CN116452580B - Notebook appearance quality detection method - Google Patents
Notebook appearance quality detection method Download PDFInfo
- Publication number
- CN116452580B CN116452580B CN202310691466.XA CN202310691466A CN116452580B CN 116452580 B CN116452580 B CN 116452580B CN 202310691466 A CN202310691466 A CN 202310691466A CN 116452580 B CN116452580 B CN 116452580B
- Authority
- CN
- China
- Prior art keywords
- key
- key position
- template
- area
- point
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title abstract description 14
- 230000007547 defect Effects 0.000 claims abstract description 60
- 238000000034 method Methods 0.000 claims description 21
- 238000010586 diagram Methods 0.000 claims description 8
- 230000000739 chaotic effect Effects 0.000 claims description 5
- 238000003708 edge detection Methods 0.000 claims description 3
- 239000004973 liquid crystal related substance Substances 0.000 description 14
- 238000004519 manufacturing process Methods 0.000 description 7
- 238000004364 calculation method Methods 0.000 description 5
- 230000002950 deficient Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/181—Segmentation; Edge detection involving edge growing; involving edge linking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
Abstract
The invention relates to the technical field of image processing, in particular to a notebook appearance quality detection method, which comprises the following steps: dividing the notebook computer keyboard image and the template image to obtain all key position areas in the notebook computer keyboard image and all template keys in the template image, obtaining the edge matching rate of the key position areas and the corresponding template keys according to the edge pixel points in the key position areas and the template keys, further obtaining the searching step length of the key position areas, searching each key position area according to the searching step length to obtain all mark pixel points in each key position area, and obtaining the defect probability of each key position area according to the connected areas formed by the mark pixel points in the key position areas and the edge matching rate, thereby obtaining the defect area. According to the invention, each key position area is searched in combination with the search step length, so that the calculated amount of template matching is reduced, and meanwhile, the appearance quality detection of the notebook is more accurate.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a method for detecting appearance quality of a notebook.
Background
Along with the continuous updating and iteration of electronic products, the automation degree in the production process of notebook computers is higher and higher, and in the early assembly production, along with the improvement of the production efficiency, the quality problem of finished products is unavoidable. The traditional manual detection method is low in detection speed and high in labor pressure, and the detection result is influenced by experience and proficiency of the inspector and some supervisor factors, so that consistency and reliability are lacked.
The keyboard is used as a more important component of the notebook computer, and the consumer is prompted to intuitively feel the computer, but in the production process, the lug characters in the keyboard frame are broken possibly due to errors of staff or machine faults, and at the moment, the whole keyboard image needs to be detected by taking a standard single keyboard as a template image.
The template matching algorithm can be used for realizing integrity detection, and a defect area can be locked according to the difference of the template and the detected image, but the calculation amount caused by the characteristic of pixel-by-pixel search of the algorithm is large, the requirement on the calculation performance of a system is strict, and the efficiency is low.
Disclosure of Invention
The invention provides a method for detecting the appearance quality of a notebook computer, which aims to solve the existing problems.
The invention discloses a notebook appearance quality detection method which adopts the following technical scheme:
one embodiment of the invention provides a method for detecting the appearance quality of a notebook, which comprises the following steps:
collecting a keyboard image and a template image of a notebook computer; dividing a keyboard image and a template image of the notebook computer to obtain all key position areas in the keyboard image of the notebook computer and all template key positions in the template image;
respectively carrying out edge detection on the notebook computer keyboard image and the template image to obtain edge pixel points in each key position area of the notebook computer keyboard image and edge pixel points in each template key position of the template image;
acquiring the edge matching rate of each key position area and the corresponding template key position according to the edge pixel points; acquiring the searching step length of each key region according to the edge matching rate of each key region and the corresponding template key; searching each key position area according to the searching step length to obtain all marked pixel points in each key position area;
acquiring all connected domains formed by all marked pixel points in each key position area; obtaining the chaotic degree of all the connected domains in each key position area; obtaining the defect probability of each key region according to the chaotic degree of all the connected regions in each key region and the edge matching rate of each key region and the corresponding template key;
and acquiring a defect area according to the defect probability.
Preferably, the splitting the keyboard image and the template image of the notebook computer to obtain all key position areas in the keyboard image of the notebook computer and all template key positions in the template image comprises the following specific steps:
dividing a keyboard image of the notebook computer according to the form, the size and the position of each key position in a keyboard design diagram of the notebook computer to obtain all key position areas in the keyboard image of the notebook computer;
dividing the template image according to the form, size and position of each key position in the keyboard design diagram of the notebook computer to obtain all key position areas in the template image, and marking the key position areas as template keys.
Preferably, the step of obtaining the edge matching rate between each key region and the corresponding template key according to the edge pixel point includes the following specific steps:
wherein, the liquid crystal display device comprises a liquid crystal display device,is the firstThe edge matching rate of the key position areas and the corresponding template key positions;is the firstThe number of edge pixel points contained in the key position areas;is the firstThe first key position areaThe projection points of the edge pixel points in the corresponding template key positions and the Euclidean distance between the edge pixel points closest to the projection points in the template key positions;as a hyperbolic tangent function;is a super parameter.
Preferably, the step of searching each key region is obtained according to the edge matching rate of each key region and the corresponding template key, and the steps include:
wherein, the liquid crystal display device comprises a liquid crystal display device,represent the firstSearching step sizes of the key position areas;is the firstThe edge matching rate of the key position areas and the corresponding template key positions;is a tangent function;to round the symbol up.
Preferably, the step of searching each key area according to the search step length, to obtain all the marked pixel points in each key area includes the following specific steps:
s1: taking the central pixel point of the key position area as an initial searching point;
s2: taking the searching step length of the key position area as a radius, and acquiring a pixel point with the Euclidean distance from the searching point as the searching step length in the cross direction of the searching point as a target pixel point, wherein the target pixel point cannot be the searching point;
acquiring an absolute value of a difference value of gray values of each target pixel point and the pixel points at the same position in the corresponding template key position, and taking the absolute value as the gray difference of each target pixel point;
marking the target pixel points with the gray differences larger than the gray threshold value to obtain marked pixel points; if the search point corresponding to the marked pixel point is also the marked pixel point, marking all the pixel points from the marked pixel point to the corresponding search point;
acquiring the average value of the gray differences of all the target pixel points, taking the target pixel point with the maximum gray difference and larger than the average value as a new search point, wherein the new search point is not taken as the target pixel point any more;
s3: and (2) repeating the step (S2) until the obtained new target pixel point exceeds the range of the key position area, or stopping iteration when no new search point exists, so as to obtain all the marked pixel points in the key position area.
Preferably, the obtaining the confusion degree of all the connected domains in each key position area includes the following specific steps:
wherein, the liquid crystal display device comprises a liquid crystal display device,represent the firstThe degree of confusion of all connected domains in the key position areas;is the firstThe first key position areaThe areas of the communicating areas;is the firstThe number of connected domains in each key region.
Preferably, the obtaining the defect probability of each key region includes the following specific steps:
wherein the method comprises the steps ofIs the firstProbability of defect for the individual bond regions;is the firstThe degree of confusion of all connected domains in the key position areas;is the firstThe first key position areaThe areas of the communicating areas;is the firstThe area of the individual key regions;is the firstThe edge matching rate of the key position areas and the corresponding template key positions;is the firstThe number of connected domains in the key position areas;is an exponential function with a base of natural constant.
Preferably, the step of obtaining the defect area according to the defect probability includes the following specific steps:
when the defect probability of the key region is larger than the defect threshold value, acquiring the average value of the areas of all connected domains in the key region, and taking the connected domain with the area larger than the average value as the defect region.
The technical scheme of the invention has the beneficial effects that: according to the invention, the notebook computer keyboard image and the template image are segmented to obtain all key position areas in the notebook computer keyboard image and all template keys in the template image, the edge matching rate of the key position areas and the corresponding template keys is obtained according to the edge pixel points in the key position areas and the template keys, the searching step length of the key position areas is obtained according to the edge matching rate, each key position area is searched according to the searching step length, all mark pixel points in each key position area are obtained, and compared with the traditional template matching method, the traversal matching is carried out on all pixel points. According to the invention, the defect probability of each key region is obtained according to the connected domain formed by the marked pixel points in the key region and the edge matching rate, so that the defect region is obtained. The invention reduces the calculated amount of template matching, improves the calculation efficiency, obtains more accurate defect areas and detects the appearance quality of the notebook computer more accurately.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart showing the steps of a method for detecting the appearance quality of a notebook according to the present invention;
FIG. 2 is a diagram of a notebook computer keyboard image;
fig. 3 is a schematic cross-direction view.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following is a specific implementation, structure, characteristics and effects of a notebook computer appearance quality detection method according to the invention with reference to the attached drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the method for detecting the appearance quality of the notebook computer provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart illustrating steps of a method for detecting appearance quality of a notebook according to an embodiment of the invention is shown, the method includes the following steps:
s001, acquiring a keyboard image and a template image of the notebook computer, and dividing the keyboard image and the template image of the notebook computer.
Arranging an industrial camera on a notebook factory detection production line, placing the industrial camera right above the production line, overlooking and shooting an internal image of a notebook keyboard, carrying out graying treatment on the internal image of the notebook keyboard for facilitating subsequent treatment, and recording the obtained gray scale map as a notebook keyboard image. An image of a keyboard of a notebook computer according to an embodiment of the present invention is shown in fig. 2.
It should be noted that, the keyboard of the notebook computer of each brand has a corresponding design diagram during production, and the key positions on the keyboard have corresponding positions, so that the keyboard image of the notebook computer can be divided to obtain all key position areas on the keyboard for facilitating subsequent detection.
In the embodiment of the invention, the keyboard image of the notebook computer is segmented according to the form, the size and the position of each key position in the keyboard design diagram of the notebook computer, so as to obtain all key position areas in the keyboard image of the notebook computer.
It should be noted that, if the tab character breaking defect exists in the key position, the edge of the key position may deviate, and the gray oval-circled area in fig. 2 is the tab character breaking defect. Therefore, the defects of the key positions in the keyboard image of the current notebook computer can be identified according to the keyboard image of the defect-free notebook computer.
In the embodiment of the invention, an industrial camera is utilized to shoot an image of a manually determined defect-free notebook computer keyboard as a template image. Dividing the template image according to the form, size and position of each key position in the keyboard design diagram of the notebook computer to obtain all key position areas in the template image, and marking the key position areas as template keys.
And respectively carrying out edge detection on the notebook computer keyboard image and the template image by using a canny operator to obtain edge pixel points in each key position area of the notebook computer keyboard image and edge pixel points in each template key position of the template image.
Thus, the key position area of the notebook computer keyboard image and the edge pixel points in the key position area are obtained, and the template key position in the template image and the edge pixel points in the template key position are obtained.
S002, obtaining the edge matching rate of each key position area and the template key position.
It should be noted that, the broken defects of the lug characters at the key positions are the changes of the internal structure of the keyboard, which can cause the deviation of the positions of the extracted edge pixel points in the key position areas, so that the edge matching rate of the key position areas and the corresponding template key positions can be quantified according to the position difference of the edge pixel points in the key position areas of the notebook computer keyboard image and the corresponding template key positions, the smaller the edge matching rate is, the more likely the defects are in the key position areas, and the accurate positions of the defects in the key position areas need to be acquired in a more accurate mode later.
In the embodiment of the invention, a rectangular coordinate system of the notebook computer keyboard image and a rectangular coordinate system of the template image are respectively constructed by taking the first pixel point at the left upper corner in the notebook computer keyboard image and the template image as the origin of coordinates, taking the horizontal right direction as the horizontal axis and taking the vertical downward direction as the vertical axis.
Projecting all edge pixel points in each key position area into a corresponding template key position, and acquiring the edge matching rate of each key position area and the corresponding template key position:
wherein, the liquid crystal display device comprises a liquid crystal display device,is the firstThe edge matching rate of the key position areas and the corresponding template key positions;is the firstThe number of edge pixel points contained in the key position areas;is the firstThe first key position areaThe Euclidean distance between the projection point of each edge pixel point in the corresponding template key position and the edge pixel point closest to the projection point in the template key position is used for measuring the first distanceThe first key position areaThe degree of positional deviation of the individual edge pixel points;as a hyperbolic tangent function;an empirical value of 0.5 for the purpose of preventingToo large, leading toConstant approaching 1, in other embodiments the practitioner may set the value of the hyper-parameter according to the actual implementation;the larger the firstThe first key position areaThe greater the degree of deviation of the individual edge pixels;the larger the firstThe greater the overall degree of deviation of all edge pixel points in the key position area, the firstThe greater the likelihood of lug character breakage in the key location area, at this time the firstThe smaller the edge matching rate of the key position areas and the corresponding template key positions is; conversely, whenThe smaller the firstThe smaller the overall deviation degree of all edge pixel points in the key position area is, the firstThe less likely that the tab character will break in the key location areas, at this pointThe greater the edge match rate of the individual key regions to the corresponding template key.
So far, the edge matching rate of each key position area and the corresponding template key position is obtained.
S003, obtaining the marked pixel points of each key position area.
It should be noted that, in order to obtain the accurate position of the defect, a search may be performed for each key region. The probability of defects in the key areas with high edge matching rate is low, the search step length can be appropriately increased to reduce the calculated amount during searching of the key areas, the probability of defects in the key areas with low edge matching rate is high, and more accurate searching is needed to determine the specific positions of the defects in the key areas with low edge matching rate, so that the search step length needs to be set to be smaller. Therefore, different search steps are required to be set for each key region in combination with the edge matching rate of the key region, and each key region is searched according to the search steps. The cross search method can improve the calculation efficiency, and the search step length of each key position area can be applied to the cross search method to obtain all pixel points which are possibly defective and mark the pixel points.
In the embodiment of the invention, the search step length of each key region is obtained according to the edge matching rate of each key region and the corresponding template key:
wherein, the liquid crystal display device comprises a liquid crystal display device,represent the firstSearching step sizes of the key position areas;is the firstThe edge matching rate of the key position areas and the corresponding template key positions;is the circumference ratio;is a super parameter;is a tangent function;rounding up the symbol; when the first isThe greater the edge matching rate of the key position areas and the corresponding template key positionsAt the time of the firstThe larger the search step of the key region, whenThe smaller the edge matching rate of the key position areas and the corresponding template key positions is, the firstThe smaller the search step size of the individual key regions.
After obtaining the searching step length of each key position area, the searching step length is applied to the cross searching method, and the embodiment of the invention uses the first stepThe following are examples of key regions:
1. first by the firstThe central pixel point of each key position area is used as an initial searching point;
2. in the first placeThe searching step length of each key position area is a radius, the pixel points with the Euclidean distance from the searching point to the searching point in the cross direction of the searching point as the searching step length are obtained, and the pixel points are taken as target pixel points. The cross direction schematic is shown in fig. 3. When a pixel point having a distance from the search point to the search step in the cross direction of the search point is a history search point, the history search point is no longer used as the target pixel point. Obtaining gray scale difference of each target pixel point:
wherein, the liquid crystal display device comprises a liquid crystal display device,is the firstThe first key regionThe gray level difference of each target pixel point, namelyThe first key regionTarget pixel point and the firstGray scale differences of pixel points at the same position in the template key positions corresponding to the key position areas;is the firstThe first key regionGray values of the target pixel points;is the firstThe first key regionThe target pixel point is at the firstGray values of pixel points at the same position in the template key positions corresponding to the key position areas;is an absolute value sign.
A gray threshold P is preset, where the embodiment is described by taking p=8 as an example, and the embodiment is not specifically limited, where P may be determined according to the specific implementation situation.
And marking the target pixel point with the gray level difference larger than the gray level threshold value P to obtain a marked pixel point. And if the search point corresponding to the marked pixel point is also the marked pixel point, marking all the pixel points from the marked pixel point to the corresponding search point.
And acquiring the average value of the gray differences of all the target pixel points, taking the target pixel point with the maximum gray difference and larger than the average value as a new search point, and taking the search point not as the target pixel point any more.
3. Repeating step 2 until the new target pixel point exceeds the firstThe iteration is stopped for a range of key regions, or when no new search points exist.
Thus far, get the firstAll of the marked pixels of the key region.
And similarly, obtaining the marked pixel point of each key position area.
S004, obtaining the defect probability of each key position area.
It should be noted that, the difference between the gray value of the marked pixel point and the gray value of the pixel point corresponding to the marked pixel point in the template key position is large, which indicates that the marked pixel point may be the pixel point corresponding to the defect. Because the keyboard is defective and the distribution of the connected domains generated by the marked pixel points in the key position areas is complex and discrete, binarization processing can be performed on the marked pixel points, and the probability of defects in each key position area is obtained according to the connected domains of the binarized marked pixel points. In the marking process, the noise point can cause deviation of the result, so that the probability of defects in each key position area is determined by combining the information quantity and the information confusion degree contained in a single connected domain.
In the embodiment of the invention, the gray value of all the marked pixel points of each key area is set to be 1, the gray values of the rest pixel points are set to be 0, a binary image of each key area is obtained, and the binary image of each key area is subjected to connected domain analysis to obtain all the connected domains formed by all the marked pixel points.
Obtaining the confusion degree of all connected domains in each key position area:
wherein, the liquid crystal display device comprises a liquid crystal display device,represent the firstThe degree of confusion of all connected domains in the key position areas;is the firstThe first key position areaThe areas of the communicating areas;is the firstThe number of connected domains in the key position areas; if at firstWhen the areas of the connected domains contained in the individual regions are relatively uniform, each connected domain is a noise point with a large probability, but not a region with a high gray level difference of a real mark, at the momentThe greater the degree of confusion of all connected domains in each key region, whereas when a defect exists in each key region, the connected domains corresponding to the defect are larger, and the connected domains corresponding to the rest noise points are smaller, at this time, the firstThe less the degree of confusion is for all connected domains in each bond region.
Obtaining the defect probability of each key region according to the chaotic degree of all connected domains in each key region:
wherein the method comprises the steps ofIs the firstProbability of defect for the individual bond regions;is the firstThe degree of confusion of all connected domains in the key position areas;is the firstThe first key position areaThe areas of the communicating areas;is the firstThe area of the individual key regions;is the firstThe edge matching rate of the key position areas and the corresponding template key positions;is the firstThe number of connected domains in the key position areas;is an exponential function with a natural constant as a base; when the first isThe area of all the connected domains in the key region is at the firstThe larger the duty ratio in the key position region, the description is thatThe more the number of pixels with larger difference in the key position areas, the more the influence of noise points is weakened by the firstThe degree of confusion of all connected domains in the key region is relative to the firstThe area of all the connected domains in the key region is at the firstThe duty ratio in the key position area is adjusted to weaken the influence of noise points and is combined with the firstThe edge matching rate of the key position areas and the corresponding template key positions enables the result of the defect probability to be more accurate.
So far, the defect probability of each key region is obtained.
S005, obtaining a defect area, and detecting the appearance quality of the notebook.
A defect threshold Q is preset, where the embodiment is described by taking q=0.5 as an example, and the embodiment is not specifically limited, where Q may be determined according to the specific implementation situation. When the defect probability of the key region is larger than the defect threshold value Q, the key region is considered to have defects, at the moment, the average value of the areas of all connected domains in the key region is obtained, and the connected domains with the areas larger than the average value are taken as defect regions. When the probability of defect of the bond region is less than the defect threshold Q, the bond region is considered to be defect-free.
Thus, the appearance quality detection of the notebook is finished, an accurate defect area is obtained, the defect area is marked, and relevant staff repair the defect area.
Through the steps, the detection of the appearance quality of the notebook is completed.
According to the embodiment of the invention, all key position areas in the notebook computer keyboard image and all template keys in the template image are obtained by dividing the notebook computer keyboard image and the template image, the edge matching rate of the key position areas and the corresponding template keys is obtained according to the edge pixel points in the key position areas and the template keys, the searching step length of the key position areas is obtained according to the edge matching rate, each key position area is searched according to the searching step length, all mark pixel points in each key position area are obtained, and compared with the traditional template matching method, the traversing matching is carried out on all pixel points. According to the invention, the defect probability of each key region is obtained according to the connected domain formed by the marked pixel points in the key region and the edge matching rate, so that the defect region is obtained. The invention reduces the calculated amount of template matching, improves the calculation efficiency, obtains more accurate defect areas and detects the appearance quality of the notebook computer more accurately.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (5)
1. The method for detecting the appearance quality of the notebook is characterized by comprising the following steps of:
collecting a keyboard image and a template image of a notebook computer; dividing a keyboard image and a template image of the notebook computer to obtain all key position areas in the keyboard image of the notebook computer and all template key positions in the template image;
respectively carrying out edge detection on the notebook computer keyboard image and the template image to obtain edge pixel points in each key position area of the notebook computer keyboard image and edge pixel points in each template key position of the template image;
acquiring the edge matching rate of each key position area and the corresponding template key position according to the edge pixel points; acquiring the searching step length of each key region according to the edge matching rate of each key region and the corresponding template key; searching each key position area according to the searching step length to obtain all marked pixel points in each key position area;
acquiring all connected domains formed by all marked pixel points in each key position area; obtaining the chaotic degree of all the connected domains in each key position area; obtaining the defect probability of each key region according to the chaotic degree of all the connected regions in each key region and the edge matching rate of each key region and the corresponding template key;
obtaining a defect area according to the defect probability;
the method comprises the following specific steps of:
wherein beta is m The edge matching rate of the m-th key position area and the corresponding template key position is obtained; n (N) m The number of the edge pixel points contained in the m key position area; d, d m,n The Euclidean distance between a projection point of an nth edge pixel point in an mth key position area in a corresponding template key position and an edge pixel point closest to the projection point in the template key position; tan () is a hyperbolic tangent function; alpha is a super parameter;
the obtaining the confusion degree of all the connected domains in each key position area comprises the following specific steps:
wherein h is m Indicating the degree of confusion of all connected domains in the mth key region; s is(s) m,t The area of the t-th communication domain in the m-th key region; t (T) m The number of the connected domains in the m-th key position area;
the obtaining the defect probability of each key area comprises the following specific steps:
wherein H is m The probability of defect for the mth bond site region; h is a m The degree of confusion of all connected domains in the mth key position area; s is(s) m,t The area of the t-th communication domain in the m-th key region; s is S m Is the area of the mth bond site region; beta m The edge matching rate of the m-th key position area and the corresponding template key position is obtained; t (T) m The number of the connected domains in the m-th key position area; exp () is an exponential function based on a natural constant.
2. The method for detecting the appearance quality of a notebook computer according to claim 1, wherein the steps of dividing the keyboard image and the template image of the notebook computer to obtain all key areas in the keyboard image of the notebook computer and all template keys in the template image comprise the following specific steps:
dividing a keyboard image of the notebook computer according to the form, the size and the position of each key position in a keyboard design diagram of the notebook computer to obtain all key position areas in the keyboard image of the notebook computer;
dividing the template image according to the form, size and position of each key position in the keyboard design diagram of the notebook computer to obtain all key position areas in the template image, and marking the key position areas as template keys.
3. The method for detecting the appearance quality of a notebook according to claim 1, wherein the step of obtaining the search step of each key region according to the edge matching rate of each key region and the corresponding template key comprises the following specific steps:
wherein f m A search step representing an mth key region; beta m The edge matching rate of the m-th key position area and the corresponding template key position is obtained; tan () is a tangent function;to round the symbol up.
4. The method for detecting the appearance quality of a notebook computer according to claim 1, wherein the step of searching each key region according to a search step length to obtain all the marked pixel points in each key region comprises the following specific steps:
s1: taking the central pixel point of the key position area as an initial searching point;
s2: taking the searching step length of the key position area as a radius, and acquiring a pixel point with the Euclidean distance from the searching point as the searching step length in the cross direction of the searching point as a target pixel point, wherein the target pixel point cannot be the searching point;
acquiring an absolute value of a difference value of gray values of each target pixel point and the pixel points at the same position in the corresponding template key position, and taking the absolute value as the gray difference of each target pixel point;
marking the target pixel points with the gray differences larger than the gray threshold value to obtain marked pixel points; if the search point corresponding to the marked pixel point is also the marked pixel point, marking all the pixel points from the marked pixel point to the corresponding search point;
acquiring the average value of the gray differences of all the target pixel points, taking the target pixel point with the maximum gray difference and larger than the average value as a new search point, wherein the new search point is not taken as the target pixel point any more;
s3: and (2) repeating the step (S2) until the obtained new target pixel point exceeds the range of the key position area, or stopping iteration when no new search point exists, so as to obtain all the marked pixel points in the key position area.
5. The method for detecting the appearance quality of a notebook according to claim 1, wherein the step of obtaining the defect area according to the probability of the defect comprises the following specific steps:
when the defect probability of the key region is larger than the defect threshold value, acquiring the average value of the areas of all connected domains in the key region, and taking the connected domain with the area larger than the average value as the defect region.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310691466.XA CN116452580B (en) | 2023-06-13 | 2023-06-13 | Notebook appearance quality detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310691466.XA CN116452580B (en) | 2023-06-13 | 2023-06-13 | Notebook appearance quality detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116452580A CN116452580A (en) | 2023-07-18 |
| CN116452580B true CN116452580B (en) | 2023-09-01 |
Family
ID=87132340
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310691466.XA Active CN116452580B (en) | 2023-06-13 | 2023-06-13 | Notebook appearance quality detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116452580B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116823811B (en) * | 2023-08-25 | 2023-12-01 | 汶上县誉诚制衣有限公司 | Functional jacket surface quality detection method |
| CN117197144B (en) * | 2023-11-08 | 2024-02-23 | 深圳市极摩客科技有限公司 | Host shell quality detection method based on artificial intelligence |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2001243236A (en) * | 2000-02-28 | 2001-09-07 | Nippon Telegr & Teleph Corp <Ntt> | Video feature extracting method, video feature extracting device, video retrieving method, video retrieving device and recording medium in which its program is recorded |
| CN103077534A (en) * | 2012-12-31 | 2013-05-01 | 南京华图信息技术有限公司 | Space-time multi-scale moving target detection method |
| CN106485255A (en) * | 2016-09-29 | 2017-03-08 | 深圳元启智能技术有限公司 | A kind of DM code positioning and the method and system of identification |
| KR101812953B1 (en) * | 2017-06-23 | 2017-12-29 | 주식회사 디앤에스 테크놀로지 | Image detecting system and method for underbody of vehicle using recognizing a car type by identifying emblem image |
| WO2021012735A1 (en) * | 2019-07-25 | 2021-01-28 | 研祥智能科技股份有限公司 | Method and system for detecting screen display defects |
| CN112884746A (en) * | 2021-02-22 | 2021-06-01 | 青岛星科瑞升信息科技有限公司 | Character defect intelligent detection algorithm based on edge shape matching |
| CN113379673A (en) * | 2021-04-28 | 2021-09-10 | 合肥联宝信息技术有限公司 | Method, device and equipment for detecting small red cap of keyboard |
| WO2022027949A1 (en) * | 2020-08-04 | 2022-02-10 | 湖南大学 | Machine vision-based detecting method and system for glass bottle bottom defects |
| CN115273088A (en) * | 2022-09-30 | 2022-11-01 | 南通慕派商贸有限公司 | Chinese character printing quality detection method based on machine vision |
| CN115330780A (en) * | 2022-10-13 | 2022-11-11 | 山东盛世恒机械制造有限公司 | Rapid detection method for slag inclusion defect of metal welding |
| CN115351598A (en) * | 2022-10-17 | 2022-11-18 | 南通钜德智能科技有限公司 | Numerical control machine tool bearing detection method |
| CN115376007A (en) * | 2022-08-10 | 2022-11-22 | 北方工业大学 | Object detection method, device, equipment, medium and computer program product |
| CN115457041A (en) * | 2022-11-14 | 2022-12-09 | 安徽乾劲企业管理有限公司 | Road quality identification and detection method |
| CN116012357A (en) * | 2023-02-09 | 2023-04-25 | 河南职业技术学院 | Automobile circular trailer hole air line defect detection method based on machine vision |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9400936B2 (en) * | 2014-12-11 | 2016-07-26 | Xerox Corporation | Methods and systems for vehicle tag number recognition |
| KR102406975B1 (en) * | 2015-05-29 | 2022-06-13 | 엘지디스플레이 주식회사 | Panel defect detection method and organic light emitting display device |
-
2023
- 2023-06-13 CN CN202310691466.XA patent/CN116452580B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2001243236A (en) * | 2000-02-28 | 2001-09-07 | Nippon Telegr & Teleph Corp <Ntt> | Video feature extracting method, video feature extracting device, video retrieving method, video retrieving device and recording medium in which its program is recorded |
| CN103077534A (en) * | 2012-12-31 | 2013-05-01 | 南京华图信息技术有限公司 | Space-time multi-scale moving target detection method |
| CN106485255A (en) * | 2016-09-29 | 2017-03-08 | 深圳元启智能技术有限公司 | A kind of DM code positioning and the method and system of identification |
| KR101812953B1 (en) * | 2017-06-23 | 2017-12-29 | 주식회사 디앤에스 테크놀로지 | Image detecting system and method for underbody of vehicle using recognizing a car type by identifying emblem image |
| WO2021012735A1 (en) * | 2019-07-25 | 2021-01-28 | 研祥智能科技股份有限公司 | Method and system for detecting screen display defects |
| WO2022027949A1 (en) * | 2020-08-04 | 2022-02-10 | 湖南大学 | Machine vision-based detecting method and system for glass bottle bottom defects |
| CN112884746A (en) * | 2021-02-22 | 2021-06-01 | 青岛星科瑞升信息科技有限公司 | Character defect intelligent detection algorithm based on edge shape matching |
| CN113379673A (en) * | 2021-04-28 | 2021-09-10 | 合肥联宝信息技术有限公司 | Method, device and equipment for detecting small red cap of keyboard |
| CN115376007A (en) * | 2022-08-10 | 2022-11-22 | 北方工业大学 | Object detection method, device, equipment, medium and computer program product |
| CN115273088A (en) * | 2022-09-30 | 2022-11-01 | 南通慕派商贸有限公司 | Chinese character printing quality detection method based on machine vision |
| CN115330780A (en) * | 2022-10-13 | 2022-11-11 | 山东盛世恒机械制造有限公司 | Rapid detection method for slag inclusion defect of metal welding |
| CN115351598A (en) * | 2022-10-17 | 2022-11-18 | 南通钜德智能科技有限公司 | Numerical control machine tool bearing detection method |
| CN115457041A (en) * | 2022-11-14 | 2022-12-09 | 安徽乾劲企业管理有限公司 | Road quality identification and detection method |
| CN116012357A (en) * | 2023-02-09 | 2023-04-25 | 河南职业技术学院 | Automobile circular trailer hole air line defect detection method based on machine vision |
Non-Patent Citations (1)
| Title |
|---|
| 无标记印刷品质量在线检测方法研究;许敏;马钺;陈帅;徐首帅;;传感器与微系统(第10期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116452580A (en) | 2023-07-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116452580B (en) | Notebook appearance quality detection method | |
| CN112949564B (en) | Pointer type instrument automatic reading method based on deep learning | |
| CN110136116B (en) | Injection pump defect detection method, device, equipment and storage medium | |
| CN109767445B (en) | High-precision PCB defect intelligent detection method | |
| CN109886952B (en) | Screen defect point detection method and device and computer readable medium | |
| CN114549835B (en) | Pointer instrument correction identification method and device based on deep learning | |
| CN111046881B (en) | Pointer type instrument reading identification method based on computer vision and deep learning | |
| CN104050446A (en) | Meter pointer image identification method based on pointer width character | |
| CN111833303A (en) | Product detection method and device, electronic equipment and storage medium | |
| CN115063618B (en) | Defect positioning method, system, equipment and medium based on template matching | |
| CN103439348A (en) | Remote controller key defect detection method based on difference image method | |
| CN110751606B (en) | Foam image processing method and system based on neural network algorithm | |
| CN107392221B (en) | Training method of classification model, and method and device for classifying OCR (optical character recognition) results | |
| US20220076404A1 (en) | Defect management apparatus, method and non-transitory computer readable medium | |
| CN101739556A (en) | Method for automatically identifying number of steel billet | |
| CN110659637A (en) | Electric energy meter number and label automatic identification method combining deep neural network and SIFT features | |
| US20230214985A1 (en) | Product recognition method, model training method, device and electronic device | |
| CN112686264A (en) | Digital instrument reading method and device, computer equipment and storage medium | |
| CN106815830B (en) | Image defect detection method | |
| CN113705564B (en) | Pointer type instrument identification reading method | |
| CN114677348A (en) | IC chip defect detection method and system based on vision and storage medium | |
| CN109141457B (en) | Navigation evaluation method and device, computer equipment and storage medium | |
| CN113379673B (en) | Method, device and equipment for detecting small red cap of keyboard | |
| CN113160220A (en) | Door handle homing and bending detection method based on deep learning | |
| CN114219753A (en) | Power equipment surface defect detection method based on deep learning and terminal |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |