Supercavitation high-pressure water jet machine cleaning quality detection method based on visual technology

Abstract

The invention belongs to the technical field of cleaning quality detection, and relates to a super-cavitation high-pressure water jet machine cleaning quality detection method based on a visual technology. By analyzing the cleaning quality evaluation index of each sub-region corresponding to each specified object and the functional detection evaluation index of each specified object, centralized management of data is facilitated, a personalized cleaning scheme is formulated according to characteristics and requirements of the data, the cleaning efficiency and quality are improved, the cleaning cost is reduced, and the cleaning efficiency is improved. The cleaning quality evaluation index change trend coefficient of each specified object and the functional detection evaluation index change trend coefficient of each specified object are analyzed, so that the accuracy of the sustainable cleaning evaluation condition of each specified object is improved, and the cleaning quality evaluation index change trend coefficient of each specified object and the functional detection evaluation index change trend coefficient of each specified object are improved. The cleaning effect can be reflected comprehensively, abnormal conditions in the cleaning process can be recognized conveniently, measures can be taken in time for treatment, and problem expansion is avoided.

Description

Technical Field

[0001] The invention belongs to the technical field of cleaning quality detection, and relates to a cleaning quality detection method of a supercavitation high-pressure water jet machine based on vision technology. Background Art

[0002] With the increasing busyness of maritime traffic, navigation marks, as important facilities to ensure the safety of ship navigation, have become more and more critical to maintain. Navigation marks are exposed to the marine environment for a long time, and various dirt such as marine organisms, oil stains, rust, etc. will adhere to the surface. These dirt not only affect the appearance of the navigation mark, but also reduce its navigation and warning functions. Traditional navigation mark cleaning quality inspection mainly relies on manual visual inspection and simple physical contact inspection, which is highly subjective and easily affected by the experience, vision and environmental factors of the inspectors. It is also difficult to make accurate judgments on some minor stains or difficult-to-reach areas. The supercavitating high-pressure water jet machine uses the strong impact force generated by the high-pressure water jet and the energy release when the cavitation collapses in the navigation mark cleaning technology, which can effectively remove various stubborn dirt on the surface of the navigation mark. However, a scientific and accurate detection method is needed to evaluate its cleaning quality effect to ensure that the navigation mark can be restored to a good working condition after cleaning.

[0003] In the prior art, there are also some related solutions involving the cleaning of the beacon surface. For example, the invention patent application of a beacon light surface defect recognition method and system with Chinese patent publication number CN118691606B, the method steps include: obtaining an HSI image of the beacon light surface to be detected and an HSI image of the standard beacon light surface, identifying all the stained pixels in the image to be detected, and replacing the original brightness value of the target pixel with the weighted brightness value of all non-stained pixels in the neighborhood of the stained pixel to obtain a new brightness value of the target pixel; using the new brightness values ​​and original brightness values ​​of all the stained pixels contained in the image to be detected, calculating the light blocking degree of the image to be detected, when the light blocking degree of the image to be detected is greater than the preset light blocking degree, determining that the beacon light surface needs to be cleaned; the present invention can accurately identify light-blocking stains in the image to be detected, and determine whether the beacon light surface needs to be cleaned according to the light blocking degree of the light-blocking stains.

[0004] Although the above scheme analyzes the light blocking degree of the beacon light surface to judge the cleaning requirement based on the HSI image of the beacon light surface and the HSI image of the standard beacon light surface, the above scheme lacks further detection and analysis of the cleaning condition of the beacon light, and thus cannot understand the use condition of the beacon light after cleaning, which is not conducive to obtaining specific problems of the beacon light in a more comprehensive and accurate manner.

[0005] Another invention patent application for a cleaning quality evaluation method with a Chinese patent publication number of CN112991326B. It includes: collecting images of parts after cleaning, preprocessing and splicing the images of parts; quantifying and scoring the color depth of the stains, and multiplying it by the area of ​​the stains to obtain the score of each stain, and the ratio of the sum of the scores of all stains on the parts to the total area of ​​the surface of the parts to obtain the score A; according to the shape and distribution of the stains, the energy consumption of the secondary washing is analyzed, and the cleaning area required for the secondary washing is calculated using morphological expansion operations to obtain the ratio of the area that does not need to be cleaned to the total area of ​​the surface of the parts to obtain the score B; weighted calculation of the score A and the score B is performed to obtain the total score of the cleaning quality of the parts. The present invention effectively solves the problems that the existing cleaning equipment lacks the function of automatically evaluating the cleaning quality, the lack of consistency in the evaluation standards, the low efficiency of the evaluation method, and the difficulty in ensuring the accuracy of the evaluation results.

[0006] Since the existing technology of navigation beacon cleaning is basically combined with image vision technology for analysis, and the above scheme analyzes the image of the cleaned object to obtain the total cleaning quality score of the cleaned object. However, it mainly detects and analyzes surface stains, ignoring the comparison and analysis of the navigation beacon before and after cleaning and after cleaning with the standard situation, which is not conducive to formulating personalized cleaning plans according to characteristics and needs, and cannot improve cleaning efficiency and quality, nor is it conducive to improving the accuracy of the sustainable cleaning evaluation of the navigation beacon, and cannot more comprehensively reflect the cleaning effect. Summary of the invention

[0007] In view of this, in order to solve the problems raised in the above background technology, a supercavitation high-pressure water jet cleaning quality detection method based on visual technology is proposed.

[0008] The purpose of the present invention can be achieved through the following technical solutions: The present invention provides a supercavitation high-pressure water jet cleaning quality detection method based on visual technology, including: S1, area division: each object that needs to be cleaned by a supercavitation high-pressure water jet machine at the current time is recorded as a designated object, an initial image of each designated object is obtained, and each designated object is divided according to a predefined principle to obtain each sub-area corresponding to each designated object.

[0009] S2. Acquisition of cleaning quality information: Real-time photography is performed on each designated object during the cleaning process to obtain images of each sub-region corresponding to each designated object before and after cleaning, and cleaning quality information of each designated object before and after cleaning is obtained based on the images, including dirt information of each sub-region corresponding to each designated object before and after cleaning and functional information of each designated object before and after cleaning.

[0010] S3. Analysis of cleaning quality information: Analyze the cleaning quality evaluation index of each sub-region corresponding to each designated object and the functional detection evaluation index of each designated object, store them in the database, and obtain the cleaning quality evaluation status of each designated object.

[0011] S4. Analysis of cleaning quality trend: Extract the cleaning quality evaluation index of each sub-area corresponding to each designated object analyzed each time and the functional detection evaluation index of each designated object from the database, analyze the changing trend coefficient of the cleaning quality evaluation index of each designated object and the changing trend coefficient of the functional detection evaluation index of each designated object, and obtain the sustainable cleaning evaluation status of each designated object.

[0012] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. The present invention divides each designated object according to a predefined principle to obtain sub-areas corresponding to each designated object, which helps to more accurately evaluate the cleaning effect and further facilitates the discovery of potential problems in the cleaning process.

[0013] 2. The present invention analyzes the cleaning quality evaluation index of each sub-area corresponding to each designated object and the functional detection evaluation index of each designated object, stores them in a database, and obtains the cleaning quality evaluation status of each designated object, which helps to realize centralized management of data, facilitates the formulation of personalized cleaning plans based on its characteristics and needs, improves cleaning efficiency and quality, reduces cleaning costs, and also helps to ensure that the cleaning process does not have a negative impact on the function of the object.

[0014] 3. The present invention analyzes the changing trend coefficient of the cleaning quality evaluation index of each designated object and the changing trend coefficient of the functional detection evaluation index of each designated object, and obtains the sustainable cleaning evaluation of each designated object. By analyzing rich data samples, the accuracy of the sustainable cleaning evaluation of each designated object is improved, and it is helpful to fully reflect the cleaning effect, and it is also beneficial to identify abnormal situations in the cleaning process, such as reduced cleaning quality, damaged functions, etc., so as to facilitate timely measures to deal with them and avoid the expansion of problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0016] Figure 1 It is a schematic diagram for implementing the method steps of the present invention.

[0017] Figure 2It is a flow chart of the cleaning quality evaluation analysis process of the present invention.

[0018] Figure 3 It is a flow chart of the sustainable cleaning evaluation situation analysis process of the present invention. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] See also Figure 1 As shown, the present invention provides a supercavitation high-pressure water jet cleaning quality detection method based on visual technology, and the specific steps are as follows: S1, area division: each object that needs to be cleaned by a supercavitation high-pressure water jet machine at the current time is recorded as a designated object, an initial image of each designated object is obtained, and each designated object is divided according to a predefined principle to obtain each sub-area corresponding to each designated object.

[0021] As a preferred feasible embodiment, the specific content of the predefined principles in each sub-region corresponding to each designated object is as follows: extracting the initial image of each designated object, dividing it according to the standard image of each designated object corresponding to the area of ​​the marked area type stored in the database, obtaining the area corresponding to each area type in the initial image of each designated object, and further re-recording it in order as the sub-region corresponding to each designated object.

[0022] It should be further explained that the area types include but are not limited to functional areas, seawater immersion areas, weathering and sun exposure areas, key inspection areas and difficult-to-clean areas.

[0023] The area type corresponding areas include areas of the same area type at different locations.

[0024] In a specific example, if the standard image of a designated object with marked region types corresponding to regions contains four region types, and the number of regions contained in the four region types is 3, 2, 1 and 3 respectively. Based on the division, the regions corresponding to the region types in the initial image of the designated object can be obtained, and the specific operation of further re-recording them in order as the sub-regions corresponding to the designated object can be: the first region type in the initial image of the designated object corresponds to the first region, the first region type corresponds to the second region, the first region type corresponds to the third region, the second region type corresponds to the first region, the second region type corresponds to the second region, the third region type corresponds to a region, the fourth region type corresponds to the first region, the fourth region type corresponds to the second region, and the fourth region type corresponds to the third region are recorded in order as the sub-regions corresponding to the designated object, that is, the number of sub-regions corresponding to the designated object is 9.

[0025] The present invention divides each designated object according to a predefined principle to obtain each sub-area corresponding to each designated object, which helps to more accurately evaluate the cleaning effect and further facilitates the discovery of potential problems in the cleaning process.

[0026] S2. Acquisition of cleaning quality information: Real-time photography is performed on each designated object during the cleaning process to obtain images of each sub-region corresponding to each designated object before and after cleaning, and cleaning quality information of each designated object before and after cleaning is obtained based on the images, including dirt information of each sub-region corresponding to each designated object before and after cleaning and functional information of each designated object before and after cleaning.

[0027] As a preferred feasible embodiment, the dirt information of each sub-region corresponding to each designated object before and after cleaning includes the number of distribution regions of each type of dirt, the area of ​​each distribution region and the color deviation value.

[0028] It needs to be further explained that the specific method for obtaining the number of distribution areas and the area of ​​each distribution area of ​​each type of dirt corresponding to each sub-area of ​​each designated object before and after cleaning is: the images of each sub-area corresponding to each designated object before cleaning are compared one by one with the images corresponding to each type of dirt stored in the database, and the types of dirt in the images of each sub-area corresponding to each designated object before cleaning are obtained, which are recorded as the types of dirt in each sub-area corresponding to each designated object before cleaning, and the images of each sub-area corresponding to each designated object before cleaning are further analyzed by image analysis software to obtain the areas of each distribution area of ​​each type of dirt in each sub-area corresponding to each designated object before cleaning, and the distribution areas of each type of dirt in each sub-area corresponding to each designated object before cleaning are counted to obtain the number of distribution areas of each type of dirt in each sub-area corresponding to each designated object before cleaning.

[0029] Similarly, the number of distribution regions and the area of ​​each distribution region of each type of dirt in each sub-region corresponding to each designated object after cleaning can be obtained.

[0030] It should be noted that the specific method for obtaining the various types of dirt in the image before cleaning for each sub-region corresponding to each designated object is as follows: if a certain area in the image before cleaning for a certain sub-region corresponding to a designated object exists in a certain area that is consistent with an image corresponding to a certain type of dirt stored in the database, then this type of dirt is recorded as the type of dirt in the image before cleaning for the sub-region corresponding to the designated object, and then the various types of dirt in the image before cleaning for each sub-region corresponding to each designated object are obtained.

[0031] The specific method for obtaining the color deviation value of each sub-region corresponding to each designated object before and after cleaning is as follows: grayscale processing is performed on the image of each sub-region corresponding to each designated object before cleaning, and the grayscale value of each pixel point corresponding to the grayscale image of each sub-region corresponding to each designated object before cleaning is obtained. ijq , where i = 1, 2, ..., a, i is the number of each specified object, a is the number of specified objects, j = 1, 2, ..., b, j is the number of each sub-region, b is the number of sub-regions, q = 1, 2, ..., Q, q is the number of each pixel, Q is the number of pixels, and the color deviation value of each sub-region corresponding to each specified object before cleaning is analyzed Gray i ' jq is the standard gray value of the qth pixel corresponding to the jth sub-region of the grayscale image before cleaning of the ith specified object extracted from the database, and ΔGray0 is the allowable difference between the set gray value and the standard gray value.

[0032] By obtaining the grayscale value of each pixel of the grayscale image corresponding to each sub-region of each designated object before cleaning, the grayscale value of each pixel of the grayscale image corresponding to each sub-region of each designated object after cleaning is obtained. Further, according to the analysis method of the color deviation value of each sub-region of each designated object before cleaning, the color deviation value of each sub-region of each designated object after cleaning can be obtained in the same way.

[0033] The functional information of each designated object before and after cleaning includes the degree of appearance damage, the reflective marking performance evaluation index and the lamp performance evaluation index.

[0034] It needs to be further explained that the specific method for obtaining the degree of appearance damage of each designated object before and after cleaning is: using image analysis software to analyze the image of each sub-region corresponding to each designated object before cleaning, to obtain the appearance damage area of ​​each sub-region corresponding to each designated object in the image before cleaning, and then respectively comparing it with the total appearance area of ​​each sub-region corresponding to each designated object in the image before cleaning, and the obtained quotient is recorded as the degree of appearance damage of each designated object before cleaning. Similarly, the degree of appearance damage of each designated object after cleaning can be obtained.

[0035] The specific method for obtaining the reflective mark performance evaluation index of each designated object before and after cleaning is as follows: the image of the reflective mark area of ​​each designated object before cleaning is obtained by screening each sub-area corresponding to each designated object in the image before cleaning, and the image is imported into the OpenCV library, and the red, green and blue (R, G, B) three-channel values ​​of each pixel point of the image of the reflective mark area of ​​each designated object before cleaning are obtained by processing, which are recorded as R if , G if , B if , where f = 1, 2, ..., F, f is the number of pixels corresponding to the reflective marking area before cleaning, F is the number of pixels corresponding to the reflective marking area before cleaning, and the red, green, and blue (R, G, B) three-channel values ​​of each pixel corresponding to the reflective marking area of ​​each designated object before cleaning are normalized to obtain the normalized corresponding value r of the red, green, and blue (R, G, B) three-channel values ​​of each pixel corresponding to the reflective marking area of ​​each designated object before cleaning. if , g if 、b if .

[0036] It should be explained that the specific content of the normalization process is: according to the calculation formula Get the normalized corresponding value r of the red, green and blue (R, G, B) three-channel value of each pixel in the image corresponding to the reflective identification area of ​​each specified object before cleaning if , g if 、b if .

[0037] According to the standard calculation formula V if =max(r if ,g if ,b if ) can obtain the brightness V of each pixel in the image corresponding to the reflective mark area of ​​each specified object before cleaning if , and the mean value of the image before cleaning can be obtained by performing mean value processing on the reflective mark area corresponding to each specified object, which is recorded as the reflective intensity of the image before cleaning corresponding to the reflective mark area of ​​each specified object. i .

[0038] According to the calculation formula Get the brightness standard deviation of the image of the reflective identification area of ​​each specified object before cleaning

[0039] Analyze the reflective marking performance evaluation index of each specified object before cleaning Among them, Inten0, They are the preset reflective intensity threshold and brightness standard deviation threshold of the reflective identification area corresponding to the specified object. Similarly, the reflective identification performance evaluation index of each specified object after cleaning can be obtained.

[0040] The specific method for obtaining the lamp performance evaluation index of each designated object before and after cleaning is as follows: the image of the lamp area of ​​each designated object before cleaning is obtained by screening from each sub-area corresponding to each designated object in the image before cleaning, and according to the method for obtaining the brightness of each pixel corresponding to the reflective identification area of ​​each designated object before cleaning, the brightness of each pixel corresponding to the image of the lamp area of ​​each designated object before cleaning can be obtained in the same way, and the brightness threshold of the pixel corresponding to the lamp area stored in the database is compared, and the pixel corresponding to the image of the lamp area of ​​each designated object before cleaning whose brightness is greater than the brightness threshold of the pixel corresponding to the lamp area is screened, and it is recorded as the luminous pixel corresponding to the image of the lamp area of ​​each designated object before cleaning, and the number of luminous pixel corresponding to the image of the lamp area of ​​each designated object before cleaning is obtained by counting, and the number of luminous pixel corresponding to the image of the lamp area of ​​each designated object before cleaning is divided by the total number of pixel corresponding to the image of the lamp area of ​​each designated object before cleaning, and the quotient obtained is recorded as the luminous range index Lum of the image of the lamp area of ​​each designated object before cleaning. i .

[0041] The brightness of each luminous pixel in the image before cleaning corresponding to the lamp area of ​​each designated object is further obtained, and the average value of the brightness of the luminous area in the image before cleaning corresponding to the lamp area of ​​each designated object is obtained by average processing, which is recorded as the luminous intensity Inten of the image before cleaning corresponding to the lamp area of ​​each designated object i ′.

[0042] Analyze the lamp performance evaluation index of each specified object before cleaning Where Inten′0 is the preset luminous intensity threshold of the lamp area corresponding to the specified object. Similarly, the lamp performance evaluation index of each specified object after cleaning can be obtained.

[0043] S3. Analysis of cleaning quality information: Analyze the cleaning quality evaluation index of each sub-region corresponding to each designated object and the functional detection evaluation index of each designated object, store them in the database, and obtain the cleaning quality evaluation status of each designated object.

[0044] As a preferred feasible embodiment, the specific analysis method of the cleaning quality evaluation index of each sub-region corresponding to each designated object is as follows: extracting the number of distribution areas of each type of dirt, the area of ​​each distribution area and the color deviation value of each sub-region corresponding to each designated object before and after cleaning, and recording them as Wherein, i=1,2,...,a, i is the number of each designated object, a is the number of designated objects, j=1,2,...,b, j is the number of each sub-region, b is the number of sub-regions, g=1,2,...,G, g is the number of each type of dirt, G is the number of dirt types, h=1,2,...,m, h is the number of each distribution area, m is the number of distribution areas.

[0045] Analyze the cleaning quality evaluation index of each sub-area corresponding to each specified object in are the quality evaluation coefficients before and after cleaning and the cleaning standard quality evaluation coefficients of the jth sub-region corresponding to the i-th specified object, respectively, and Among them, m0, S0, and CDev0 are the minimum difference in the number of dirt distribution areas, the minimum difference in the area of ​​the distribution area, and the minimum difference in the color deviation value of the preset object before and after cleaning, respectively. Among them, μ0, μ1, and μ2 are the minimum values ​​set to prevent the denominator from being 0, and their units are dimensionless, square meters, and dimensionless, respectively, and β1 and β2 are the weight factors of the evaluation coefficients before and after cleaning and the evaluation coefficients of the cleaning standard, respectively. It should be noted that the preset objects are navigation marks placed in advance.

[0046] In a specific example, β1=0.4 and β2=0.6.

[0047] The pre- and post-cleaning evaluation coefficient focuses on the cleaning effect of each sub-area of ​​each specified object before and after cleaning. By comparing the status before and after cleaning, the effect of the cleaning operation can be intuitively reflected. However, since the cleaning effect may be affected by many factors, such as cleaning method, cleaning time, etc., this coefficient is important but may not be the only evaluation criterion. Therefore, the weight factor of the pre- and post-cleaning evaluation coefficient is assigned to 0.4.

[0048] The cleaning standard evaluation coefficient is evaluated based on certain cleaning standards or specifications. These standards or specifications are usually formulated based on industry experience, scientific research or regulatory requirements, and are more authoritative and instructive. When evaluating the cleaning effect, giving this coefficient a higher weight can ensure the accuracy and reliability of the evaluation results. Therefore, the weight factor of the cleaning standard evaluation coefficient is assigned to 0.6.

[0049] As a preferred feasible embodiment, the specific analysis method of the functional detection evaluation index of each designated object is: extract the appearance damage degree, reflective label performance evaluation index and lamp performance evaluation index of each designated object before and after cleaning, and record them as Analyze the functional test evaluation indicators of each specified object in are the functional test evaluation coefficients before and after cleaning and the cleaning standard functional test evaluation coefficients of the i-th specified object, respectively, and Dam0, RefM0, and NavL0 are the preset minimum difference in appearance damage before and after cleaning, the minimum difference in reflective sign performance evaluation index, and the minimum difference in lamp performance evaluation index, respectively. Wherein μ3 is the minimum value set to prevent the denominator from being 0, and its units are dimensionless.

[0050] As a preferred feasible embodiment, the specific method for obtaining the cleaning quality evaluation status of each designated object is: comparing the cleaning quality evaluation index of each sub-region corresponding to each designated object with the preset cleaning quality evaluation index threshold, screening the sub-regions corresponding to each designated object whose cleaning quality evaluation index is greater than the cleaning quality evaluation index threshold, and recording them as the qualified cleaning sub-regions corresponding to each designated object, and then counting the number of qualified cleaning sub-regions corresponding to each designated object, and performing ratio processing on the ratio with the total number of sub-regions corresponding to each designated object; if the ratio is greater than the set ratio threshold, the cleaning quality evaluation status of the sub-region corresponding to each designated object is recorded as qualified; otherwise, the cleaning quality evaluation status of the sub-region corresponding to each designated object is recorded as unqualified.

[0051] The functional detection evaluation index of each designated object is compared with the preset functional detection evaluation index threshold. If the functional detection evaluation index of a designated object is greater than the functional detection evaluation index threshold, the functional detection evaluation situation of the designated object is recorded as qualified; otherwise, the functional detection evaluation situation of the designated object is recorded as unqualified, thereby obtaining the functional detection evaluation situation of each designated object.

[0052] The cleaning quality evaluation of each designated object whose cleaning quality evaluation of the corresponding sub-area and the functional test evaluation are both qualified is recorded as qualified.

[0053] It needs to be further explained that the cleaning quality evaluation analysis process flow chart is as follows Figure 2 shown.

[0054] The present invention analyzes the cleaning quality evaluation index of each sub-area corresponding to each designated object and the functional detection evaluation index of each designated object, stores them in a database, and obtains the cleaning quality evaluation status of each designated object, which helps to realize centralized management of data, is conducive to formulating personalized cleaning plans according to its characteristics and needs, improves cleaning efficiency and quality, reduces cleaning costs, and also helps to ensure that the cleaning process does not have a negative impact on the function of the object.

[0055] S4. Analysis of cleaning quality trend: Extract the cleaning quality evaluation index of each sub-area corresponding to each designated object analyzed each time and the functional detection evaluation index of each designated object from the database, analyze the changing trend coefficient of the cleaning quality evaluation index of each designated object and the changing trend coefficient of the functional detection evaluation index of each designated object, and obtain the sustainable cleaning evaluation status of each designated object.

[0056] As a preferred feasible embodiment, the specific analysis method of the change trend coefficient of the cleaning quality evaluation index of each designated object is: extract the cleaning quality evaluation index Clean corresponding to each sub-area of ​​each designated object analyzed in each time r ' ij , where r = 1, 2, ..., R, r is the number of each analysis, R is the number of analyses, and the change trend coefficient of the cleaning quality evaluation index of each specified object is analyzed

[0057] Where Clean(′ R-1 ) ij 、Clean′ Rij 、Clean(′ r+1 ) ij are the cleaning quality evaluation indexes of the jth sub-area corresponding to the ith specified object in the (R-1)th analysis, the Rth analysis, and the (r+1)th analysis, respectively. δ1 and δ2 are the weight coefficients corresponding to the set recent cleaning quality evaluation index and long-term cleaning quality evaluation index, respectively.

[0058] It should be further explained that the recent cleaning quality evaluation index is the current cleaning quality evaluation index and the previous cleaning quality evaluation index.

[0059] The long-term cleaning quality evaluation index is the historical cleaning quality evaluation index with the last cleaning time point as the boundary.

[0060] In a specific example, δ1=0.67, δ2=0.33.

[0061] The recent cleaning quality directly reflects the effect and efficiency of the current cleaning operation, which is of great significance for ensuring the normal operation of each designated object. Therefore, the weight coefficient corresponding to the recent cleaning quality evaluation index is assigned to 0.67, which can more directly reflect the current status of the cleaning work and facilitate timely discovery and resolution of problems.

[0062] Long-term cleaning quality focuses more on the long-term impact of the cleaning effect on each specified object. Although the impact of long-term cleaning quality may not be as intuitive and urgent as that of recent cleaning quality, it is equally important for maintaining the long-term stable operation of equipment, reducing failure rates and maintenance costs, etc. Therefore, the weight coefficient corresponding to the long-term cleaning quality evaluation index is assigned a value of 0.33.

[0063] As a preferred feasible embodiment, the specific analysis method of the functional detection evaluation index change trend coefficient of each designated object is: extract the functional detection evaluation index Fun of each designated object analyzed each time r ' i , analyze the trend coefficient of functional detection evaluation indicators of each specified object Among them, Fun(′ R-1 ) i 、Fun′ Ri 、Fun(′ r+1 ) i They are the functional detection evaluation indicators of the i-th specified object in the (R-1)th analysis, the Rth analysis, and the (r+1)th analysis, respectively.

[0064] As a preferred feasible embodiment, the specific method of obtaining the sustainable cleaning evaluation status of each designated object is: comparing the cleaning quality evaluation index change trend coefficient and the functional detection evaluation index change trend coefficient of each designated object with the preset cleaning quality evaluation index change trend coefficient threshold and the functional detection evaluation index change trend coefficient threshold respectively; if the cleaning quality evaluation index change trend coefficient and the functional detection evaluation index change trend coefficient of a designated object are both smaller than the cleaning quality evaluation index change trend coefficient threshold and the functional detection evaluation index change trend coefficient threshold, the sustainable cleaning evaluation status of the designated object is recorded as unsustainable; otherwise, the sustainable cleaning evaluation status of the designated object is recorded as sustainable.

[0065] It is necessary to further explain that the sustainable cleaning evaluation situation analysis process flow chart is as follows Figure 3 shown.

[0066] As a preferred feasible embodiment, the method uses a database during the execution process to store the cleaning quality evaluation indicators of each sub-area corresponding to each specified object analyzed each time and the functional detection evaluation indicators of each specified object, store the standard images of each specified object corresponding to the area of ​​the marked area type, store the corresponding images of each type of dirt, store the standard grayscale values ​​of each pixel point of the grayscale image of each sub-area corresponding to each specified object before cleaning, and store the brightness threshold of the pixel point corresponding to the lighting area.

[0067] The present invention analyzes the changing trend coefficient of the cleaning quality evaluation index of each designated object and the changing trend coefficient of the functional detection evaluation index of each designated object, and obtains the sustainable cleaning evaluation situation of each designated object. By analyzing rich data samples, the accuracy of the sustainable cleaning evaluation situation of each designated object is improved, and it is helpful to fully reflect the cleaning effect, and it is also beneficial to identify abnormal situations in the cleaning process, such as reduced cleaning quality, damaged functions, etc., so as to facilitate timely measures to deal with them and avoid the expansion of problems.

[0068] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they shall all fall within the protection scope of the present invention.

Claims

1. A method for detecting the cleaning quality of a supercavitation high-pressure water jet machine based on visual technology, characterized in that: include: S1, region division: each object that needs to be cleaned by the supercavitation high-pressure water jet machine at the current time is recorded as each designated object, an initial image of each designated object is obtained, and each designated object is divided according to a predefined principle to obtain each sub-region corresponding to each designated object; S2, obtaining cleaning quality information: taking real-time photos of each designated object during the cleaning process, obtaining images of each sub-region corresponding to each designated object before and after cleaning, and obtaining cleaning quality information of each designated object before and after cleaning, including dirt information of each sub-region corresponding to each designated object before and after cleaning and functional information of each designated object before and after cleaning; S3, cleaning quality information analysis: analyzing the cleaning quality evaluation index of each sub-region corresponding to each designated object and the functional detection evaluation index of each designated object, storing them in the database, and obtaining the cleaning quality evaluation status of each designated object; S4. Analysis of cleaning quality trend: Extract the cleaning quality evaluation index of each sub-area corresponding to each designated object analyzed each time and the functional detection evaluation index of each designated object from the database, analyze the changing trend coefficient of the cleaning quality evaluation index of each designated object and the changing trend coefficient of the functional detection evaluation index of each designated object, and obtain the sustainable cleaning evaluation status of each designated object.

2. The method for detecting the cleaning quality of a supercavitation high-pressure water jet machine based on visual technology according to claim 1 is characterized in that: The specific contents of the predefined principles in each sub-region corresponding to each designated object by dividing each designated object according to the predefined principles are as follows: Extract the initial image of each designated object, divide it into standard images of each designated object according to the area corresponding to the marked area type stored in the database, obtain the area corresponding to each area type in the initial image of each designated object, and further re-record it in order as each sub-area corresponding to each designated object.

3. The method for detecting the cleaning quality of a supercavitation high-pressure water jet machine based on visual technology according to claim 1 is characterized in that: The dirt information of each sub-region corresponding to each designated object before and after cleaning includes the number of distribution regions of each type of dirt, the area of ​​each distribution region, and the color deviation value; The functional information of each designated object before and after cleaning includes the degree of appearance damage, the reflective marking performance evaluation index and the lamp performance evaluation index.

4. The method for detecting the cleaning quality of a supercavitation high-pressure water jet machine based on visual technology according to claim 3 is characterized in that: The specific analysis method of the cleaning quality evaluation index of each sub-area corresponding to each designated object is as follows: Extract the number of distribution areas of each type of dirt, the area of ​​each distribution area, and the color deviation value of each sub-area of ​​each specified object before and after cleaning, and record them as Wherein, i=1,2,...,a, i is the number of each designated object, a is the number of designated objects, j=1,2,...,b, j is the number of each sub-region, b is the number of sub-regions, g=1,2,...,G, g is the number of each type of dirt, G is the number of dirt types, h=1,2,...,m, h is the number of each distribution area, m is the number of distribution areas; Analyze the cleaning quality evaluation index of each sub-area corresponding to each specified object in are the quality evaluation coefficients before and after cleaning and the cleaning standard quality evaluation coefficients of the jth sub-region corresponding to the i-th specified object, respectively, and Among them, m0, S0, and CDev0 are the minimum difference in the number of dirt distribution areas, the minimum difference in the area of ​​the distribution area, and the minimum difference in the color deviation value of the preset object before and after cleaning, respectively. Among them, μ0, μ1, and μ2 are the minimum values ​​set to prevent the denominator from being 0, and their units are dimensionless, square meters, and dimensionless respectively. β1 and β2 are the weight factors of the set before and after cleaning evaluation coefficients and cleaning standard evaluation coefficients respectively.

5. The method for detecting the cleaning quality of a supercavitation high-pressure water jet machine based on visual technology according to claim 4 is characterized in that: The specific analysis method of the functional detection evaluation index of each specified object is as follows: The appearance damage degree, reflective sign performance evaluation index and lamp performance evaluation index of each specified object before and after cleaning are extracted and recorded as Analyze the functional test evaluation indicators of each specified object in are the functional test evaluation coefficients before and after cleaning and the cleaning standard functional test evaluation coefficients of the i-th specified object, respectively, and Dam0, RefM0, and NavL0 are the preset minimum difference in appearance damage before and after cleaning, the minimum difference in reflective sign performance evaluation index, and the minimum difference in lamp performance evaluation index, respectively. Wherein μ3 is the minimum value set to prevent the denominator from being 0, and its units are dimensionless.

6. The method for detecting the cleaning quality of a supercavitation high-pressure water jet machine based on visual technology according to claim 5 is characterized in that: The specific method for obtaining the cleaning quality evaluation of each designated object is as follows: The cleaning quality evaluation index of each sub-region corresponding to each designated object is compared with the preset cleaning quality evaluation index threshold, and the sub-regions corresponding to each designated object whose cleaning quality evaluation index is greater than the cleaning quality evaluation index threshold are screened, and recorded as the qualified cleaning sub-regions corresponding to each designated object, and then the number of qualified cleaning sub-regions corresponding to each designated object is counted, and the ratio is processed with the total number of sub-regions corresponding to each designated object. If the ratio is greater than the set ratio threshold, the cleaning quality evaluation of the sub-region corresponding to each designated object is recorded as qualified, otherwise, the cleaning quality evaluation of the sub-region corresponding to each designated object is recorded as unqualified; The functional detection evaluation index of each designated object is compared with the preset functional detection evaluation index threshold value. If the functional detection evaluation index of a designated object is greater than the functional detection evaluation index threshold value, the functional detection evaluation situation of the designated object is recorded as qualified. Otherwise, the functional detection evaluation situation of the designated object is recorded as unqualified, thereby obtaining the functional detection evaluation situation of each designated object; The cleaning quality evaluation of each designated object whose cleaning quality evaluation of the corresponding sub-area and the functional test evaluation are both qualified is recorded as qualified.

7. The method for detecting the cleaning quality of a supercavitation high-pressure water jet machine based on visual technology according to claim 5 is characterized in that: The specific analysis method of the change trend coefficient of the cleaning quality evaluation index of each specified object is: Extract the cleaning quality evaluation index Clean for each sub-area of ​​each specified object analyzed in each analysis r ' ij , where r = 1, 2, ..., R, r is the number of each analysis, R is the number of analyses, and the change trend coefficient of the cleaning quality evaluation index of each specified object is analyzed Where Clean(′ R-1 ) ij 、Clean′ Rij 、Clean(′ r+1 ) ij are the cleaning quality evaluation indexes of the jth sub-area corresponding to the ith specified object in the (R-1)th analysis, the Rth analysis, and the (r+1)th analysis, respectively. δ1 and δ2 are the weight coefficients corresponding to the set recent cleaning quality evaluation index and long-term cleaning quality evaluation index, respectively.

8. The method for detecting the cleaning quality of a supercavitation high-pressure water jet machine based on visual technology according to claim 7 is characterized in that: The specific analysis method of the functional detection evaluation index change trend coefficient of each specified object is as follows: Extract the functional detection evaluation index Fun of each specified object in each analysis r ' i , analyze the trend coefficient of functional detection evaluation indicators of each specified object Among them, Fun(′ R-1 ) i 、Fun′ Ri 、Fun(′ r+1 ) i They are the functional detection evaluation indicators of the i-th specified object in the (R-1)th analysis, the Rth analysis, and the (r+1)th analysis, respectively.

9. The method for detecting the cleaning quality of a supercavitation high-pressure water jet machine based on visual technology according to claim 8 is characterized in that: The specific method for obtaining the sustainable cleaning evaluation of each designated object is as follows: The cleaning quality evaluation index change trend coefficient and the functional detection evaluation index change trend coefficient of each designated object are compared with the preset cleaning quality evaluation index change trend coefficient threshold and the functional detection evaluation index change trend coefficient threshold respectively. If the cleaning quality evaluation index change trend coefficient and the functional detection evaluation index change trend coefficient of a designated object are both smaller than the cleaning quality evaluation index change trend coefficient threshold and the functional detection evaluation index change trend coefficient threshold, the sustainable cleaning evaluation situation of the designated object is recorded as unsustainable; otherwise, the sustainable cleaning evaluation situation of the designated object is recorded as sustainable.

10. The method for detecting the cleaning quality of a supercavitation high-pressure water jet machine based on visual technology according to claim 1 is characterized in that: The method uses a database during the execution process to store the cleaning quality evaluation indicators of each sub-area corresponding to each designated object analyzed each time and the functional detection evaluation indicators of each designated object, store the standard images of each designated object corresponding to the area of ​​the marked area type, store the corresponding images of each type of dirt, store the standard grayscale values ​​of each pixel point of the grayscale image of each sub-area corresponding to each designated object before cleaning, and store the brightness threshold of the pixel point corresponding to the lighting area.

Images