CN117974576A

CN117974576A - Image quality detection method, device, computer equipment and storage medium

Info

Publication number: CN117974576A
Application number: CN202410039945.8A
Authority: CN
Inventors: 柳锐; 易振彧; 潘阳山; 莫宇; 刘枢; 吕江波; 沈小勇
Original assignee: Shenzhen Smartmore Technology Co Ltd
Current assignee: Shenzhen Smartmore Technology Co Ltd
Priority date: 2024-01-09
Filing date: 2024-01-09
Publication date: 2024-05-03

Abstract

The application relates to an image quality detection method, an image quality detection device, computer equipment and a storage medium. The method comprises the following steps: performing contour detection on the target object in the acquired image to be detected to obtain a contour curve of the target object; determining a gradient amplitude image of an image to be detected; the gradient amplitude image is an image representing the degree of variation between adjacent pixel points in the image to be detected; for each contour point on the contour curve, determining the normal direction of the targeted contour point; determining a gradient amplitude maximum point corresponding to the aimed contour point according to the gradient amplitude image and the normal direction; and determining the imaging quality of the image to be detected according to the maximum point of the gradient amplitude corresponding to each contour point on the contour curve. By adopting the application, the efficiency and the accuracy of image quality detection can be improved.

Description

Image quality detection method, device, computer equipment and storage medium

Technical Field

The present application relates to the field of computer technologies, and in particular, to an image quality detection method, an image quality detection device, a computer device, and a storage medium.

Background

In the field of electronics manufacturing, printed circuit boards are an integral part of modern electronic devices. Quality detection of printed circuit boards is an important step in ensuring proper operation of electronic devices, reducing cost, and improving quality and reliability. Therefore, detecting the imaging quality of a printed circuit board is critical to the performance and reliability of an electronic device.

However, at present, for the imaging quality detection of the printed circuit board, a complex detection process is often required and accompanied by manual intervention, so that the detection efficiency of the imaging quality detection is low, and the detection result is not stable and accurate.

Disclosure of Invention

In view of the foregoing, it is desirable to provide an image quality detection method, apparatus, computer device, computer readable storage medium, and computer program product that can achieve an improvement in efficiency and accuracy of image quality detection.

In a first aspect, the present application provides an image quality detection method, including:

performing contour detection on the target object in the acquired image to be detected to obtain a contour curve of the target object;

determining a gradient amplitude image of an image to be detected; the gradient amplitude image is an image representing the degree of variation between adjacent pixel points in the image to be detected;

for each contour point on the contour curve, determining the normal direction of the targeted contour point;

Determining a gradient amplitude maximum point corresponding to the aimed contour point according to the gradient amplitude image and the normal direction;

And determining the imaging quality of the image to be detected according to the maximum point of the gradient amplitude corresponding to each contour point on the contour curve.

In a second aspect, the present application provides an image quality detection apparatus comprising:

The detection module is used for carrying out contour detection on the target object in the acquired image to be detected to obtain a contour curve of the target object;

The first determining module is used for determining a gradient amplitude image of the image to be detected; the gradient amplitude image is an image representing the degree of variation between adjacent pixel points in the image to be detected;

The second determining module is used for determining the normal direction of each contour point on the contour curve;

The third determining module is used for determining a gradient amplitude maximum point corresponding to the aimed contour point according to the gradient amplitude image and the normal direction;

and the fourth determining module is used for determining the imaging quality of the image to be detected according to the gradient amplitude maximum point corresponding to each contour point on the contour curve.

In a third aspect, the application provides a computer device comprising a memory storing a computer program and a processor implementing the steps of the method described above when the processor executes the computer program.

In a fourth aspect, the present application provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the method described above.

In a fifth aspect, the application provides a computer program product comprising a computer program which, when executed by a processor, implements the steps of the method described above.

The image quality detection method, the image quality detection device, the computer equipment, the computer readable storage medium and the computer program product can obtain the contour curve of the target object in the image to be detected by the contour detection method, and can provide more accurate and flexible edge detection results compared with the traditional threshold method. And acquiring a gradient amplitude image corresponding to the image to be detected, analyzing the maximum gradient amplitude point corresponding to each contour point according to the normal direction of each contour point on the contour curve, and finally determining the imaging quality of the image to be detected. Compared with the traditional method, the method has the advantages that the calculation process in the image quality detection can be simplified, the detection efficiency is improved, the manual intervention is not needed, the influence of subjective factors is eliminated, and the stability and the accuracy of the detection result are improved.

Drawings

Fig. 1 is an application environment diagram of an image quality detection method according to an embodiment of the present application;

fig. 2 is a schematic flow chart of an image quality detection method according to an embodiment of the present application;

FIG. 3 is a schematic diagram of an image to be detected according to an embodiment of the present application;

FIG. 4 is a schematic diagram of a binary image according to an embodiment of the present application;

FIG. 5 is a schematic diagram of a gradient magnitude accumulation value curve according to an embodiment of the present application;

Fig. 6 is a schematic diagram of an inner layer wire in a printed circuit board according to an embodiment of the present application;

FIG. 7 is a schematic illustration of vignetting effect according to an embodiment of the present application;

Fig. 8 is a block diagram of an image quality detecting apparatus according to an embodiment of the present application;

FIG. 9 is a diagram illustrating an internal architecture of a computer device according to an embodiment of the present application;

FIG. 10 is an internal block diagram of another computer device according to an embodiment of the present application;

fig. 11 is an internal structural diagram of a computer-readable storage medium according to an embodiment of the present application.

Detailed Description

The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.

The image quality detection method provided by the embodiment of the application can be applied to an application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a communication network. The data storage system may store data that the server 104 needs to process. The data storage system may be integrated on the server 104 or may be located on a cloud or other network server. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, internet of things devices, portable wearable devices, and the internet of things devices may be smart speakers, smart televisions, smart air conditioners, smart vehicle devices, etc. The portable wearable device may be a smart watch, smart bracelet, headset, or the like. The server 104 may be implemented as a stand-alone server or as a server cluster of multiple servers.

As shown in fig. 2, an embodiment of the present application provides an image quality detection method, and the method is applied to the terminal 102 or the server 104 in fig. 1 for illustration. It is understood that the computer device may include at least one of a terminal and a server. The method comprises the following steps:

S202, performing contour detection on the target object in the acquired image to be detected to obtain a contour curve of the target object.

The image to be detected is not particularly limited, and the image to be detected can be set according to actual needs. In some embodiments, the image to be detected is a central area cut out from the original image of the printed circuit board, and it is understood that the size of the central area may be set according to actual needs. For example 200 x 200 pixels. Contour curves refer to curves representing the boundary of an object or target in a two-dimensional image, the contour curves being obtained by separating the object from the background. In general, a contour curve may be defined as a series of ordered two-dimensional coordinate points that are sequentially connected to form a closed curve that follows the boundary of an object or target, with each coordinate point representing a pixel on the boundary of the object or target. The application does not limit the target object specifically, and can be set according to actual needs. For example, the target object is a substrate portion on a printed circuit board.

Specifically, referring to fig. 3, firstly, according to the actual detection requirement, an image of a certain area is selected from the original image of the printed circuit board as an image to be detected, and the image to be detected mainly includes a substrate 302 and a wire 304. Wherein, the wires are also called circuit traces or wires, which are used to connect circuit paths between electronic components for transmitting current and signals. The substrate refers to the body portion that carries the electronic components and wires. And taking the substrate in the image to be detected as a target object according to the actual detection requirement. And then carrying out graying treatment on the image to be detected to obtain a corresponding gray image, and carrying out image segmentation on the gray image by using a machine learning method so as to separate the target object from the background. For example, HRNet (a specific network architecture used for image segmentation in deep learning) is used to perform image segmentation on the gray-scale image, so as to obtain a corresponding binary image. And then using findContours functions of an OpenCV (open source computer vision library) (functions used for searching contours in images in the open source computer vision library) or other methods to obtain contour points of the target object in the binary image, and further forming a closed curve, namely, a contour curve of the target object.

S204, determining a gradient amplitude image of the image to be detected; the gradient amplitude image is an image representing the variation degree between adjacent pixel points in the image to be detected.

The gradient amplitude image is an image calculated according to gradient information of the image. Each pixel of the gradient magnitude image represents the degree of variation between adjacent pixels in the image. In an image, the gray value of each pixel point represents the luminance or color information of that point. The gradient amplitude image shows the change condition of the gray value of the adjacent pixel points. If the gray value of the pixel point around a pixel point changes greatly, the gradient amplitude of the pixel point will be larger; and if the gray value of the surrounding pixel changes less, the gradient magnitude of the pixel will be smaller.

Gradient magnitude images are typically used to represent edges, textures, or other detailed information in the image. The larger gradient magnitude generally corresponds to a boundary in the image or an edge of the object. This is because there is often a place where the gray value changes greatly at the boundary or edge, while the gray value changes less for the pixels in other areas. By observing the gradient magnitude image, we can locate and analyze edges and textures in the image.

Specifically, firstly, based on the image to be detected, selecting an average value method, a weighted average method, a maximum value method and the like according to actual needs, converting the image to be detected into a gray image, and if the image is already the gray image, skipping the step. And (3) applying a Sobel operator to the gray image, and calculating the change condition of gray values of pixel points around each pixel point. The horizontal direction, the vertical direction or both directions can be selected for calculation as desired. And performing square sum evolution operation on the gradient result obtained by calculating the Sobel operator to obtain the gradient amplitude of each pixel point. And (3) carrying out graying treatment on the obtained gradient amplitude values to ensure that all gradient amplitude values are in a proper range (for example, 0 to 255), so that the result can be correctly displayed. Finally, the obtained gradient amplitude map can be used for displaying edge information in the image, and the position with large gradient amplitude indicates the existence of edges.

S206, determining the normal direction of each contour point on the contour curve.

The normal direction refers to a direction tangential to a curved surface or curve and perpendicular to the curved surface or curve. For a surface in three-dimensional space, the normal direction is the normal vector to a point on the surface, which indicates the orientation of the surface at that point. In plane geometry, the normal direction of the curve is perpendicular to the tangent of the curve, i.e. the slopes of the two are opposite numbers, and the normal direction at a certain point on the curve can be obtained by calculating the slope of the tangent at the certain point and then the opposite number of the slope.

Specifically, for each contour point, calculating the normal direction, that is, calculating the direction perpendicular to the contour curve, the coordinate difference of the adjacent contour points can be used to obtain the segmentation direction, and then the obtained segmentation direction is rotated by 90 degrees, so that the corresponding normal direction of each contour point can be obtained.

In some embodiments, the coordinate difference of the adjacent contour points is calculated, and a tangential direction vector can be obtained. For each tangential direction vector, rotating it by 90 degrees gives a normal direction vector. The rotation of 90 degrees can be achieved simply by exchanging the x (horizontal) and y (vertical) coordinates of the vector and taking the negative of one of the exchanged x-axis and y-axis coordinates. For each normal direction vector, normalization processing is performed. The vector is divided by its length such that the length of the vector becomes 1, still maintaining the same direction. The resulting normal direction can be used to describe the vertical direction of the contour curve.

S208, determining a gradient amplitude maximum point corresponding to the aimed contour point according to the gradient amplitude image and the normal direction.

Where the maximum gradient magnitude point refers to the point at which the magnitude of the gradient (i.e., the rate of change of the function or data set at each point) is greatest across a function or data set. The gradient magnitude maximum point typically represents the local maximum on the function or the steepest point on the data set. In the fields of mathematics and computer vision, the maximum point of gradient magnitude is often used to find the edge or feature point of an image. By calculating the gradient of each point on the function or dataset and finding the point with the greatest gradient magnitude, important locations on the function or dataset can be quickly located.

Specifically, in the gradient magnitude image, the following operation is performed for each contour point, taking one of the contour points as an example. Firstly, determining the corresponding normal direction of the contour point, and then taking the contour point as a starting point (starting from the contour point), and searching the point with the largest gradient amplitude, namely the gradient amplitude maximum point, leftwards and rightwards along the corresponding normal direction. It will be appreciated that the above procedure may be implemented using a maximum search or other method.

S210, determining the imaging quality of the image to be detected according to the maximum gradient amplitude point corresponding to each contour point on the contour curve.

The imaging quality of the image to be detected can be defined as the visual definition, detail retention degree, distortion degree and other aspects of the image. Specifically, the imaging quality represents the indicators of the definition, noise level, sharpness, brightness uniformity, color accuracy, and the like of the image. It will be appreciated that the definition of imaging quality is not particularly limited in the present application, and may be set according to actual needs.

Specifically, in the gradient amplitude image, finding the maximum gradient amplitude point corresponding to each contour point. Firstly, each contour point is traversed, the accumulated value corresponding to each contour point is stored in an array, and the array is 'line diffusion function', which is used for representing gradient change conditions along the normal direction of the contour, and the definition and the contrast of the image are reflected. Then, in the LSF-like array, the peak value and the two-edge value are accumulated to be used as the image quality evaluation value of the image to be detected. And finally, comparing the obtained image quality evaluation value of the image to be detected with a preset threshold value to judge whether the image to be detected is a defect.

Therefore, in the embodiment of the application, firstly, peak values and two edge values are accumulated to replace the original complex function fitting and variance solving, so that the peak values and the two edge values are used as the image quality evaluation value of the image to be detected, the calculation speed is improved, and the detection efficiency is improved. And finally, starting from the contour point, searching the point with the maximum gradient amplitude, namely the point with the maximum gradient amplitude, leftwards and rightwards along the corresponding normal direction respectively by determining the normal direction corresponding to the contour point to perform gradient analysis, so that manual intervention can be avoided, the influence of subjective factors is eliminated, the stability and the accuracy of a detection result are improved, and the imaging quality of the image to be detected is estimated more accurately.

In some embodiments, performing contour detection on a target object in an acquired image to be detected to obtain a contour curve of the target object, including:

Determining a binary image of a target object in the acquired image to be detected;

performing initial contour detection on the binary image of the target object to obtain a plurality of initial contour points;

for each two initial contour points in the plurality of initial contour points, determining a line segment formed by taking the two initial contour points as endpoints, and deleting other initial contour points when other initial contour points are arranged in the line segment;

and determining the contour curve of the target object according to the rest initial contour points in the initial contour points.

Specifically, a binary image of a target object in an image to be detected is acquired, the binary image is taken as an input image, contour extraction is performed on the binary image through an OpenCV (contour extraction algorithm), a frame of the binary image is taken as a first boundary, and an initial pixel point is selected on the first boundary. And scanning the binary image by using a raster scanning method from the initial pixel point, and judging the type of the pixel point through the gray value of the pixel point and the gray values of adjacent pixel points when the pixel point with the gray value of not 0 is scanned. The types of pixel points mainly include two types, namely an outer boundary starting point and a hole boundary starting point. And performing boundary tracking from a boundary starting point, taking the boundary starting point as a current point, and adding the current point into the contour point set. The acquisition of the neighborhood pixel point of the current point can be realized by checking a plurality of adjacent pixel points around the current point. Finding the next neighborhood pixel point to be accessed, and checking whether the next neighborhood pixel point is in the image boundary. If it is within the boundary, it is added to the set of contour points and the current point is updated to the next neighborhood pixel point. If the next neighborhood pixel is not within the boundary, indicating that the endpoint of the boundary has been reached. At this time, the contour point set is added as a complete contour to the contour set. And removed from the set of contour points. Repeating the steps until all the boundaries are found, and further determining the contour curve of the target object.

In some embodiments, a binary image of a target object in an image to be detected is obtained, and an edge detection or contour detection algorithm is performed on the binary image of the target object to obtain a plurality of initial contour points. And connecting pixel points between every two points in the obtained multiple initial contour points serving as endpoints through a linear equation to form a line segment. For each line segment it is checked whether there are further initial contour points. If there are other initial contour points, these points are deleted from the plurality of initial contour points, leaving only two endpoints. And finally, determining a final contour curve of the target object through methods such as interpolation or optimization algorithm according to the rest points in the plurality of initial contour points.

It can be seen that in the present embodiment, the boundaries are sampled at intervals of n, which is a positive integer greater than or equal to 1. From the starting point, the method continuously advances through the boundary tracking algorithm, and moves to the nth point in the neighborhood pixel points each time. The coordinates of each nth point need only be stored to represent the entire boundary. While other intermediate points do not need to be stored. By sampling simplification, the consumption of data storage is reduced to a certain extent, and the efficiency of the boundary tracking algorithm is improved.

In some embodiments, determining the binary image of the target object in the acquired image to be detected includes:

Determining an acquired gray level image of an image to be detected;

Inputting a gray level image of an image to be detected into a preset high-resolution model for processing, and outputting a class label corresponding to each pixel point in the image to be detected;

And determining a binary image of the target object in the image to be detected according to the class label corresponding to each pixel point in the image to be detected.

Specifically, referring to fig. 4, first, the image to be detected is subjected to graying processing, so as to obtain a corresponding gray image. Then image segmentation is performed on the gray-scale image by HRNet (specific network architecture for image segmentation in deep learning), and the specific process comprises: first, an image to be detected is converted into a grayscale image. The conversion may be performed using common gray scale conversion formulas, such as averaging, weighted averaging, or using functions provided by an open source library such as OpenCV. A high resolution model is then designed and trained for the image segmentation task using a deep learning approach. And inputting the gray level image of the image to be detected into a high-resolution model trained in advance. And predicting by using the high-resolution model to obtain a class label corresponding to each pixel point in the image to be detected. The probability value output by the model can be converted into a two-class label of which the pixel belongs to the target or the background. Finally, the target object in the image to be detected is extracted according to the class label corresponding to each pixel point, and a binary image 402 of the target object is generated. The pixel points with the probability value larger than the threshold value are considered to belong to the target object, and other pixel points are considered to belong to the background, so that the binary image is generated.

Therefore, in the embodiment, by using the preset high-resolution model, more detail information can be extracted, and the target object and the background can be better distinguished, so that the precision and accuracy of image segmentation are effectively improved. And the threshold parameters in the high-resolution model can be adjusted according to actual needs to improve the application range of the high-resolution model.

In some embodiments, determining a gradient magnitude image of an image to be detected includes:

Respectively carrying out first convolution operation on each pixel point in the image to be detected through a preset horizontal direction operator template to obtain a horizontal gradient amplitude value corresponding to each pixel point in the image to be detected;

respectively performing second convolution operation on each pixel point in the image to be detected through a preset vertical direction operator template to obtain a vertical gradient amplitude value corresponding to each pixel point in the image to be detected;

according to the horizontal gradient amplitude and the vertical gradient amplitude corresponding to each pixel point in the image to be detected, determining the gradient amplitude corresponding to each pixel point in the image to be detected;

And determining a gradient amplitude image of the image to be detected according to the gradient amplitude corresponding to each pixel point in the image to be detected.

Specifically, first, we need to obtain operator templates in the horizontal and vertical directions. Common operator templates include Sobel operators, prewitt operators (Prewitt operators), and the like, which can perform convolution operation on an image, so as to extract gradient information of each pixel point in the image. Then, a horizontal direction operator template is applied to each pixel point of the image to be detected, and a first convolution operation is performed. The horizontal gradient amplitude corresponding to each pixel point in the image can be obtained. Next, a second convolution operation is performed by applying the vertical direction operator template to each pixel point of the image to be detected as well. The vertical gradient amplitude corresponding to each pixel point in the image can be obtained. The gradient amplitude corresponding to each pixel point can be determined through the horizontal gradient amplitude and the vertical gradient amplitude, and the gradient amplitude can be obtained through calculating the square sum of the two gradient amplitudes and the square sum of the two gradient amplitudes. Finally, a gradient magnitude image may be formed based on the gradient magnitude of each pixel, where a larger magnitude represents a more pronounced edge.

It can be seen that in this embodiment, the edges in the image can be more accurately located by calculating the horizontal and vertical gradient magnitudes, respectively, and solving the gradient magnitudes in combination. The method has the advantages of simplicity, easiness in implementation, wide application range and the like.

In some embodiments, determining the gradient amplitude value corresponding to each pixel point in the image to be detected according to the horizontal gradient amplitude value and the vertical gradient amplitude value corresponding to each pixel point in the image to be detected, includes:

Square processing is carried out on the horizontal gradient amplitude of each pixel point in the image to be detected, so as to obtain a first square value;

square processing is carried out on the vertical gradient amplitude value of the pixel point to obtain a second square value;

adding the first square value and the second square value to obtain a third square value;

and (5) taking root numbers for the third square values to obtain gradient amplitude values corresponding to the aimed pixel points.

Specifically, the following operation is performed for each pixel point in the image to be detected, taking one pixel point as an example. First, the horizontal gradient magnitude of the pixel point is calculated by applying a horizontal operator template such as Sobel operator. And squaring the horizontal gradient amplitude to obtain a first square value. And calculating the vertical gradient amplitude of the pixel point by applying an operator template in the vertical direction. The vertical gradient magnitude is then squared to obtain a second square value. Then, the first square value and the second square value are added to obtain a third square value. And finally, carrying out square root processing on the third square value to obtain the gradient amplitude value of the pixel point.

It can be seen that in this embodiment, the total gradient amplitude of the edge is obtained by squaring the horizontal and vertical gradient amplitudes, then adding the two and performing square root processing. The processing mode can further enhance gradient information of the edge, so that the edge is more remarkable in the image, and subsequent processing and analysis are facilitated.

In some embodiments, the gradient magnitude image is a curved image of a plurality of gradient magnitude points; each gradient amplitude point is a point formed by the corresponding pixel point and the gradient amplitude of the corresponding pixel point in the image to be detected; according to the gradient amplitude image and the normal direction, determining a gradient amplitude maximum point corresponding to the aimed contour point comprises the following steps:

Passing through pixel gradient points corresponding to the aimed contour points in the gradient amplitude image, and drawing a target normal along the normal direction;

searching a point with the maximum gradient amplitude in the gradient amplitude image along the target normal line to obtain a gradient amplitude maximum point corresponding to the aimed contour point.

Specifically, in the gradient magnitude image, the pixel points corresponding to the contour points are passed, and the target normal line is drawn along the normal line direction. This gives a straight line tangent to the contour point and perpendicular to the contour. The point with the largest gradient magnitude is looked up left and right in the gradient magnitude image along the target normal. In the course of moving along the target normal direction, the gradient magnitude of each pixel point is checked, and the point having the largest gradient magnitude is recorded.

It can be seen that in the present embodiment, by calculating the normal direction to which the contour point corresponds, a straight line tangent to the contour point and perpendicular to the contour can be determined. Meanwhile, the point with the maximum gradient amplitude value is searched in the gradient amplitude value image, and the most obvious gradient change point can be obtained. These steps can provide more accurate edge information with higher accuracy.

In some embodiments, determining the imaging quality of the image to be detected according to the gradient magnitude maximum point corresponding to each contour point on the contour curve includes:

searching the adjacent points of the maximum point of each gradient amplitude in the gradient amplitude image;

According to the gradient amplitude of each adjacent point of the maximum point of the gradient amplitude, determining the gradient amplitude accumulated value corresponding to each contour point on the contour curve;

And determining the imaging quality of the image to be detected according to the gradient amplitude accumulated value corresponding to each contour point on the contour curve.

Specifically, it is to be readily understood that the "points adjacent to each gradient magnitude maximum point" is not particularly limited, and may be one adjacent point on the left and right of each gradient magnitude maximum point, that is, one adjacent point on the left and one adjacent point on the right, or may be one adjacent point on the left and one adjacent point on the right of each gradient magnitude maximum point, that is, one adjacent point on the left and one adjacent point on the right. The following operations are performed for each contour point in the image to be detected, taking one of the contour points as an example. After finding the maximum gradient amplitude point corresponding to the contour point, finding the pixel points adjacent to the maximum gradient amplitude point, and adding the gradient amplitude values of the adjacent pixel points and the gradient amplitude value of the maximum gradient amplitude point. For example, after finding the highest point of the gradient, taking 3 points (7 points in total) on the left and right sides of the highest point, and accumulating the gradient magnitudes of the points to obtain the accumulated value corresponding to the contour point. The accumulated value of each contour point is then stored in an array, which is the "line spread function" for representing the gradient change along the contour normal, reflecting the sharpness and contrast of the image. In the LSF-like array, the peak value and the two side values are accumulated to be used as an image quality evaluation value of the image to be detected. And finally, comparing the obtained image quality evaluation value of the image to be detected with a preset threshold value to judge whether the image to be detected is a defect.

Therefore, in the embodiment of the application, peak value taking and two-edge value accumulating are used for replacing the original complex function fitting and variance solving, and are used as the image quality evaluation value of the image to be detected, so that the calculation speed is improved, and the detection efficiency is improved.

In some embodiments, determining the imaging quality of the image to be detected according to the respective gradient magnitude accumulated value of each contour point on the contour curve includes:

Determining a gradient amplitude accumulated value curve comprising a plurality of amplitude accumulated points according to the gradient amplitude accumulated value corresponding to each contour point on the contour curve;

Searching an amplitude accumulation point with the maximum gradient amplitude accumulation value in the gradient amplitude accumulation value curve to obtain a target amplitude accumulation point;

And determining the imaging quality of the image to be detected according to the target amplitude accumulation point and a point adjacent to the target amplitude accumulation point in the gradient amplitude accumulation value curve.

Specifically, referring to fig. 5, after the accumulated value of each contour point is stored in an array "LSF-like", the array "LSF-like" is converted into a gradient magnitude accumulated value curve according to the difference of the accumulated value of each contour point, and it is easy to understand that the height of the gradient magnitude accumulated value curve reflects the magnitude of the accumulated value corresponding to each contour point. Therefore, the highest point in the gradient amplitude accumulated value curve can be found out to obtain the largest gradient amplitude accumulated value, and the contour point 502 corresponding to the largest gradient amplitude accumulated value is the target amplitude accumulated point. And then searching for the first adjacent point at the left side of the target amplitude accumulation point. The first adjacent point 506 to the left that satisfies the condition is found by searching to the left of the target amplitude accumulation point. The first adjacent point to the right 504 is found to the right of the target amplitude accumulation point.

Therefore, in the embodiment of the application, by searching the peak point and the left and right adjacent points in the gradient amplitude accumulated value curve, more accurate target positioning and rich image characteristic information can be provided, so that the effects of image processing and analysis are improved, and the searching is started to the left side or the right side, and the searching is ended only by searching the first adjacent point. Therefore, no matter how complicated the gradient amplitude accumulated value curve is, the searching can be completed rapidly, the computing efficiency is improved, and the computer resource is saved to a great extent.

In some embodiments, determining the imaging quality of the image to be detected based on the target amplitude accumulation point and a point in the gradient amplitude accumulation value curve adjacent to the target amplitude accumulation point comprises:

acquiring a defect threshold curve drawn in advance in a preset coordinate system;

Drawing a distribution curve in a preset coordinate system according to the target amplitude accumulation points and points adjacent to the target amplitude accumulation points in the gradient amplitude accumulation value curve;

And determining the imaging quality of the image to be detected according to the relative position relation between the distribution curve and the defect threshold curve.

In some embodiments, the setting may be performed according to the actual application scenario and needs. Presetting a defect threshold curve according to the definition and contrast requirements of an image to be detected, wherein the defect threshold curve is used for judging whether the image to be detected has defects or not, and specifically comprises the following steps: and drawing a distribution curve in a preset coordinate system by using the obtained target amplitude accumulation points and points adjacent to the target amplitude accumulation points in the gradient amplitude accumulation value curve, wherein the target amplitude accumulation points and the points adjacent to the target amplitude accumulation points in the gradient amplitude accumulation value curve are all positioned on the distribution curve. And determining whether the image to be detected has defects according to the relative position relation of the preset defect threshold curve and the distribution curve in the preset coordinate system. For example, if the distribution curve is below the defect threshold curve, the image to be detected is determined to have a defect. Otherwise, the method has no defect. The setting can be specifically performed according to actual needs.

Therefore, in the embodiment of the application, the comprehensive, flexible and reliable imaging quality detection value can be obtained through the position distribution relation between the distribution curve and the defect threshold curve, the accuracy and the robustness of defect detection are improved, and in the whole detection process, complex function fitting and variance calculation are not needed, so that the detection efficiency and stability are greatly improved.

Referring to fig. 6, fig. 6 is a schematic diagram of an inner conductor in a printed circuit board. The inner conductor in the printed circuit board means the conductor section located inside the printed circuit board. Typically, multilayer printed circuit boards have inner conductor layers between top and bottom layers. The inner layer wire is realized by paving copper foil in the inner layer through hole or blind hole. For transmitting electrical signals or power signals between the different layers and may implement complex circuit connections. Inner conductors are typically used to transmit high frequency signals, ground, power lines, etc.

Referring to fig. 7, fig. 7 is a schematic illustration of vignetting effect. The vignetting effect refers to a phenomenon in which bright middle and dark sides appear in the edge region of an image. In the fields of photography and image processing, it is used to refer to the case where the brightness of the edges of an image is uneven due to gradual decrease of light or characteristics of an optical system.

It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.

Based on the same inventive concept, the embodiment of the application also provides an image quality detection device. The implementation of the solution provided by the device is similar to the implementation described in the above method, so the specific limitation of the embodiment of the image quality detection device or devices provided below may be referred to the limitation of the image quality detection method hereinabove, and will not be repeated here.

As shown in fig. 8, an embodiment of the present application provides an apparatus 800, including:

The detection module 802 is configured to perform contour detection on a target object in the acquired image to be detected, so as to obtain a contour curve of the target object;

A first determining module 804, configured to determine a gradient magnitude image of an image to be detected; the gradient amplitude image is an image representing the degree of variation between adjacent pixel points in the image to be detected;

a second determining module 806, configured to determine, for each contour point on the contour curve, a normal direction of the contour point;

A third determining module 808, configured to determine, according to the gradient magnitude image and the normal direction, a gradient magnitude maximum point corresponding to the targeted contour point;

a fourth determining module 810, configured to determine the imaging quality of the image to be detected according to the gradient magnitude maximum point corresponding to each contour point on the contour curve.

In some embodiments, in performing contour detection on a target object in an acquired image to be detected to obtain a contour curve of the target object, the detection module 802 is specifically configured to:

In some embodiments, in determining the binary image of the target object in the acquired image to be detected, the detection module 802 is specifically configured to:

Determining an acquired gray level image of an image to be detected;

In some embodiments, in determining the gradient magnitude image of the image to be detected, the first determining module 804 is specifically configured to:

In some embodiments, in determining the gradient amplitude value corresponding to each pixel point in the image to be detected according to the horizontal gradient amplitude value and the vertical gradient amplitude value corresponding to each pixel point in the image to be detected, the first determining module 804 is specifically configured to:

In some embodiments, the gradient magnitude image is a curved image of a plurality of gradient magnitude points; each gradient amplitude point is a point formed by the corresponding pixel point and the gradient amplitude of the corresponding pixel point in the image to be detected; in determining a gradient magnitude maximum point corresponding to the targeted contour point from the gradient magnitude image and the normal direction, the third determining module 808 is specifically configured to:

In some embodiments, in determining the imaging quality of the image to be detected according to the gradient magnitude maximum point corresponding to each contour point on the contour curve, the fourth determining module 810 is specifically configured to:

In some embodiments, in determining the imaging quality of the image to be detected according to the gradient magnitude accumulated value corresponding to each contour point on the contour curve, the fourth determining module 810 is specifically configured to:

In some embodiments, in determining the imaging quality of the image to be detected according to the target amplitude accumulation point and a point adjacent to the target amplitude accumulation point in the gradient amplitude accumulation value curve, the fourth determining module 810 is specifically configured to:

The respective modules in the above-described image quality detecting apparatus may be implemented in whole or in part by software, hardware, and a combination thereof. The above modules may be embedded in hardware or independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

In some embodiments, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 9. The computer device includes a processor, a memory, an Input/Output interface (I/O) and a communication interface. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface is connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used for storing data related to image quality detection. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement the steps in the image quality detection method described above.

In some embodiments, a computer device is provided, which may be a terminal, and the internal structure of which may be as shown in fig. 10. The computer device includes a processor, a memory, an input/output interface, a communication interface, a display unit, and an input device. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless mode can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement the steps in the image quality detection method described above. The display unit of the computer device is used for forming a visual picture, and can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen; the input device of the computer equipment can be a touch layer covered on a display screen, can also be keys, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.

It will be appreciated by those skilled in the art that the structures shown in fig. 9 or 10 are merely block diagrams of portions of structures associated with aspects of the application and are not intended to limit the computer device to which aspects of the application may be applied, and that a particular computer device may include more or fewer components than those shown, or may combine certain components, or may have a different arrangement of components.

In some embodiments, a computer device is provided, comprising a memory storing a computer program and a processor implementing the steps of the method embodiments described above when the computer program is executed.

In some embodiments, an internal structural diagram of a computer-readable storage medium is provided as shown in fig. 11, the computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method embodiments described above.

In some embodiments, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the steps of the method embodiments described above.

It should be noted that, the user information (including but not limited to user equipment information, user personal information, etc.) and the data (including but not limited to data for analysis, stored data, presented data, etc.) related to the present application are information and data authorized by the user or sufficiently authorized by each party, and the collection, use and processing of the related data need to comply with the related laws and regulations and standards of the related country and region.

Those skilled in the art will appreciate that implementing all or part of the above-described methods may be accomplished by way of a computer program, which may be stored on a non-transitory computer readable storage medium and which, when executed, may comprise the steps of the above-described embodiments of the methods. Any reference to memory, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magneto-resistive random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (PHASE CHANGE Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in various forms such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), etc. The databases referred to in the embodiments provided herein may include at least one of a relational database and a non-relational database. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processor referred to in the embodiments provided in the present application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, or the like, but is not limited thereto.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The foregoing examples illustrate only a few embodiments of the application, which are described in detail and are not to be construed as limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of the application should be assessed as that of the appended claims.

Claims

1. An image quality detection method, comprising:

performing contour detection on an obtained target object in an image to be detected to obtain a contour curve of the target object;

determining a gradient amplitude image of the image to be detected; the gradient amplitude image is an image representing the degree of variation between adjacent pixel points in the image to be detected;

And determining the imaging quality of the image to be detected according to the maximum gradient amplitude point corresponding to each contour point on the contour curve.

2. The method according to claim 1, wherein the performing contour detection on the target object in the acquired image to be detected to obtain a contour curve of the target object includes:

For each two initial contour points in the plurality of initial contour points, determining a line segment formed by taking the two initial contour points as endpoints, and deleting other initial contour points when the line segment has the other initial contour points;

3. The method according to claim 2, wherein determining the binary image of the target object in the acquired image to be detected comprises:

Determining an acquired gray level image of an image to be detected;

Inputting the gray level image of the image to be detected into a preset high-resolution model for processing, and outputting a class label corresponding to each pixel point in the image to be detected;

4. The method of claim 1, wherein said determining a gradient magnitude image of said image to be detected comprises:

respectively performing a second convolution operation on each pixel point in the image to be detected through a preset vertical direction operator template to obtain a vertical gradient amplitude value corresponding to each pixel point in the image to be detected;

Determining the gradient amplitude value corresponding to each pixel point in the image to be detected according to the horizontal gradient amplitude value and the vertical gradient amplitude value corresponding to each pixel point in the image to be detected;

5. The method of claim 4, wherein determining the respective gradient magnitude for each pixel in the image to be detected according to the respective horizontal gradient magnitude and the respective vertical gradient magnitude for each pixel in the image to be detected comprises:

square processing is carried out on the horizontal gradient amplitude value of each pixel point in the image to be detected, so as to obtain a first square value;

And taking root numbers for the third square values to obtain gradient amplitude values corresponding to the aimed pixel points.

6. The method of claim 1, wherein determining a gradient magnitude maximum point corresponding to the targeted contour point from the gradient magnitude image and the normal direction comprises:

7. The method according to claim 1, wherein determining the imaging quality of the image to be detected according to the gradient magnitude maximum point corresponding to each contour point on the contour curve comprises:

searching the adjacent points of each gradient amplitude maximum point in the gradient amplitude image;

According to the gradient amplitude of each adjacent point of the maximum gradient amplitude point, determining the gradient amplitude accumulated value corresponding to each contour point on the contour curve;

8. The method of claim 7, wherein determining the imaging quality of the image to be detected based on the respective gradient magnitude accumulation value for each contour point on the contour curve comprises:

9. The method of claim 8, wherein determining the imaging quality of the image to be detected based on the target amplitude accumulation point and a point in the gradient amplitude accumulation value curve adjacent to the target amplitude accumulation point comprises:

drawing a distribution curve in the preset coordinate system according to the target amplitude accumulation point and a point adjacent to the target amplitude accumulation point in the gradient amplitude accumulation value curve;

10. An image quality detecting apparatus, comprising:

a second determining module, configured to determine, for each contour point on the contour curve, a normal direction of the contour point;

and a fourth determining module, configured to determine an imaging quality of the image to be detected according to a gradient amplitude maximum point corresponding to each contour point on the contour curve.

11. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 9 when the computer program is executed.

12. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 9.