of the circular operator is r=(n / m) / 4, where n is the minimum division required for accuracy and m is the actual length corresponding to the pixel length.

[0013] Optionally, the binarization segmentation of the image containing the target includes: performing binarization segmentation of the image containing the target using a U-Net network.

[0014] Optionally, the connected target region is an irregularly shaped target region.

[0015] Optionally, the target is the optical adhesive of the display panel, and the irregularly shaped target area is a strip-shaped target area with irregular edges.

[0016] A second aspect of the present invention provides a target measuring device, comprising:

[0017] The segmentation module is used to perform binarization segmentation on an image containing a target, resulting in a binary image containing a background region and a connected target region.

[0018] The erosion module is used to iteratively perform erosion operations based on preset operators on the target regions in the binarized image until a termination condition is met, wherein the termination condition is that the number of target regions contained in the binarized image is not one.

[0019] The determination module is used to determine the minimum pixel width of the target based on the number of times the erosion operation is executed and the parameters of the preset operator.

[0020] A third aspect of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method provided in the first aspect of the present invention.

[0021] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, constitutes the method provided in the first aspect of the present invention.

[0022] The beneficial effects of this invention are as follows:

[0023] The technical solution described in this invention significantly improves measurement efficiency while ensuring the measurement accuracy of the minimum width of targets such as strip-shaped optical adhesives with irregular edges, and the measurement accuracy can be set. Attached Figure Description

[0024] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0025] Figure 1 A flowchart of a target measurement method provided by an embodiment of the present invention is shown.

[0026] Figure 2 A schematic diagram of the U-Net network structure is shown.

[0027] Figure 3 A schematic diagram of a binarized image is shown.

[0028] Figure 4 Showing the Figure 3 The diagram shows the result of performing three erosion operations on the binarized image.

[0029] Figure 5 Another flowchart of the target measurement method provided in an embodiment of the present invention is shown.

[0030] Figure 6 A schematic diagram of the target measuring device provided in an embodiment of the present invention is shown.

[0031] Figure 7 A schematic diagram of the structure of a computer system for implementing the target measurement device provided in an embodiment of the present invention is shown. Detailed Implementation

[0032] To more clearly illustrate the present invention, the following description, in conjunction with embodiments and accompanying drawings, further explains the invention. Similar components in the drawings are indicated by the same reference numerals. Those skilled in the art should understand that the specific description below is illustrative rather than restrictive and should not be construed as limiting the scope of protection of the present invention.

[0033] In scenarios such as display panel testing, it is necessary to measure whether the width of the strip of optically clear adhesive (OCA) used to bond the film layer meets the requirements. Since the coated strip of OCA has irregular edges, the measurement focuses on whether its minimum width meets the requirements. The inventors have found that existing measurement methods, such as minimum width measurement based on target edge detection, suffer from problems such as excessive preparatory work and low measurement efficiency.

[0034] Morphological processing in image processing mainly targets binarized images, including erosion and dilation operations. Erosion and dilation operations are usually used in combination, such as opening and closing operations: the operation of erosion followed by dilation is called opening operation, which can eliminate small targets, separate targets in thin areas, and smooth the boundaries of larger targets; the operation of dilation followed by erosion is called closing operation, which can fill small holes in the target, connect neighboring targets, and smooth the boundaries.

[0035] The inventors discovered that performing an erosion operation solely on the target region of a binarized image reduces and thins the target region (i.e., compresses the target). Therefore, one embodiment of this invention provides a target measurement method based on morphological erosion operations that is fundamentally different from existing measurement methods based on target edge detection in terms of measurement scheme and principle, in order to obtain the minimum pixel width of the target.

[0036] like Figure 1 As shown, an embodiment of the present invention provides a target measurement method, which includes the following steps:

[0037] S101. Perform binarization segmentation on the image containing the target to obtain a binarized image containing a background region and a connected target region.

[0038] In one possible implementation, the connected target region is an irregularly shaped target region.

[0039] In one possible implementation, the target is the optical adhesive of the display panel, and the irregularly shaped target area is a strip-shaped target area with irregular edges.

[0040] In one specific example, the display device including the display panel can be any product or component with display functionality, such as electronic paper, mobile phone, tablet computer, television, monitor, laptop computer, digital photo frame, or navigator.

[0041] In a specific example, the target is a strip of optically clear adhesive (OCA) used to bond the film layer of a display panel, which has irregular edges. An image containing the target is obtained by an image acquisition device, such as an industrial camera, on the test line, capturing an image of the area of ​​the display panel coated with the OCA; the target area in the image is a simply connected region.

[0042] In one possible implementation, the binarization segmentation of the image containing the target includes: performing binarization segmentation of the image containing the target using a U-Net network.

[0043] In a specific example, the U-Net network is a trained U-Net network. The training image set includes multiple images containing images of the transparent optical adhesive of the display panel as targets. The images containing the transparent optical adhesive of the display panel are input into the trained U-Net network, which outputs a binarized image, for example, where the transparent optical adhesive area of ​​the display panel is 1 and the background area other than the transparent optical adhesive area is 0. The U-Net network structure is U-shaped, as shown below. Figure 2As shown, its essence is an improved fully convolutional neural network (FCN). The U-Net network includes an encoder network 201 for downsampling and a decoder network 202 for upsampling. The input image is reduced by multiple convolutions through the encoder network 201 to obtain multiple feature maps, which are then expanded by multiple deconvolutions through the decoder network 202. During this process, the feature maps are also concatenated with those corresponding to the encoder network 201 to combine deep and shallow features, refine the image, and facilitate the acquisition of features in different dimensions of the input image, thereby improving the performance of image segmentation and object recognition. If the two feature maps to be concatenated are of different sizes, after copying the feature map corresponding to the encoder network 201, they need to be cropped before being transmitted to the decoder network 202 for concatenation.

[0044] like Figure 2 As shown in this example:

[0045] In order from shallow to deep, the encoder network 201 includes three downsampling layers and the decoder network 202 includes three upsampling layers. It should be noted that the number of downsampling layers and the number of upsampling layers are not limited to three, but the number of downsampling layers and the number of upsampling layers should be the same.

[0046] Between encoder network 201 and decoder network 202, the network is arranged from shallow to deep (… Figure 2 Three feature propagation layers, 2031, 2032, and 2033, were established in the order from top to bottom on the left side of the middle section.

[0047] It should be noted that the U-Net network may also include a classification layer (not shown in the figure) after the decoder network 202. By setting the classifier, the probability of pixels in the feature map corresponding to the image data belonging to different categories can be calculated. This enables the prediction of the category of pixels in the feature map corresponding to the image data, thereby performing pixel classification, i.e., image segmentation, to achieve target recognition.

[0048] For encoder network 201, feature extraction of image data can be performed through three downsampling layers, and the feature map obtained by feature extraction can be transmitted to decoder network 202 through three feature propagation layers 2031, 2032 and 2033.

[0049] Specifically, in this example, the downsampling layer includes two convolutional layers and one pooling layer, which is a max pooling layer. Image data is input to the shallowest downsampling layer in the encoder network 201. Figure 2 (Top left of the image) The input image data is convolved through the shallowest downsampling layer (e.g.) Figure 2The diagram shows two convolutions to obtain the feature map corresponding to the shallowest downsampling layer, which is then processed by downsampling (max pooling downsampling) and input into the next shallowest downsampling layer.

[0050] According to the order from shallow to deep ( Figure 2 Following the order from top to bottom on the left side, the other two downsampling layers in the encoder network 201 are traversed to obtain the feature maps corresponding to the other two downsampling layers.

[0051] Between encoder network 201 and decoder network 202, the feature maps mentioned above are propagated through feature propagation layers 2031, 2032 and 2033, respectively.

[0052] The output of the deepest downsampling layer is processed by convolution. Figure 2 After the bottom horizontal hollow arrow indicates this, the feature map is directly transmitted to the deepest upsampling layer in the decoder network 202. Figure 2 (As indicated by the solid upward arrow at the bottom right of the middle section).

[0053] In encoder network 201, horizontal hollow arrows represent convolution processing performed by convolutional layers, and downward solid arrows represent max pooling downsampling processing performed by max pooling layers.

[0054] As the encoder network deepens, the feature extraction of image data gradually shifts from local description to global description, thereby more accurately describing the image data and ensuring the accuracy of image segmentation.

[0055] For the decoder network 202, it not only extracts features from the image data through several upsampling operations, but also performs feature fusion on the image data.

[0056] Specifically, in this example, the first upsampling layer (the deepest upsampling layer) includes a transposed convolutional layer (also called deconvolution or up-convolution), and the other upsampling layers include a transposed convolutional layer and two convolutional layers. The deepest upsampling layer performs transposed convolutional upsampling on the feature map obtained after convolution processing of the output of the deepest downsampling layer to obtain the first feature map to be fused.

[0057] The first feature map to be fused is input into the second deepest upsampling layer, fused with the feature map corresponding to the deepest downsampling layer transmitted through the propagation layer 2033, and then subjected to convolution processing (e.g. Figure 2 The diagram shows two convolutions to obtain the second feature map to be fused.

[0058] In order from darkest to lightest ( Figure 2(From bottom to top on the right side of the middle) traverse the other two upsampling layers to obtain the feature maps corresponding to the upsampling layers.

[0059] Once the traversal is completed, the feature map corresponding to the image data is obtained from the feature map corresponding to the shallowest upsampling layer.

[0060] In decoder network 202, horizontal hollow arrows represent convolution processing performed by convolutional layers, and upward solid arrows represent transposed convolution upsampling processing performed by transposed convolutional layers.

[0061] Through the above process, the combination of encoder and decoder networks not only effectively reduces the computational load of image segmentation and improves segmentation efficiency, but also fully guarantees the accuracy of image segmentation.

[0062] In addition, other segmentation networks such as fully convolutional neural networks (FCN) and instance segmentation models (Mask RCNN) can also be used in this embodiment to perform binarization segmentation on images containing targets, which will not be elaborated here.

[0063] Continuing with the previous example, the binarized image obtained by binarizing the image containing the transparent optical adhesive of the display panel using the U-Net network is as follows: Figure 3 As shown, Figure 3 In the binarized image shown, a connected region with a diagonal pattern is the target region 301, and the remaining blank regions are the background regions 302.

[0064] S102. Iteratively perform an erosion operation based on a preset operator on the target region in the binarized image until a termination condition is met, wherein the termination condition is that the number of target regions contained in the binarized image is not one.

[0065] In a specific example, when the target width varies across different locations (e.g., the target is a strip of transparent optical adhesive with irregular edges on a display panel), if the initial binarized image contains a single connected target region (i.e., the number of target regions is one), during the iterative erosion operation, the first occurrence of the binarized image containing a target region other than one includes two scenarios: The first, more common scenario, is that the number of target regions in the binarized image is greater than one. This indicates that after at least one erosion operation, a break has occurred in the target region, and the break must correspond to the position with the minimum target width. Therefore, the erosion operation can be terminated at this point to determine the minimum target pixel width. The second scenario, a less probable event, is that the number of target regions in the binarized image changes directly from one to zero (i.e., the target region is eroded away, and there is no target region in the binarized image). In this case, the erosion operation should also be terminated to determine the minimum target pixel width.

[0066] In one possible implementation, the preset operator is a circular operator.

[0067] By adopting this possible implementation method, the target measurement method provided in this embodiment has isotropic characteristics. It can measure the minimum pixel width of targets such as strip-shaped optical adhesive with irregular edges without the need to correct the image angle and give a measurement coordinate system, which can simplify the preliminary work of measurement.

[0068] In addition, if the direction of extension of the target, such as a strip of transparent optical adhesive with irregular edges on a display panel, is known, operators of other shapes, such as squares, can be used for etching operations.

[0069] In a specific example, for Figure 3 The schematic diagram shown is an example of the result of iteratively performing an erosion operation based on a circular operator on a binarized image. Figure 4 As shown, Figure 4 As shown, the closed boundary 401 is Figure 3 The target region 301 is a connected target region. After the first erosion operation on the target region defined by the closed boundary 401, the target region obtained is a target region defined by the closed boundary 402. After the second erosion operation on the target region defined by the closed boundary 402, the target region obtained is a target region defined by the closed boundary 403. After the third erosion operation on the target region defined by the closed boundary 403, the target region obtained is two target regions defined by the two closed boundaries 404. That is, after the third erosion operation, the termination condition that the number of target regions contained in the binarized image is not one is met. At this time, the erosion operation can be terminated and the subsequent determination of the minimum pixel width of the target can be carried out.

[0070] S103. Determine the minimum pixel width of the target based on the number of times the erosion operation is executed and the parameters of the preset operator.

[0071] In one possible implementation, the parameter of the preset operator is the radius of the circular operator. That is, the minimum pixel width of the target is determined based on the number of times the erosion operation is performed and the radius of the circular operator. The minimum pixel width of the target is, for example, the pixel width at the narrowest point of a strip of transparent optical adhesive with irregular edges on a display panel in the image.

[0072] In one possible implementation, the radius r of the circular operator is r=(n / m) / 4, where n is the minimum division required for precision and m is the actual length corresponding to the pixel length.

[0073] In a specific example, the unit of the minimum division *n* for the accuracy requirement and the unit of the actual length *m* corresponding to the pixel length are both, for example, millimeters. The actual length *m* corresponding to the pixel length is related to image acquisition parameters such as image resolution and can be obtained through pre-calibration; while the radius *r* of the circular operator is related to the minimum division *n* for the accuracy requirement, enabling the measurement accuracy to be settable. For the above formula, since each erosion operation is performed on the width in two relative directions, the radius *r* of the circular operator needs to be divided by 2. Furthermore, since the minimum division *n* for the accuracy requirement is estimated using a bisection method, the radius *r* of the circular operator is half of the minimum division *n* for the accuracy requirement; therefore, the radius *r* of the circular operator needs to be divided by 2 again.

[0074] In one possible implementation, the minimum pixel width 'a' of the target is determined as a∈[c×r×2-2r, c×r×2], that is, the range of values ​​for 'a' is c×r×2-r±r, where r is the radius of the circular operator and c is the number of times the erosion operation is executed.

[0075] That is, if the erosion operation is executed once, the minimum pixel width 'a' of the target is determined to be in the range [0, 2r]; if the erosion operation is executed twice, the minimum pixel width 'a' of the target is determined to be in the range [r, 3r], and so on. Given a minimum resolution 'n' for the specified accuracy requirement, the measured value of the minimum pixel width 'a' of the target is only related to the number of erosion operations executed, 'c'. The smaller the minimum resolution 'n' for the specified accuracy requirement, the smaller the radius 'r' of the circular operator, and the higher the measurement accuracy. Based on the above, if the erosion operation is executed once, it indicates that the radius 'r' of the circular operator is large, resulting in low measurement accuracy. In this case, the radius 'r' of the circular operator can be adjusted (for example, if the radius 'r' of the circular operator is determined by r = (n / m) / 4, the radius 'r' of the circular operator can be adjusted by adjusting the minimum resolution 'n' for the accuracy requirement), and the iterative erosion operation can be re-executed to measure the minimum pixel width of the target again with higher measurement accuracy.

[0076] Understandably, given the minimum pixel width 'a' of the target, and since the actual length 'm' corresponding to the known pixel length is known, the minimum actual width of the target can be obtained from the minimum pixel width 'a'.

[0077] In addition, if the operator is a square operator, then the value of its side length j can be j / 2==(n / m) / 4, and the value range of the minimum pixel width a of the target can be calculated by a∈[c×j×2-2j, c×j×2], that is, the value range of a is c×j×2-j±j.

[0078] It is understood that in this embodiment, the shape of the target is not limited to a strip shape. For example, irregular targets such as circular or rectangular shapes also have minimum pixel widths representing the minimum diameter and minimum side length, respectively. It should be noted that the target measurement method provided in this embodiment is particularly suitable for detecting the minimum pixel width of irregularly shaped targets, but it is also suitable for detecting the width of regular-shaped targets such as strips and circles.

[0079] In a specific example, based on the above implementation methods, the flow of the target measurement method provided in this embodiment is as follows: Figure 5 As shown.

[0080] The target measurement method provided in this embodiment has isotropic characteristics. It can measure the minimum pixel width of targets such as strip-shaped optical adhesive with irregular edges without the need to correct the image angle or give a measurement coordinate system. This simplifies the preliminary work of measurement, greatly improves the measurement efficiency while ensuring the accuracy of the minimum target width measurement, and the measurement accuracy can be set.

[0081] like Figure 6 As shown, another embodiment of the present invention provides a target measuring device, comprising:

[0082] The segmentation module 601 is used to perform binarization segmentation on an image containing a target to obtain a binarized image containing a background region and a connected target region;

[0083] The erosion module 602 is used to iteratively perform an erosion operation based on a preset operator on the target region in the binarized image until a termination condition is met, wherein the termination condition is that the number of target regions contained in the binarized image is not one.

[0084] The determination module 603 is used to determine the minimum pixel width of the target based on the number of times the erosion operation is executed and the parameters of the preset operator.

[0085] It should be noted that the principle and working process of the target measurement device provided in this embodiment are similar to the target measurement method described above. The relevant parts can be referred to the above description and will not be repeated here.

[0086] like Figure 7 As shown, a computer system suitable for implementing the target measuring device provided in the above embodiments includes a central processing module (CPU), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) or a program loaded from a storage portion into a random access memory (RAM). Various programs and data required for the operation of the computer system are also stored in the RAM. The CPU, ROM, and RAM are connected via a bus. An input / output (I / O) interface is also connected to the bus.

[0087] The following components are connected to the I / O interface: input sections including keyboards, mice, etc.; output sections including liquid crystal displays (LCDs) and speakers, etc.; storage sections including hard disks, etc.; and communication sections including network interface cards such as LAN cards and modems. The communication sections perform communication processing via networks such as the Internet. Drives are also connected to the I / O interface as needed. Removable media, such as disks, optical disks, magneto-optical disks, semiconductor memories, etc., are installed on the drive as needed so that computer programs read from them can be installed into the storage section as required.

[0088] Specifically, according to this embodiment, the process described in the flowchart above can be implemented as a computer software program. For example, this embodiment includes a computer program product comprising a computer program tangibly embodied on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium.

[0089] The flowcharts and schematic diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of the system, method, and computer program product of this embodiment. In this regard, each block in the flowchart or schematic diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the schematic diagram and / or flowchart, and combinations of blocks in the schematic diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0090] The modules described in this embodiment can be implemented in software or hardware. These modules can also be housed in a processor; for example, a processor may be described as including a segmentation module, an erosion module, and a determination module. The names of these modules do not necessarily limit the functionality of the module itself. For example, the determination module may also be described as an "acquisition module" or a "measurement module."

[0091] In another aspect, this embodiment also provides a non-volatile computer storage medium. This non-volatile computer storage medium can be the non-volatile computer storage medium included in the above-described device in the above embodiments, or it can be a separate non-volatile computer storage medium not installed in the terminal. The non-volatile computer storage medium stores one or more programs. When the one or more programs are executed by a device, the device: performs binarization segmentation on an image containing a target to obtain a binarized image containing a background region and a connected target region; iteratively performs an erosion operation based on a preset operator on the target region in the binarized image until a termination condition is met, wherein the termination condition is that the number of target regions contained in the binarized image is not one; and determines the minimum pixel width of the target based on the number of times the erosion operation is executed and the parameters of the preset operator.

[0092] In the description of this invention, it should be noted that the terms "upper," "lower," etc., indicating the orientation or positional relationship are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Unless otherwise expressly specified and limited, the terms "installed," "connected," and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication between two elements. For those skilled in the art, the specific meaning of the above terms in this invention can be understood according to the specific circumstances.

[0093] It should also be noted that in the description of this invention, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0094] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. For those skilled in the art, other variations or modifications can be made based on the above description. It is impossible to exhaustively list all the implementation methods here. All obvious variations or modifications derived from the technical solutions of the present invention are still within the protection scope of the present invention.

Claims

1. A target measurement method, characterized in that, include: Binarization segmentation is performed on the image containing the target to obtain a binary image containing a background region and a connected target region; The target region in the binarized image is iteratively eroded using a preset operator until a termination condition is met, wherein the number of target regions contained in the binarized image is not one. The minimum pixel width of the target is determined based on the number of times the erosion operation is executed and the parameters of the preset operator. The preset operator is a circular operator; The parameter of the preset operator is the radius of the circular operator; The minimum pixel width 'a' of the target is determined as a∈[c×r×2-2r,c×r×2], where r is the radius of the circular operator and c is the number of times the erosion operation is executed.

2. The method according to claim 1, characterized in that, The radius r of the circular operator is r=(n / m) / 4, where n is the minimum division required for precision and m is the actual length corresponding to the pixel length.

3. The method according to claim 1, characterized in that, The binarization segmentation of the image containing the target includes: performing binarization segmentation of the image containing the target using a U-Net network.

4. The method according to claim 1, characterized in that, The connected target region is an irregularly shaped target region.

5. The method according to claim 4, characterized in that, The target is the optical adhesive of the display panel, and the irregularly shaped target area is a strip-shaped target area with irregular edges.

6. A target measuring device, characterized in that, include: The segmentation module is used to perform binarization segmentation on an image containing a target, resulting in a binary image containing a background region and a connected target region. The erosion module is used to iteratively perform erosion operations based on preset operators on the target regions in the binarized image until a termination condition is met, wherein the termination condition is that the number of target regions contained in the binarized image is not one. The determination module is used to determine the minimum pixel width of the target based on the number of times the erosion operation is executed and the parameters of the preset operator; The preset operator is a circular operator; The parameter of the preset operator is the radius of the circular operator; The minimum pixel width 'a' of the target is determined as a∈[c×r×2-2r,c×r×2], where r is the radius of the circular operator and c is the number of times the erosion operation is executed.

7. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1-5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-5.

Images