CN116977337B - Waterproof aluminum alloy door and window detection method based on machine vision - Google Patents
Waterproof aluminum alloy door and window detection method based on machine vision Download PDFInfo
- Publication number
- CN116977337B CN116977337B CN202311229235.3A CN202311229235A CN116977337B CN 116977337 B CN116977337 B CN 116977337B CN 202311229235 A CN202311229235 A CN 202311229235A CN 116977337 B CN116977337 B CN 116977337B
- Authority
- CN
- China
- Prior art keywords
- image
- pixel point
- difference
- window
- aluminum alloy
- 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
- 229910000838 Al alloy Inorganic materials 0.000 title claims abstract description 75
- 238000001514 detection method Methods 0.000 title claims abstract description 37
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 110
- 238000006073 displacement reaction Methods 0.000 claims abstract description 75
- 238000012360 testing method Methods 0.000 claims abstract description 75
- 238000000034 method Methods 0.000 claims description 42
- 238000005507 spraying Methods 0.000 claims description 11
- 230000014509 gene expression Effects 0.000 claims description 7
- 238000012937 correction Methods 0.000 claims description 6
- 239000000047 product Substances 0.000 description 9
- 238000011410 subtraction method Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 239000011521 glass Substances 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 239000005357 flat glass Substances 0.000 description 1
- 230000002706 hydrostatic effect Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000011056 performance test Methods 0.000 description 1
- 239000012466 permeate Substances 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 239000007921 spray Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000000007 visual effect 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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M3/00—Investigating fluid-tightness of structures
- G01M3/02—Investigating fluid-tightness of structures by using fluid or vacuum
- G01M3/04—Investigating fluid-tightness of structures by using fluid or vacuum by detecting the presence of fluid at the leakage point
-
- 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
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
Abstract
The invention relates to the field of machine vision, in particular to a machine vision-based waterproof aluminum alloy door and window detection method, which comprises the following steps: acquiring a test image sequence; constructing an initial difference image and a corrected difference image according to the difference of two adjacent test images; further calculating the displacement correlation of each pixel point to obtain a displacement correlation image, obtaining a corrected displacement correlation image, and obtaining the water drop change percentage and the difference coefficient of each pixel point according to the gray information of the corresponding pixel point of the initial difference image and the corrected difference image; acquiring a water seepage drop region image by combining the difference coefficient, the initial difference image, the corrected difference image and the corrected displacement correlation image, and acquiring a water resistance index according to the area of the water seepage drop region and the occurrence time of the water seepage drop region; and the waterproof detection of the waterproof aluminum alloy doors and windows is completed by combining the waterproof index. Therefore, the accurate detection of the waterproof performance of the waterproof aluminum alloy door and window is realized based on machine vision.
Description
Technical Field
The application relates to the field of machine vision, in particular to a waterproof aluminum alloy door and window detection method based on machine vision.
Background
The door and window with firmness and good water tightness is an effective method for resisting rain. The good watertight performance can effectively prevent outdoor rainwater from entering when the door and window face the attack of heavy rain, and the indoor environment is kept dry. If the outer window is poor in waterproof performance, even if the door and window are closed, rainwater can permeate into a room through the door and window by means of wind power, so that indoor moisture is increased, walls are wet, mold and corrosion are even caused, and living comfort of people is affected.
Therefore, the waterproof performance of the aluminum alloy doors and windows needs to be detected, and the waterproof performance is generally higher as the water tightness level is higher. At present, most of the watertight performance detection of the aluminum alloy doors and windows utilizes human eyes to capture leakage areas, and the problems of low efficiency, large influence of human factors and the like exist. The difference method used for detecting the general moving object is difficult to ensure that the corresponding pixel points of the subtracted images are positioned on the same target point in space, and if the moving object does fine movement, the problem of incomplete detection of the moving object can also occur.
In summary, the invention provides a machine vision-based waterproof aluminum alloy door and window detection method, which is used for acquiring image data before and after an aluminum alloy water tightness test, extracting an initial difference image in combination with analysis of the images before and after the water tightness test, correcting the initial difference image, acquiring displacement similarity of each region to obtain a displacement correlation matrix, constructing a difference coefficient, and acquiring a water seepage drop region according to the difference coefficient and the corrected displacement correlation matrix, thereby realizing accurate detection analysis of the waterproof aluminum alloy door and window.
Disclosure of Invention
In order to solve the technical problems, the invention provides a machine vision-based waterproof aluminum alloy door and window detection method to solve the existing problems.
The machine vision-based waterproof aluminum alloy door and window detection method provided by the invention adopts the following technical scheme:
the invention provides a machine vision-based waterproof aluminum alloy door and window detection method, which comprises the following steps of:
collecting multi-frame aluminum alloy door and window images before and after water spraying, and recording the multi-frame aluminum alloy door and window images as a test image sequence;
for any two adjacent test images in the test image sequence, constructing an initial difference image according to the difference of the two adjacent test images; obtaining a corrected differential image according to the gray scale difference of the adjacent test image pixel points; taking the displacement change of each pixel point in the front and rear adjacent images as the pixel displacement quantity of each pixel point, combining the initial difference image and the corrected difference image to correspond to the pixel displacement quantity change in the local neighborhood of each pixel point to obtain the displacement correlation degree of the pixel point at each position, forming a displacement correlation degree image by the displacement correlation degree of each pixel point, acquiring the corrected displacement correlation degree image according to the displacement correlation degree image, and obtaining the water drop change percentage of each pixel point according to the gray information in the local neighborhood of each pixel point corresponding to the initial difference image and the corrected difference image; obtaining a difference coefficient of the pixel point according to the water drop change percentage of the pixel point; acquiring a leakage water drop area image by combining the difference coefficient, the initial difference image, the corrected difference image and the corrected displacement correlation image; acquiring leakage water drop area images corresponding to any two adjacent test images in the test image sequence;
obtaining a waterproof index according to the area of the water seepage drop area and the occurrence time of the water seepage drop area in each water seepage drop area image; and the waterproof detection of the waterproof aluminum alloy doors and windows is completed by combining the waterproof index.
Further, the constructing an initial differential image according to the differences of the two adjacent test images includes:
and regarding two adjacent test images, taking the difference value of the gray value of each pixel point in the test image of the next frame and the gray value of each pixel point in the corresponding position in the test image of the previous frame as the gray value of each pixel point in the initial difference image, and obtaining the initial difference image.
Further, the corrected difference image is obtained according to the gray scale difference of the adjacent test image pixels, and the expression is:
in the method, in the process of the invention,representing test images、The corresponding corrected difference image is then displayed,represent the firstIn a test imageThe gray value of the pixel point is located,represent the firstIn a test imageThe gray value of the pixel point.
Further, the combining the initial difference image and the corrected difference image to correspond to the change of the pixel displacement in the pixel area window to obtain the displacement correlation degree of the pixel at each position, including:
for each pixel point of the corresponding position of the initial difference image and the corrected difference image;
calculating covariance of pixel displacement of a pixel point neighborhood window of the initial difference image and a pixel point neighborhood window of a corresponding position of the corrected difference image;
respectively calculating pixel displacement variance of a pixel point neighborhood window of the initial difference image and a pixel point neighborhood window of the corresponding position of the corrected difference image, and obtaining the product of the two variances;
and taking the ratio of the covariance to the product as the displacement correlation degree of the pixel point at the corresponding position.
Further, the obtaining the corrected displacement correlation image according to the displacement correlation image has the expression:
in the method, in the process of the invention,in order to correct the displacement correlation image,in order to displace the correlation image,is a correction parameter greater than 1.
Further, the drop change percentage of each pixel point includes:
and respectively calculating the sum value of gray values in the pixel point neighborhood window at the same position of the initial difference image and the corrected difference image, obtaining the absolute value of the difference value of the two sums and the maximum value of the two sums, and taking the ratio of the absolute value to the maximum value as the water drop change percentage of the pixel point.
Further, the obtaining the difference coefficient of the pixel according to the drop change percentage of the pixel includes:
when the water drop change percentage of the pixel point is smaller than or equal to the water drop change percentage threshold value, the difference coefficient of the pixel point is 0; when the water drop change percentage of the pixel point is larger than the water drop change percentage threshold value, the difference coefficient of the pixel point is 1.
Further, the step of obtaining the image of the leaking water drop area by combining the difference coefficient, the initial difference image, the corrected difference image and the corrected displacement correlation image includes:
and calculating the gray difference value of each pixel point in the corrected difference image and the initial difference image, calculating the product of the gray value of each pixel point in the corrected displacement related image and the difference coefficient, and taking the product of the gray difference value and the product of 1 minus the product as the gray value of each pixel point in the leaked water drop area image.
Further, the obtaining the waterproof index according to the area of the water seepage drop area and the occurrence time of the water seepage drop area in the image of each water seepage drop area comprises:
the method comprises the steps of obtaining the area of a leaking water drop area by combining all leaking water drop area images, obtaining a serial number corresponding to a test image which appears first in the leaking water drop area in a test image sequence, and obtaining an indoor and outdoor pressure difference absolute value of an aluminum alloy door window at the corresponding collection time of the test image of the serial number;
and calculating the product of the serial number and the area of the seepage water drop area, and taking the normalized value of the ratio of the absolute value of the pressure difference to the product as a waterproof index.
Further, the waterproof detection for completing the waterproof aluminum alloy doors and windows by combining the waterproof indexes comprises the following steps: when the waterproof index is larger than the waterproof index threshold, the aluminum alloy door and window is qualified in waterproof; otherwise, the aluminum alloy door and window is unqualified in waterproof.
The invention has at least the following beneficial effects:
according to the method, an initial differential image is built according to the difference of images before and after the water tightness test and the image subtraction method, a corrected differential image is built according to displacement to solve the problem of pixel point correspondence of the image subtraction method, the part which does not have similarity between two moving differential images is eliminated by utilizing the correlation of the two images, and the similar part is reserved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a machine vision-based waterproof aluminum alloy door and window detection method provided by the invention;
FIG. 2 is a current frame image provided by the present invention;
FIG. 3 is a view showing an adjacent next frame image of the current frame image provided by the present invention;
FIG. 4 is an initial difference image provided by the present invention;
FIG. 5 is a corrected difference image provided by the present invention;
fig. 6 is an image of a weeping drop zone provided by the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description is given below of the specific implementation, structure, characteristics and effects of the machine vision-based waterproof aluminum alloy door and window detection method according to the invention in combination with the accompanying 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 specific scheme of the machine vision-based waterproof aluminum alloy door and window detection method provided by the invention is specifically described below with reference to the accompanying drawings.
The invention provides a machine vision-based waterproof aluminum alloy door and window detection method.
Specifically, the following machine vision-based waterproof aluminum alloy door and window detection method is provided, please refer to fig. 1, and the method comprises the following steps:
and S001, collecting image data before and after the water tightness detection of the aluminum alloy doors and windows, and preprocessing to obtain a test image sequence.
The embodiment aims to detect the waterproofness of the aluminum alloy according to the analysis of front and rear images of the aluminum alloy doors and windows before and after the water tightness detection. And (3) carrying out watertight performance detection on the aluminum alloy door and window test piece by using a simulated static pressure box method, and under the action of a stable pressure difference, uniformly spraying water on the outdoor side of the aluminum alloy door and window, so as to determine the seepage capability of the aluminum alloy door and window test piece. It should be noted that, the simulation hydrostatic tank method and the detailed process are known techniques, and are not included in the protection scope of the present embodiment, and are not described in detail herein.
Then, fixing the camera position at the indoor side, collecting the watertight performance test video of the aluminum alloy door and window on the pressure box, wherein the obtained video stream data can be regarded as continuous multi-frame image data with short time intervals, and the frame rate of the video data can be kept to be 30FPS, namely, 30 images are shot per second as one video stream data. The specific image capturing frame rate, camera view angle, etc. can be set by the practitioner according to the actual situation, and the embodiment of the length of the captured video and the number of image frames is not limited. In this embodiment, the image of the aluminum alloy door and window collected before water spraying and the image of the multi-frame aluminum alloy door and window collected after water spraying are both recorded as test images, and each test image is preprocessed, where the preprocessing in this embodiment includes image denoising and image enhancement, and the specific preprocessing process is a known technology, which is not in the protection scope of this embodiment, and is not described in detail herein. And obtaining a test image sequence according to each preprocessed test image.
So far, the method can be used for obtaining the images of the aluminum alloy doors and windows before and after water tightness detection, and the test image sequence is obtained and used as the basic data of the detection of the waterproof aluminum alloy doors and windows.
Step S002: and analyzing the acquired test image sequence to detect the seepage area of the aluminum alloy door and window.
Because the water drenching device detects the outdoor side of aluminum alloy door and window, the camera of indoor side shoots the image when drenching through transparent door and window glass, when pressure boost and drenching to certain degree in the pressure tank, aluminum alloy door and window can produce the seepage phenomenon, appears in aluminum alloy door and window's indoor side surface can appear the drop of water.
Considering that water on an aluminum alloy door and window is continuously moving after water spraying, an aluminum alloy image shot before water spraying can be regarded as a background area. Because the test image sequence is obtained from video data, the aluminum alloy images before water spraying are also obtained under the same visual angle and frame rate, and therefore, the image characteristics of the aluminum alloy images before water spraying in the test image sequence are related to parameters such as the frame number, the position, the speed and the like of the aluminum alloy images after water spraying in the test sequence. Because the video is generated by real-time data, the different positions of the test image sequence have different degrees of variability. The differences and correlations were used to analyze the water drops leaking from the inside of the water spray area. Referring to fig. 2 and 3, fig. 2 is a current frame image, and fig. 3 is a next frame image adjacent to the current frame image. Referring to fig. 4, fig. 4 is an initial difference image obtained by performing a difference between a current frame image and an adjacent next frame image.
Constructing an initial differential image according to the difference of two adjacent images in a test image sequence, and acquiring the gray value of each pixel point of the initial differential image, wherein the gray value expression of each pixel point of the initial differential image is as follows:
in the method, in the process of the invention,representing the first of a sequence of test imagesSheet and the firstIn an initial differential image of a sheetThe gray value at the location is a function of the gray value at the location,、respectively represent the first in the test image sequenceStretch and firstSheet test image coordinatesGray values of pixels of (a).
Repeating the method in the embodiment, acquiring the gray value of each pixel point in the initial differential image, and acquiring the initial differential image.
The gray value corresponding to each pixel point in the initial difference image is the first gray value in the test image sequenceTo the firstThe gray level change value of the sheet, therefore, when the gray level is changed, it can be considered that the firstThe pixel point at the position on the test image is the pixel point of the motion area. When the leaked water drops appear in the indoor glass area of the aluminum alloy door and window, the moment of appearance can be regarded as a movement area, and after the moment of appearance, the movement displacement of the water drops on the inner side of the inner chamber is smaller than the movement displacement of the water flow on the outer side at the same time because the impact force of the water spraying is small. Since the initial difference image is difficult to ensure that the corresponding pixel points of the subtracted image are located on the same target point in space. So that it is corrected.
Because there is motion of an object or thing in the image, when the actual corresponding pixel positions of the frame difference method are subtracted, the subtraction is not necessarily performed on the same pixels corresponding to the object or thing, for example, a moving automobile, the positions of the automobiles in the first frame image and the second frame image are changed, and the direct subtraction is not the subtraction of the same automobile positions. Thus, the gradient is used for correction to improve the accuracy of the differential image. Referring to fig. 5, fig. 5 is a corrected difference image. The Sobel operator is used for obtaining the gradient direction and the gradient amplitude of each initial differential image pixel point, and the gradient amplitude of each pixel point is the pixel point from the first pointTo the firstThe displacement distance of the test image. Assuming that the displacement generated by the water drop movement pixel point isCorrecting the difference image:
in the method, in the process of the invention,representing test images、The corresponding corrected difference image is then displayed,represent the firstIn a test imageThe gray value of the pixel point is located,represent the firstIn a test imageThe gray value of the pixel point is located,is a pixel pointAnd generating the shifted pixel point positions. By this method a modified differential image sequence can be obtained from the test image sequence.
By studying the initial sequence of difference images, it was found that the position of the weeping drop in two adjacent frames in the sequence did not have a mutation. I.e. there is a continuity in space and time. If the first isIndividual test imagesAt the position ofWhere there is a water drop, the water drop will appear at the firstIndividual test imagesIn (a)In a small neighborhood of, i.eWhere it is located. The method is characterized by comprising the following steps of (a) preparing a product of the formula (a)) As a test imageIn (a)The pixel displacement of each pixel point in the two adjacent frames of images is used as the pixel displacement of each pixel point. Therefore, the position of the region where the water seepage drop is located can be judged through the correlation between the image regions.
Further, in this embodiment, the change condition of the displacement amount of the pixel point at each position is analyzed, for each pixel point, a neighborhood window is obtained with each pixel point as the center, and according to the pixel displacement change condition of each pixel point in the domain window, the displacement change correlation degree between the corresponding pixel points in the initial difference image and the corrected difference image is analyzed, so that the displacement correlation degree is added and constructed in this embodiment, and the displacement correlation degree of the pixel point at each position is obtained:
in the method, in the process of the invention,representing the displacement correlation of pixel points at (x, y) for characterizing the initial differential imageThe pixel point is corresponding to the correction difference imagePixel displacement correlation degree between pixel points; the size of the neighborhood window can be set by the practitioner, and in the embodiment, the neighborhood window is set as followsA square;representing the calculation of the covariance of the data,for the pixel neighborhood window at the initial difference image (x, y) and the corresponding position of the corrected difference imageCovariance of pixel displacement of pixel point neighborhood window;、respectively expressed by、The standard deviation of the pixel displacement in the neighborhood window with the pixel point as the center. It should be noted that, in the present embodiment, the following description is givenIs a pixel point at the same position in different images.
And finally obtaining a displacement correlation image with the same size as the original image according to the displacement correlation of each pixel point. To eliminate dissimilar parts in two images as much as possible and keep similar parts, in order to improve the contrast of the displacement correlation image and improve the significance of the pixel points with higher displacement correlation, in this embodiment, the displacement correlation of each pixel point in the displacement correlation image is optimized, and a corrected displacement correlation image is constructed to increase the image contrast, where the corrected displacement correlation image is:
in the method, in the process of the invention,in order to correct the displacement correlation image,in order to displace the correlation image,for correction parameters greater than 1, the practitioner can set the correction parameters to be at his own discretion, in order to improve the contrast of the corrected displacement correlation image。
The contrast of the corrected displacement correlation image becomes larger, which is favorable for eliminating the part with dissimilar water drop distribution between the two images and enhancing the part with similar water drop distribution. It should be noted that when the water drop profile of the indoor side of the aluminum alloy door and window overlaps with the water flow profile of the outdoor side of the aluminum alloy door and window, the correlation degree is weaker, which is unfavorable for retaining the similar part, so that the embodiment will construct a difference coefficient to improve the extraction precision of the leaked water drop region and avoid the problem of false detection of the leaked water drop pixel point in the glass region. In this embodiment, first, a drop change percentage is constructed for each pixel of the initial differential image by relatively correcting the change condition of each pixel of the differential image, and a differential coefficient of each pixel is obtained according to the drop change percentage. The expressions of the drop change percentages and the difference coefficients in this embodiment are respectively:
in the method, in the process of the invention,is the differential coefficient of the pixel point at (x, y),representing the percentage of drop change at the pixel point at (x, y),) Respectively representing the sum of gray values in the pixel point neighborhood window at the same position of the initial differential image and the corrected differential image,as a function of the maximum value,for the drop percentage change threshold, the practitioner can set himself, and in this example, the checked value is 0.5. When the drop change percentage of the overlapping region of the moving object is equal to or smaller than the drop change percentage threshold, the difference coefficient constructed in this embodiment may be used to disable weak correlation, i.e., weaken correlation, of the leaking drops in the overlapping region, so as to highlight the leaking drops.
Referring to fig. 6, fig. 6 is an image of a region of leaking water droplets. And eliminating a part with dissimilar distribution between the two images by utilizing a difference coefficient, and reserving the part with similar gradient distribution to obtain a leakage water drop area image:
in the method, in the process of the invention,in order to image the area of the leaking water drop,in order to correct the difference image,for the initial difference image to be a picture of the original difference image,in order to correct the displacement correlation image,is a differential coefficient.
Repeating the method in the embodiment, obtaining corresponding seepage water drop area images according to any two adjacent test images, and merging all seepage water drop area images to obtain the final seepage water drop area.
And S003, constructing a waterproof index according to the extracted seepage water drop area, and analyzing the waterproof performance of the aluminum alloy door and window.
According to the method, the seepage water drop area in the detection process of the aluminum alloy door and window can be accurately extracted, the area of the seepage water body area is obtained, the waterproof index of the aluminum alloy door and window is further constructed and used for quantitatively evaluating the waterproof performance of the aluminum alloy door and window, and the waterproof index expression is specifically as follows:
in the method, in the process of the invention,the serial number corresponding to the test image in which the weeping drop area first appears in the test image sequence,sequence numberAluminum alloy door and window with test images corresponding to acquisition timeThe absolute value of the pressure difference between the indoor and the outdoor is S is the area of the water drop leakage area,is waterproof performance index.
The waterproof index construction logic specifically comprises: if the water drops appear quickly and last longer, the water resistance of the door and window is poor, and the water resistance index is smaller; conversely, if the water drops appear later or for a shorter duration, indicating a better water resistance of the door and window, the greater the water resistance index. If the pressure difference between the indoor and the outdoor is large, that is, the outside has obvious air flow or wind force effect, and the leaked water drops are still less or not, the waterproof performance of the door and the window is good, and the waterproof index is larger. Conversely, if the pressure difference between the indoor and outdoor is small or the number of leaked water drops is large, the water-proof performance of the door and window may be poor, and the water-proof index is smaller. If the area of the leaked water drops is large, it may indicate that the window and door has poor waterproof performance, and the water drops may penetrate through large gaps or cracks, so that the waterproof index is smaller. Conversely, if the area of the leaked water drops is smaller, it may indicate that the window and door has better waterproof performance, and the waterproof index is larger.
Further, in this embodiment, the waterproof index threshold is set to 0.6, and the embodiment is not limited, and the embodiment can be set by the practitioner according to the actual situation. When the waterproof index of the aluminum alloy door and window is larger than 0.6, the aluminum alloy door and window is good in waterproof quality, and the aluminum alloy door and window is regarded as qualified in waterproof quality; when the waterproof index of the aluminum alloy is smaller than 0.6, the aluminum alloy door and window is proved to be unqualified in waterproof performance and needs to be reprocessed and overhauled.
The method provided by the embodiment of the invention can be used for accurately detecting the waterproof performance of the aluminum alloy, detecting and extracting the leakage water drop area according to the change characteristics of the test images in the test image sequence, quantitatively analyzing the waterproof performance of the aluminum alloy, and has higher detection precision and smaller calculated amount.
According to the method, an initial differential image is built according to the difference of images before and after the water tightness test and the image subtraction method, a corrected differential image is built according to displacement to solve the problem of pixel point correspondence of the image subtraction method, the part which does not have similarity between two moving differential images is eliminated by utilizing the correlation of the two images, and the similar part is reserved.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. And the foregoing description has been directed to specific embodiments of this specification. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and the same or similar parts of each embodiment are referred to each other, and each embodiment mainly describes differences from other embodiments.
The above embodiments are only for illustrating the technical solution of the present application, and are not limiting; the technical solutions described in the foregoing embodiments are modified or some of the technical features are replaced equivalently, so that the essence of the corresponding technical solutions does not deviate from the scope of the technical solutions of the embodiments of the present application, and all the technical solutions are included in the protection scope of the present application.
Claims (7)
1. The machine vision-based waterproof aluminum alloy door and window detection method is characterized by comprising the following steps of:
collecting multi-frame aluminum alloy door and window images before and after water spraying, and recording the multi-frame aluminum alloy door and window images as a test image sequence;
for any two adjacent test images in the test image sequence, constructing an initial difference image according to the difference of the two adjacent test images; obtaining a corrected differential image according to the gray scale difference of the adjacent test image pixel points; taking the displacement change of each pixel point in the front and rear adjacent images as the pixel displacement quantity of each pixel point, combining the initial difference image and the corrected difference image to correspond to the change of the pixel displacement quantity in the window of the field of each pixel point to obtain the displacement correlation degree of the pixel point at each position, forming a displacement correlation degree image by the displacement correlation degree of each pixel point, acquiring the corrected displacement correlation degree image according to the displacement correlation degree image, and obtaining the water drop change percentage of each pixel point according to the gray information in the local adjacent area of each pixel point corresponding to the initial difference image and the corrected difference image; obtaining a difference coefficient of the pixel point according to the water drop change percentage of the pixel point; acquiring a leakage water drop area image by combining the difference coefficient, the initial difference image, the corrected difference image and the corrected displacement correlation image; acquiring leakage water drop area images corresponding to any two adjacent test images in the test image sequence;
obtaining a waterproof index according to the area of the water seepage drop area and the occurrence time of the water seepage drop area in each water seepage drop area image; the waterproof detection of the waterproof aluminum alloy doors and windows is completed by combining the waterproof index;
the method for acquiring the water drop change percentage of each pixel point comprises the following steps: respectively calculating the sum value of gray values in a pixel point neighborhood window at the same position of an initial difference image and a corrected difference image, obtaining the absolute value of the difference value of the two sums and the maximum value of the two sums, and taking the ratio of the absolute value to the maximum value as the water drop change percentage of the pixel point;
the method for obtaining the difference coefficient of the pixel point according to the water drop change percentage of the pixel point comprises the following steps:
when the water drop change percentage of the pixel point is smaller than or equal to the water drop change percentage threshold value, the difference coefficient of the pixel point is 0; when the water drop change percentage of the pixel point is larger than the water drop change percentage threshold value, the difference coefficient of the pixel point is 1;
wherein, obtaining the waterproof index according to the area of the water seepage drop area and the occurrence time of the water seepage drop area in each water seepage drop area image, comprising:
the method comprises the steps of obtaining the area of a leaking water drop area by combining all leaking water drop area images, obtaining a serial number corresponding to a test image which appears first in the leaking water drop area in a test image sequence, and obtaining an indoor and outdoor pressure difference absolute value of an aluminum alloy door window at the corresponding collection time of the test image of the serial number;
and calculating the product of the serial number and the area of the seepage water drop area, and taking the normalized value of the ratio of the absolute value of the pressure difference to the product as a waterproof index.
2. The machine vision-based waterproof aluminum alloy door and window detection method according to claim 1, wherein the constructing an initial differential image according to the difference between two adjacent test images comprises:
and regarding two adjacent test images, taking the difference value of the gray value of each pixel point in the test image of the next frame and the gray value of each pixel point in the corresponding position in the test image of the previous frame as the gray value of each pixel point in the initial difference image, and obtaining the initial difference image.
3. The machine vision-based waterproof aluminum alloy door and window detection method according to claim 1, wherein the corrected difference image is obtained according to the gray scale difference of the adjacent test image pixels, and the expression is:in (1) the->Representing test image +.>Corresponding corrected difference image, +.>Indicate->In the test image +.>At the pixel pointThe gray-scale value of the gray-scale value,indicate->In the test image +.>The gray value of the pixel point.
4. The machine vision-based waterproof aluminum alloy door and window detection method according to claim 1, wherein the combining the initial difference image and the corrected difference image to correspond to the change of the pixel displacement in the pixel field window to obtain the displacement correlation degree of the pixel at each position comprises:
for each pixel point of the corresponding position of the initial difference image and the corrected difference image;
calculating covariance of pixel displacement of a pixel point neighborhood window of the initial difference image and a pixel point neighborhood window of a corresponding position of the corrected difference image;
respectively calculating pixel displacement variance of a pixel point neighborhood window of the initial difference image and a pixel point neighborhood window of the corresponding position of the corrected difference image, and obtaining the product of the two variances;
and taking the ratio of the covariance to the product as the displacement correlation degree of the pixel point at the corresponding position.
5. The machine vision-based waterproof aluminum alloy door and window detection method according to claim 1, wherein the corrected displacement correlation image is obtained according to the displacement correlation image, and the expression is:
in (1) the->For correcting the displacement correlation image +.>For displacement correlation image, ++>Is a correction parameter greater than 1.
6. The machine vision-based waterproof aluminum alloy door and window detection method according to claim 1, wherein the obtaining the leaked water drop area image by combining the difference coefficient, the initial difference image, the corrected difference image and the corrected displacement correlation image comprises:
and calculating the gray difference value of each pixel point in the corrected difference image and the initial difference image, calculating the product of the gray value of each pixel point in the corrected displacement related image and the difference coefficient, and taking the product of the gray difference value and the product of 1 minus the product as the gray value of each pixel point in the leaked water drop area image.
7. The machine vision-based waterproof aluminum alloy door and window detection method according to claim 1, wherein the completion of waterproof detection of the waterproof aluminum alloy door and window by combining the waterproof index comprises: when the waterproof index is larger than the waterproof index threshold, the aluminum alloy door and window is qualified in waterproof; otherwise, the aluminum alloy door and window is unqualified in waterproof.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311229235.3A CN116977337B (en) | 2023-09-22 | 2023-09-22 | Waterproof aluminum alloy door and window detection method based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311229235.3A CN116977337B (en) | 2023-09-22 | 2023-09-22 | Waterproof aluminum alloy door and window detection method based on machine vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116977337A CN116977337A (en) | 2023-10-31 |
| CN116977337B true CN116977337B (en) | 2024-01-16 |
Family
ID=88471632
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311229235.3A Active CN116977337B (en) | 2023-09-22 | 2023-09-22 | Waterproof aluminum alloy door and window detection method based on machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116977337B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150093043A1 (en) * | 2013-10-02 | 2015-04-02 | National Applied Research Laboratories | Method of Evaluating Image Correlation with Speckle Patter |
| CN107610102A (en) * | 2017-08-24 | 2018-01-19 | 东南大学 | A kind of Displacement measuring method based on Tikhonov regularizations |
| CN116309570A (en) * | 2023-05-18 | 2023-06-23 | 山东亮马新材料科技有限公司 | Titanium alloy bar quality detection method and system |
-
2023
- 2023-09-22 CN CN202311229235.3A patent/CN116977337B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150093043A1 (en) * | 2013-10-02 | 2015-04-02 | National Applied Research Laboratories | Method of Evaluating Image Correlation with Speckle Patter |
| CN107610102A (en) * | 2017-08-24 | 2018-01-19 | 东南大学 | A kind of Displacement measuring method based on Tikhonov regularizations |
| CN116309570A (en) * | 2023-05-18 | 2023-06-23 | 山东亮马新材料科技有限公司 | Titanium alloy bar quality detection method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116977337A (en) | 2023-10-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107563347B (en) | Passenger flow counting method and device based on TOF camera | |
| CN110929566B (en) | Human face living body detection method based on visible light and near infrared binocular camera | |
| CN108388885B (en) | Multi-person close-up real-time identification and automatic screenshot method for large live broadcast scene | |
| CN106600625A (en) | Image processing method and device for detecting small-sized living thing | |
| CN101587541B (en) | Character recognition method based on human body contour outline | |
| CN106780542A (en) | A kind of machine fish tracking of the Camshift based on embedded Kalman filter | |
| CN104834893A (en) | Front-view pedestrian gait period detection method | |
| CN111027432B (en) | Gait feature-based visual following robot method | |
| CN108921062B (en) | Gait recognition method combining multiple gait features and cooperative dictionary | |
| CN102063727B (en) | Covariance matching-based active contour tracking method | |
| CN104794449A (en) | Gait energy image acquisition method based on human body HOG (histogram of oriented gradient) features and identity identification method | |
| CN113378649A (en) | Identity, position and action recognition method, system, electronic equipment and storage medium | |
| CN115375581A (en) | Dynamic visual event stream noise reduction effect evaluation method based on event time-space synchronization | |
| CN115375733A (en) | Snow vehicle sled three-dimensional sliding track extraction method based on videos and point cloud data | |
| CN111582036A (en) | Cross-view-angle person identification method based on shape and posture under wearable device | |
| CN116977337B (en) | Waterproof aluminum alloy door and window detection method based on machine vision | |
| CN109102520A (en) | The moving target detecting method combined based on fuzzy means clustering with Kalman filter tracking | |
| CN103065329A (en) | Space rope tying robot camera automatic motion detection and compensation method | |
| Kuang et al. | An effective skeleton extraction method based on Kinect depth image | |
| CN109285148B (en) | Infrared weak and small target detection method based on heavily weighted low rank and enhanced sparsity | |
| CN104537690B (en) | One kind is based on the united moving spot targets detection method of maximum time index | |
| CN101241546A (en) | Method for compensating for gait binary value distortion | |
| CN110502968B (en) | Method for detecting infrared small and weak moving target based on track point space-time consistency | |
| CN113158942A (en) | Segmentation algorithm and device for detecting motion human behavior | |
| CN110738684A (en) | target tracking method based on correlation filtering fusion convolution residual learning |
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 |