A method for detecting washing performance of a dishwasher based on fluorescence imaging

Abstract

The application discloses a kind of based on fluorescent imaging's dish washer washing performance detection method, comprising: based on the temperature time parameter of the dish washer to be tested to the pollutant and the contaminated tableware is simulated washing, obtains washed product, according to the three-dimensional fluorescence spectrum of washed pollutant and the fluorescence image of the fluorescent imaging system acquisition washed tableware, according to the physical parameter of washed tableware to the pollution contact surface is modeled, using curved surface model to the pollution area in fluorescence image is corrected, obtains correction area, based on correction area extraction washed tableware in the total pixel area of pollution area, and calculate and obtain washing performance score.The method utilizes fluorescent technology to solve the problem that it is difficult to identify when pollutant and table surface color are similar, gives washing performance score evaluation mode from pixel level, which helps to promote the intelligent development of dish washer washing performance test.

Description

Technical Field

[0001] This invention relates to the field of environmental science, and in particular to a method for testing the cleaning performance of a dishwasher based on fluorescence imaging. Background Technology

[0002] With the improvement of people's living standards in my country, dishwashers have become a powerful tool for efficient living and production. However, the current national standards for evaluating the cleaning performance of dishwashers are subject to subjective factors and have low testing efficiency. Therefore, it is of great significance to improve the scientific nature of national standards and enhance the international competitiveness of my country's national standards by using imaging to replace visual recognition and combining image processing algorithms to establish a quantifiable dishwasher cleaning performance testing technology.

[0003] Fluorescence imaging technology has wide applications in analytical chemistry, food science and other fields. Fluorescence is a photoluminescence phenomenon with specificity. The analyte will only emit fluorescence when it contains fluorescent substances. Fluorescent substances will emit fluorescence under the excitation light of a specific ultraviolet wavelength. Therefore, by selecting a suitable fluorescence excitation source, it is possible to ensure that foodborne contaminants on the surface of tableware emit fluorescence while the background of the tableware does not respond with fluorescence, thereby improving the accuracy of contaminant identification. In addition, there will be errors when the analyte in three-dimensional space is projected onto the two-dimensional plane of the camera sensor. The curved surface characteristics of some tableware surfaces will cause distortion of the area of ​​contaminants in the image. Therefore, it is necessary to establish a surface correction algorithm to reduce the imaging distortion of contaminants. Summary of the Invention

[0004] The purpose of this invention is to provide a method for detecting the cleaning performance of a dishwasher based on fluorescence imaging.

[0005] To achieve the above objectives, the present invention is implemented according to the following technical solution:

[0006] The first aspect of this invention provides a method for detecting the cleaning performance of a dishwasher based on fluorescence imaging, comprising:

[0007] Based on the temperature and time parameters of the washing mode obtained from the dishwasher under test, the contaminants and contaminated tableware are simulated to be washed without spraying by static baking according to the temperature and time parameters, and the washed products are obtained.

[0008] The optimal wavelength parameters for fluorescence imaging are determined based on the three-dimensional fluorescence spectrum of the washed contaminants. A fluorescence imaging system is then built based on the optimal wavelength parameters, and fluorescence images of the washed tableware are acquired using the fluorescence imaging system.

[0009] The contaminated contact surface is modeled based on the physical parameters of the washed tableware, and the contaminated area in the fluorescence image is corrected using the surface model to obtain the corrected area.

[0010] The total pixel area of ​​the contaminated area in the washed dishes is extracted based on the corrected area, and the cleaning performance score is calculated based on the total pixel area and the dishwasher performance test standard.

[0011] Furthermore, a method for obtaining the contaminant and the contaminated tableware includes:

[0012] Obtain the dishwasher performance test standard. According to the standard, take uncooked starchy stains and place them in a non-stick pan. Pour in cold water and milk in a 3:1 ratio to obtain a stain mixture. Heat the stain mixture to a boiling point and simmer it over low heat. Take 40% of the simmered stain mixture and place it in clean tableware to prepare undried contaminants.

[0013] Using 0.5% of the undried contaminant as the minimum dripping value, the undried contaminant was transferred using a dropper and dripped onto each clean tableware in a gradient manner, distributing the dripping points on the inner bottom and inner wall of the tableware. This process was repeated 3 times to prepare undried contaminated tableware. The undried contaminant and the undried contaminated tableware were then dried to obtain the contaminant and the contaminated tableware.

[0014] Further, a method for obtaining the washed contaminated product includes:

[0015] Based on the temperature and time parameters of the washing mode obtained from the dishwasher under test, the contaminants and contaminated tableware are simulated to be washed without spraying by static baking according to the temperature and time parameters, and the washed products are obtained.

[0016] Based on the temperature and time parameters of the pre-wash, main wash, rinse, and drying modes obtained from the dishwasher under test, cold water was added to the contaminated tableware so that the cold water covered the contaminants on the inner surface of the tableware. The drying device was set according to the temperature and time parameters of the pre-wash, main wash, and rinse modes. The contaminants and the contaminated tableware with added cold water were placed into the drying devices set in different ways for settling. After settling, the contaminated tableware was taken out, the water in the contaminated tableware was poured out, and it was set to stand together with the contaminants under the temperature and time parameters of drying. After settling, the washed contaminants and washed tableware were obtained and regarded as washed products.

[0017] Furthermore, the method for obtaining the three-dimensional fluorescence spectrum includes:

[0018] The surface layer of the washed product was extracted to remove contaminants. The sample surface layer was then filled into the sample cell of a fluorescence spectrophotometer and placed inside the front support. The incident angle was set to 30 degrees. The excitation wavelength range of the three-dimensional fluorescence spectrum was 200-600 nm with a step size of 10 nm, and the emission wavelength range was 200-700 nm with a step size of 10 nm. The slit width of the excitation and emission monochromators was 5 nm, the detector voltage was 700 V, and the scan rate was 30000 nm / min. The three-dimensional fluorescence spectrum of the sample surface layer and the maximum fluorophore in the fluorescence intensity were obtained. The optimal excitation wavelength and the optimal emission wavelength were obtained based on the maximum fluorophore and used as the optimal wavelength parameters.

[0019] Furthermore, the method for constructing the fluorescence imaging system includes:

[0020] A fluorescence imaging system is constructed based on the optimal wavelength parameters of three-dimensional fluorescence spectroscopy. The system includes an industrial color camera 1, an LED ultraviolet light source 2, a black sample plate 3, tableware 4, a light source controller 5, and a computer 6.

[0021] The tableware 4 is placed on the upper surface of the black sample plate 3. An LED ultraviolet light source 2 is set above the black sample plate 3. An industrial color camera 1 is vertically arranged above the LED ultraviolet light source 2. The lens of the industrial color camera 1 faces the tableware 4 through the top opening of the LED ultraviolet light source 2. The industrial color camera 1 and the computer 6 are connected by a data cable. The LED ultraviolet light source 2 is driven by a light source controller 5.

[0022] The LED ultraviolet light source 2 is a ring light source with an outer diameter of 120mm and an inner diameter of 50mm. The luminous area of ​​the LED ultraviolet light source 2 is larger than the maximum diameter of the tableware 4. The industrial color camera 1 is embedded with a bandpass filter with a full width at half maximum (FWHM) of 170nm.

[0023] Further, the method for obtaining the fluorescence image includes:

[0024] Based on the washed products, the washed tableware is used as the sample of tableware 4 to obtain sample 4. Sample 4 is placed on the upper surface of the black sample plate 3 of the fluorescence imaging system with the opening of sample 4 facing upward and aligned with the coaxial center of the LED ultraviolet light source 2. The distance between the lower end of the LED ultraviolet light source 2 and the upper surface of the black sample plate 3 is adjusted to be greater than the height of sample 4. The fluorescence image of sample 4 is acquired using the industrial color camera 1 to obtain the fluorescence image of the washed tableware.

[0025] Furthermore, the method for obtaining the corrected image includes:

[0026] A fluorescence image of the washed tableware is acquired. The fluorescence image is then converted to HSV color space, the saturation channel is extracted, and background interference is suppressed to obtain a de-interference image. Based on the de-interference image, the pixels on the surface of the washed tableware are assigned zero values, and binarization is performed according to the fluorescence index. The connected regions on the binarized de-interference image are calculated to obtain the contaminated areas. The fluorescence index calculation formula is as follows:

[0027] ;

[0028] in The fluorescence index, G , B , R These represent the image mean values ​​for the green, red, and blue color channels, respectively.

[0029] The pixel area of ​​the contaminated area and the vertex coordinates and pixel area of ​​the minimum bounding rectangle are extracted. Physical parameters are obtained based on the inner radius and internal height of the washed tableware. These physical parameters are used to model the contaminated contact surface. The equation of the surface model is as follows:

[0030] ;

[0031] in R Let the inner radius be , H For internal height, q The curvature of the contaminated contact surface of the tableware;

[0032] The slope of the tangent line is calculated by differentiating the equation of the surface model, and then the arc length from the vertex of the contaminated region to the image center 0 is calculated using the slope of the tangent line. The calculation formula is as follows:

[0033] ;

[0034] The vertex coordinates are corrected based on the arc length, and the correction formula is as follows:

[0035] ;

[0036] in Let x be the horizontal correction coordinate of the smallest bounding rectangle in the nth contaminated region. The vertical correction coordinate of the smallest bounding rectangle in the nth contaminated region;

[0037] The correction area is calculated based on the correction coordinates, and the formula for calculating the correction area is as follows:

[0038] ;

[0039] in Let n be the corrected area of ​​the nth contaminated area. The pixel area of ​​the contaminated region is given by the length of the minimum bounding rectangle. Multiply by width Calculated It is the pixel area of ​​the smallest bounding rectangle of the nth contaminated region. It is the pixel area of ​​the smallest bounding rectangle after the nth contaminated area is corrected.

[0040] Furthermore, the method for obtaining the curvature of the contaminated contact surface includes:

[0041] Obtain fluorescence images and physical parameters of washed tableware. Use the 80% to 100% range of the inner radius from the physical parameters as the cross-sectional radius. Extract contamination sampling points within the inner wall height corresponding to the cross-sectional radius. Calculate the curvature of the contaminated contact surface and the local curvature of the contaminated area based on the contamination sampling points. The calculation formula is as follows:

[0042] ;

[0043] in To improve the curvature of the contamination contact surface, R Let the inner radius be , H For internal height, Where is the cross-sectional radius, For contamination sampling points, The coefficient of the quadratic term;

[0044] Based on the coefficient of the quadratic term The reference height is obtained by reverse calculation, where Using the reference height as a reference, a weighting factor is calculated based on the reference height. The formula for the weighting factor is:

[0045] ;

[0046] in Let the initial value be the differential curvature. The weighting factor for contaminated sampling point i;

[0047] The differential curvature is iteratively corrected using the formula for the curvature of the contaminated contact surface by using weighting factors. The final value of the differential curvature is used as the local curvature of the contaminated area. The area where the absolute value of the local curvature of the contaminated area is greater than the interquartile value is obtained. The absolute value of the local curvature of the contaminated area is used as the correction weight to perform a second correction on the correction area through weighted proportional correction.

[0048] Furthermore, the method for calculating the cleaning performance score based on the fluorescence image includes:

[0049] The pixel area of ​​the corresponding calibrated area in the washed tableware is obtained based on different temperature and time parameters. The pixel area is converted into physical area using a preset pinhole camera model to obtain the total area of ​​contaminants. The physical area scale is millimeters. The total area of ​​contaminants is scored according to the dishwasher performance test standard. If the total area of ​​contaminants is 0 square millimeters, the score is 5. If the total area of ​​contaminants is greater than 0 and less than or equal to 4, the score is 4. If the total area of ​​contaminants is greater than 4 and less than or equal to 20, the score is 3. If the total area of ​​contaminants is greater than 20 and less than or equal to 50, the score is 2. If the total area of ​​contaminants is greater than 50 and less than or equal to 200, the score is 1. If the total area of ​​contaminants is greater than 200, the score is 0.

[0050] A second aspect of the present invention provides a dishwasher cleaning performance testing system based on fluorescence imaging, comprising:

[0051] Washing simulation module: used to obtain temperature and time parameters of the washing mode based on the dishwasher under test, and to perform spray-free simulated washing of contaminants and contaminated tableware by static baking according to the temperature and time parameters, to obtain the washed products;

[0052] Fluorescence image acquisition module: used to determine the optimal wavelength parameters for fluorescence imaging based on the three-dimensional fluorescence spectrum of washed contaminants, build a fluorescence imaging system based on the optimal wavelength parameters, and use the fluorescence imaging system to acquire fluorescence images of washed tableware;

[0053] The contamination data acquisition module is used to model the contamination contact surface based on the physical parameters of the washed tableware, and use the surface model to correct the contamination area in the fluorescence image to obtain the corrected area.

[0054] Cleaning performance calculation module: used to extract the total pixel area of ​​the contaminated area in the washed dishes based on the correction area, and calculate the cleaning performance score according to the total pixel area and the dishwasher performance test standard.

[0055] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects:

[0056] The method for testing the cleaning performance of dishwashers under fluorescent imaging provided by this invention solves the problem of difficulty in identifying contaminants when their color is similar to that of the tableware surface by using fluorescence technology. It provides a pixel-level scoring method for evaluating cleaning performance, which is more objective than manual scoring. This method helps to promote the intelligent development of dishwasher cleaning performance testing and is of great significance to improving the scientific nature and international competitiveness of my country's national standards. Attached Figure Description

[0057] Figure 1 This is a flowchart illustrating the steps of a dishwasher cleaning performance testing method based on fluorescence imaging in an embodiment of the present invention.

[0058] Figure 2 The three-dimensional fluorescence spectrum of the oat contaminant sample after washing;

[0059] Figure 3 This is a schematic diagram of the dishwasher cleaning performance detection system based on fluorescence imaging according to the present invention.

[0060] Figure 4a The smallest bounding rectangle of the contaminated area before and after correction;

[0061] Figure 4b The images show the area of ​​oat contamination before and after correction in 15 washed samples of contaminated rice bowls.

[0062] Figure 5 The cleaning performance score is calculated based on fluorescence images and the cleaning performance score is calculated based on GB / T 20290-2024.

[0063] The attached figures are labeled as follows: 1. Industrial color camera, 2. LED ultraviolet light source, 3. Black sample plate, 4. Tableware, 5. Light source controller, 6. Computer. Detailed Implementation

[0064] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0065] Reference Figure 1 As shown, this invention provides a method for detecting the cleaning performance of a dishwasher based on fluorescence imaging, comprising:

[0066] Based on the temperature and time parameters of the washing mode obtained from the dishwasher under test, the contaminants and contaminated tableware are simulated to be washed without spraying by static baking according to the temperature and time parameters, and the washed products are obtained.

[0067] The optimal wavelength parameters for fluorescence imaging are determined based on the three-dimensional fluorescence spectrum of the washed contaminants. A fluorescence imaging system is then built based on the optimal wavelength parameters, and fluorescence images of the washed tableware are acquired using the fluorescence imaging system.

[0068] The contaminated contact surface is modeled based on the physical parameters of the washed tableware, and the contaminated area in the fluorescence image is corrected using the surface model to obtain the corrected area.

[0069] The total pixel area of ​​the contaminated area in the washed dishes is extracted based on the corrected area, and the cleaning performance score is calculated based on the total pixel area and the dishwasher performance test standard.

[0070] In the actual evaluation, the GB / T 20290-2024 standard for performance testing of household electric dishwashers was used. Oatmeal was used as a starch stain. 50g of uncooked oat flakes were placed in a non-stick pan, and 750mL of cold water and 250mL of milk were poured into the pan to obtain a stain mixture. The mixture was heated to the boiling point and then simmered for 10 minutes. 20g of the mixture was then poured into a clean rice bowl. The clean rice bowl was a deep-mouthed dish with an inner radius of 56mm and an internal height of 41mm to prepare undried contaminants. 0.1g, 0.2g, 0.3g, 0.4g, and 0.5g of the undried contaminants were transferred with a dropper and dripped onto the inner wall of 5 clean rice bowls. The drips should be distributed on the bottom and inner wall of the bowls to prepare undried contaminated rice bowls. The above steps were repeated 3 times to prepare a total of 15 undried contaminated rice bowls.

[0071] The undried contaminant and the undried contaminated rice bowl were placed in an oven at 80°C for 2 hours to dry, thus obtaining the contaminant and the contaminated rice bowl respectively.

[0072] In the actual evaluation, the temperature and time settings of the energy-saving wash mode of a commercially available dishwasher were extracted. The energy-saving wash mode of the dishwasher surveyed had a pre-wash temperature of 25℃, a pre-wash time of 20min, a main wash temperature of 50℃, a main wash time of 90min, a rinsing temperature of 52℃, a rinsing time of 25min, a drying temperature of 38.5℃, and a drying time of 70min. Cold water was added to the contaminated rice bowl, covering all the contaminants on the inner surface, while no cold water was added to the contaminants. The contaminants and the contaminated rice bowl were placed in an oven to simulate the washing process. The bowl was left to stand at oven temperatures of 25℃, 50℃, and 52℃ for 20min, 90min, and 25min, respectively. Then the water in the contaminated rice bowl was poured out, and the bowl was left to stand at 38.5℃ for 70min along with the contaminants in the oven. This yielded a sample of washed rice bowl and a sample of washed contaminants.

[0073] In the actual assessment, the surface layer of the washed contaminant sample was used to acquire three-dimensional fluorescence spectra using a Hitachi F-7000 fluorescence spectrophotometer. A front-surface structure was employed, with an incident angle of 30°. The excitation wavelength range of the three-dimensional fluorescence spectrum was 200-600 nm with a step size of 10 nm, and the emission wavelength range was also 200-700 nm with a step size of 10 nm. The slit width of the excitation and emission monochromators was 5 nm, the detector voltage was 700 V, and the scan rate was 30000 nm / min. The sample surface layer was filled into the sample cell and placed inside the front-surface support before three-dimensional fluorescence spectroscopy acquisition began. The three-dimensional fluorescence spectrum obtained from the washed oat contaminant sample is shown below. Figure 2 As shown, the washed oat contaminant sample contained three main fluorescence emission peaks, with corresponding excitation and emission wavelengths of 300 nm and 320 nm, 335 nm and 430 nm, and 450 nm and 510 nm, respectively. The excitation and emission wavelengths of the fluorophore with the highest fluorescence intensity were determined to be approximately 335 nm and 430 nm.

[0074] In the actual evaluation, considering the cost of the ultraviolet LED light source, a center wavelength of 365nm was selected for the LED light source in the fluorescence imaging system. Based on the selected 365nm LED light source, a dishwasher cleaning performance testing system based on fluorescence imaging was built. Figure 3 The system shown acquires fluorescence images of a washed rice bowl sample 4. The washed rice bowl sample 4 is placed on a black sample plate 3, which is coaxial with the LED ultraviolet light source 2. The distance between the industrial color camera 1 and the black sample plate 3 is adjusted to 303mm, and the distance between the LED ultraviolet light source 2 and the black sample plate 3 is adjusted to 270mm. Fluorescence images of sample 4 are acquired using the industrial color camera 1. Each image has a size of 1280×1024 and is referred to as the original image.

[0075] In the actual evaluation, the original image was converted to grayscale and binarized. Background removal was then performed on the original image, which was then converted to an HSV color channel image. Specular reflection from the LED UV light source 2 on the surface of sample 4 was removed, resulting in an image with light source interference removed. Using this image, the contaminated area was extracted by setting pixels belonging to the bowl surface to 0, and the fluorescence index was then used to further refine the analysis. ExG The image is binarized, and the connected regions on the binarized image are calculated, which are the contaminated regions. The curvature of the contaminated contact surface is 6, the inner radius of the rice bowl is 56mm, and the inner height is 41mm. The correction area is obtained based on the contaminated area and the surface model of the rice bowl. The minimum bounding rectangle of the contaminated region before and after correction is shown in the figure. Figure 4a As shown; the areas of oat contamination in 15 washed, contaminated rice bowls before and after correction are as follows. Figure 4bAs shown, the cross-sectional radius is extracted from a fluorescence image of a washed rice bowl. Pixels in the contaminated area within the interval The inner wall height is between 0.8R and R, specifically between 44.8mm and 56mm. A total of 100 valid sampling points were extracted. Based on the pollution sampling points, the curvature of the pollution contact surface and the local curvature of the pollution area were calculated, and the curvature of the pollution contact surface was obtained. The value is 5.8, and the coefficient of the quadratic term is obtained. It is 0.35, combined with The reference height is obtained by reverse calculation, and the initial value of the differential curvature is obtained based on the reference height. for Iteratively correct the differential curvature to obtain The value is 5.9, and the final value of the differential curvature is... Since the rice bowl has been washed, the correction area is 25mm. 2 The region with a local curvature absolute value greater than the interquartile range is identified. The local curvature absolute value is then used as a correction weight to perform a secondary correction on the area to be corrected using a weighted proportional correction. ;

[0076] In practical evaluation, the cross-sectional radius of a standard rice bowl is extracted from the fluorescence image. Pixels in the contaminated area within the interval The inner wall height is between 0.7R and R, specifically between 42mm and 60mm. The sampling depth was 30 mm, and a total of 80 valid sampling points were extracted. The curvature of the contaminated contact surface was calculated. The value is 4.2, and the coefficient of the quadratic term is obtained. It is 0.22, combined with The reference height is obtained by reverse calculation, and the initial value of the differential curvature is obtained based on the reference height. for Iteratively correct the differential curvature to obtain The value is 4.3, and the final value of the differential curvature is... Since the rice bowl has been washed, the correction area is 20mm. 2 The region with a local curvature absolute value greater than the interquartile range is identified. The local curvature absolute value is then used as a correction weight to perform a secondary correction on the area to be corrected using a weighted proportional correction. ;

[0077] In practical evaluation, a pinhole camera model is used to convert pixel area into physical area. According to Table 3 of GB / T20290-2024, the rice bowl is scored based on the total area of ​​contaminants. A score is given when the total area of ​​contaminants is 0 mm. 2The score is 5 when the total area of ​​contaminants is greater than 0 and less than or equal to 4, 4 when the total area of ​​contaminants is greater than 4 and less than or equal to 20, 3 when the total area of ​​contaminants is greater than 20 and less than or equal to 50, 2 when the total area of ​​contaminants is greater than 50 and less than or equal to 200, 1 when the total area of ​​contaminants is greater than 50 and less than or equal to 200, and 0 when the total area of ​​contaminants is greater than 200. The cleaning performance score calculated based on the fluorescence image and the cleaning performance score calculated based on GB / T 20290-2024 are shown in Figure 4.

[0078] In this embodiment, the method for obtaining the contaminant and the contaminated tableware includes:

[0079] Obtain the dishwasher performance test standard. According to the standard, take uncooked starchy stains and place them in a non-stick pan. Pour in cold water and milk in a 3:1 ratio to obtain a stain mixture. Heat the stain mixture to a boiling point and simmer it over low heat. Take 40% of the simmered stain mixture and place it in clean tableware to prepare undried contaminants.

[0080] Using 0.5% of the undried contaminant as the minimum dripping value, the undried contaminant was transferred using a dropper and dripped onto each clean tableware in a gradient manner, distributing the dripping points on the inner bottom and inner wall of the tableware. This process was repeated 3 times to prepare undried contaminated tableware. The undried contaminant and the undried contaminated tableware were then dried to obtain the contaminant and the contaminated tableware.

[0081] In this embodiment, the method for obtaining the washed contaminated product includes:

[0082] Based on the temperature and time parameters of the washing mode obtained from the dishwasher under test, the contaminants and contaminated tableware are simulated to be washed without spraying by static baking according to the temperature and time parameters, and the washed products are obtained.

[0083] Based on the temperature and time parameters of the pre-wash, main wash, rinse, and drying modes obtained from the dishwasher under test, cold water was added to the contaminated tableware so that the cold water covered the contaminants on the inner surface of the tableware. The drying device was set according to the temperature and time parameters of the pre-wash, main wash, and rinse modes. The contaminants and the contaminated tableware with added cold water were placed into the drying devices set in different ways for settling. After settling, the contaminated tableware was taken out, the water in the contaminated tableware was poured out, and it was set to stand together with the contaminants under the temperature and time parameters of drying. After settling, the washed contaminants and washed tableware were obtained and regarded as washed products.

[0084] In this embodiment, the method for obtaining the three-dimensional fluorescence spectrum includes:

[0085] The surface layer of the washed product was extracted to remove contaminants. The sample surface layer was then filled into the sample cell of a fluorescence spectrophotometer and placed inside the front support. The incident angle was set to 30 degrees. The excitation wavelength range of the three-dimensional fluorescence spectrum was 200-600 nm with a step size of 10 nm, and the emission wavelength range was 200-700 nm with a step size of 10 nm. The slit width of the excitation and emission monochromators was 5 nm, the detector voltage was 700 V, and the scan rate was 30000 nm / min. The three-dimensional fluorescence spectrum of the sample surface layer and the maximum fluorophore in the fluorescence intensity were obtained. The optimal excitation wavelength and the optimal emission wavelength were obtained based on the maximum fluorophore and used as the optimal wavelength parameters.

[0086] In this embodiment, the method for constructing the fluorescence imaging system includes:

[0087] A fluorescence imaging system is constructed based on the optimal wavelength parameters of three-dimensional fluorescence spectroscopy. The system includes an industrial color camera 1, an LED ultraviolet light source 2, a black sample plate 3, tableware 4, a light source controller 5, and a computer 6.

[0088] The tableware 4 is placed on the upper surface of the black sample plate 3. An LED ultraviolet light source 2 is set above the black sample plate 3. An industrial color camera 1 is vertically arranged above the LED ultraviolet light source 2. The lens of the industrial color camera 1 faces the tableware 4 through the top opening of the LED ultraviolet light source 2. The industrial color camera 1 and the computer 6 are connected by a data cable. The LED ultraviolet light source 2 is driven by a light source controller 5.

[0089] The LED ultraviolet light source 2 is a ring light source with an outer diameter of 120mm and an inner diameter of 50mm. The luminous area of ​​the LED ultraviolet light source 2 is larger than the maximum diameter of the tableware 4. The industrial color camera 1 is embedded with a bandpass filter with a full width at half maximum (FWHM) of 170nm.

[0090] In this embodiment, the method for obtaining the fluorescence image includes:

[0091] Based on the washed products, the washed tableware is used as the sample of tableware 4 to obtain sample 4. Sample 4 is placed on the upper surface of the black sample plate 3 of the fluorescence imaging system with the opening of sample 4 facing upward and aligned with the coaxial center of the LED ultraviolet light source 2. The distance between the lower end of the LED ultraviolet light source 2 and the upper surface of the black sample plate 3 is adjusted to be greater than the height of sample 4. The fluorescence image of sample 4 is acquired using the industrial color camera 1 to obtain the fluorescence image of the washed tableware.

[0092] In this embodiment, the method for obtaining the corrected image includes:

[0093] A fluorescence image of the washed tableware is acquired. The fluorescence image is then converted to HSV color space, the saturation channel is extracted, and background interference is suppressed to obtain a de-interference image. Based on the de-interference image, the pixels on the surface of the washed tableware are assigned zero values, and binarization is performed according to the fluorescence index. The connected regions on the binarized de-interference image are calculated to obtain the contaminated areas. The fluorescence index calculation formula is as follows:

[0094] ;

[0095] in The fluorescence index, G , B , R These represent the image mean values ​​for the green, red, and blue color channels, respectively.

[0096] The pixel area of ​​the contaminated area and the vertex coordinates and pixel area of ​​the minimum bounding rectangle are extracted. Physical parameters are obtained based on the inner radius and internal height of the washed tableware. These physical parameters are used to model the contaminated contact surface. The equation of the surface model is as follows:

[0097] ;

[0098] in R Let the inner radius be , H For internal height, q The curvature of the contaminated contact surface of the tableware;

[0099] The slope of the tangent line is calculated by differentiating the equation of the surface model, and then the arc length from the vertex of the contaminated region to the image center 0 is calculated using the slope of the tangent line. The calculation formula is as follows:

[0100] ;

[0101] The vertex coordinates are corrected based on the arc length, and the correction formula is as follows:

[0102] ;

[0103] in Let x be the horizontal correction coordinate of the smallest bounding rectangle in the nth contaminated region. The vertical correction coordinate of the smallest bounding rectangle in the nth contaminated region;

[0104] The correction area is calculated based on the correction coordinates, and the formula for calculating the correction area is as follows:

[0105] ;

[0106] in Let n be the corrected area of ​​the nth contaminated area. The pixel area of ​​the contaminated region is given by the length of the minimum bounding rectangle. Multiply by width Calculated It is the pixel area of ​​the smallest bounding rectangle of the nth contaminated region. It is the pixel area of ​​the smallest bounding rectangle after the nth contaminated area is corrected.

[0107] In this embodiment, the method for obtaining the curvature of the contaminated contact surface includes:

[0108] Obtain fluorescence images and physical parameters of washed tableware. Use the 80% to 100% range of the inner radius from the physical parameters as the cross-sectional radius. Extract contamination sampling points within the inner wall height corresponding to the cross-sectional radius. Calculate the curvature of the contaminated contact surface and the local curvature of the contaminated area based on the contamination sampling points. The calculation formula is as follows:

[0109] ;

[0110] in To improve the curvature of the contamination contact surface, R Let the inner radius be , H For internal height, Where is the cross-sectional radius, For contamination sampling points, The coefficient of the quadratic term;

[0111] Based on the coefficient of the quadratic term The reference height is obtained by reverse calculation, where Using the reference height as a reference, a weighting factor is calculated based on the reference height. The formula for the weighting factor is:

[0112] ;

[0113] in Let the initial value be the differential curvature. The weighting factor for contaminated sampling point i;

[0114] The differential curvature is iteratively corrected using the formula for the curvature of the contaminated contact surface by using weighting factors. The final value of the differential curvature is used as the local curvature of the contaminated area. The area where the absolute value of the local curvature of the contaminated area is greater than the interquartile value is obtained. The absolute value of the local curvature of the contaminated area is used as the correction weight to perform a second correction on the correction area through weighted proportional correction.

[0115] In this embodiment, the method for calculating the cleaning performance score based on the fluorescence image includes:

[0116] The pixel area of ​​the corresponding calibrated area in the washed tableware is obtained based on different temperature and time parameters. The pixel area is converted into physical area using a preset pinhole camera model to obtain the total area of ​​contaminants. The physical area scale is millimeters. The total area of ​​contaminants is scored according to the dishwasher performance test standard. If the total area of ​​contaminants is 0 square millimeters, the score is 5. If the total area of ​​contaminants is greater than 0 and less than or equal to 4, the score is 4. If the total area of ​​contaminants is greater than 4 and less than or equal to 20, the score is 3. If the total area of ​​contaminants is greater than 20 and less than or equal to 50, the score is 2. If the total area of ​​contaminants is greater than 50 and less than or equal to 200, the score is 1. If the total area of ​​contaminants is greater than 200, the score is 0.

[0117] A second aspect of the present invention also provides a dishwasher cleaning performance detection system based on fluorescence imaging, comprising:

[0118] Washing simulation module: used to obtain temperature and time parameters of the washing mode based on the dishwasher under test, and to perform spray-free simulated washing of contaminants and contaminated tableware by static baking according to the temperature and time parameters, to obtain the washed products;

[0119] Fluorescence image acquisition module: used to determine the optimal wavelength parameters for fluorescence imaging based on the three-dimensional fluorescence spectrum of washed contaminants, build a fluorescence imaging system based on the optimal wavelength parameters, and use the fluorescence imaging system to acquire fluorescence images of washed tableware;

[0120] The contamination data acquisition module is used to model the contamination contact surface based on the physical parameters of the washed tableware, and use the surface model to correct the contamination area in the fluorescence image to obtain the corrected area.

[0121] Cleaning performance calculation module: used to extract the total pixel area of ​​the contaminated area in the washed dishes based on the correction area, and calculate the cleaning performance score according to the total pixel area and the dishwasher performance test standard.

[0122] The above description is merely an example and illustration of the structure of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the structure of the invention or exceed the scope defined in the claims, all of which should fall within the protection scope of the present invention.

Claims

1. A method for detecting the cleaning performance of a dishwasher based on fluorescence imaging, characterized in that, Includes the following steps: Based on the temperature and time parameters of the washing mode obtained from the dishwasher under test, the contaminants and contaminated tableware are simulated to be washed without spraying by static baking according to the temperature and time parameters, and the washed products are obtained. The optimal wavelength parameters for fluorescence imaging are determined based on the three-dimensional fluorescence spectrum of the washed contaminants. A fluorescence imaging system is then built based on the optimal wavelength parameters, and fluorescence images of the washed tableware are acquired using the fluorescence imaging system. A contaminated contact surface is modeled based on the physical parameters of the washed tableware. The contaminated area in the fluorescence image is corrected using the surface model to obtain the corrected area; including: A fluorescence image of the washed tableware is acquired. The fluorescence image is then converted to HSV color space, the saturation channel is extracted, and background interference is suppressed to obtain a de-interference image. Based on the de-interference image, the pixels on the surface of the washed tableware are assigned zero values, and binarization is performed according to the fluorescence index. The connected regions on the binarized de-interference image are calculated to obtain the contaminated areas. The fluorescence index calculation formula is as follows: ； in The fluorescence index, , , These represent the image mean values ​​for the green, red, and blue color channels, respectively. The pixel area of ​​the contaminated area and the vertex coordinates and pixel area of ​​the minimum bounding rectangle are extracted. Physical parameters are obtained based on the inner radius and internal height of the washed tableware. These physical parameters are used to model the contaminated contact surface. The equation of the surface model is as follows: ； in Let the inner radius be , For internal height, The curvature of the contaminated contact surface of the tableware; The slope of the tangent line is calculated by differentiating the equation of the surface model, and then the arc length from the vertex of the contaminated region to the image center O is calculated using the slope of the tangent line. The calculation formula is as follows: ； The vertex coordinates are corrected based on the arc length, and the correction formula is as follows: ； in For the first The horizontal correction coordinates of the smallest bounding rectangle in each polluted area. For the first The longitudinal correction coordinates of the smallest bounding rectangle in each polluted area; The correction area is calculated based on the correction coordinates, and the formula for calculating the correction area is as follows: ； in For the first Corrected area of ​​each polluted area The pixel area of ​​the contaminated region is given by the length of the minimum bounding rectangle. Multiply by width Calculated It is the first The pixel area of ​​the smallest bounding rectangle of a contaminated region. It is the first The pixel area of ​​the smallest bounding rectangle after correction of each contaminated area; including: Obtain fluorescence images and physical parameters of washed tableware. Use the 80% to 100% range of the inner radius from the physical parameters as the cross-sectional radius. Extract contamination sampling points within the inner wall height corresponding to the cross-sectional radius. Calculate the curvature of the contaminated contact surface and the local curvature of the contaminated area based on the contamination sampling points. The calculation formula is as follows: ； in To improve the curvature of the contamination contact surface, Let the inner radius be , For internal height, Where is the cross-sectional radius, For contamination sampling points, The coefficient of the quadratic term; Based on the coefficient of the quadratic term The reference height is obtained by reverse calculation, where Using the reference height as a reference, a weighting factor is calculated based on the reference height. The formula for the weighting factor is: ； in Let the initial value be the differential curvature. The weighting factor for contaminated sampling point i; The differential curvature is iteratively corrected using the formula for the curvature of the contaminated contact surface by using weighting factors. The final value of the differential curvature is used as the local curvature of the contaminated area. The area where the absolute value of the local curvature of the contaminated area is greater than the interquartile value is obtained. The absolute value of the local curvature of the contaminated area is used as the correction weight to perform a second correction on the correction area through weighted proportional correction. The total pixel area of ​​the contaminated area in the washed dishes is extracted based on the corrected area, and the cleaning performance score is calculated based on the total pixel area and the dishwasher performance test standard.

2. The method for detecting the cleaning performance of a dishwasher based on fluorescence imaging according to claim 1, characterized in that, A method for obtaining the contaminant and the contaminated tableware includes: Obtain the dishwasher performance test standard. According to the standard, take uncooked starchy stains and place them in a non-stick pan. Pour in cold water and milk in a 3:1 ratio to obtain a stain mixture. Heat the stain mixture to a boiling point and simmer it over low heat. Take 40% of the simmered stain mixture and place it in clean tableware to prepare undried contaminants. Using 0.5% of the undried contaminant as the minimum dripping value, the undried contaminant was transferred using a dropper and dripped onto each clean tableware in a gradient manner, distributing the dripping points on the inner bottom and inner wall of the tableware. This process was repeated 3 times to prepare undried contaminated tableware. The undried contaminant and the undried contaminated tableware were then dried to obtain the contaminant and the contaminated tableware.

3. The method for detecting the cleaning performance of a dishwasher based on fluorescence imaging according to claim 1, characterized in that, A method for obtaining the washed contaminated product includes: Based on the temperature and time parameters of the washing mode obtained from the dishwasher under test, the contaminants and contaminated tableware are simulated to be washed without spraying by static baking according to the temperature and time parameters, and the washed products are obtained. Based on the temperature and time parameters of the pre-wash, main wash, rinse, and drying modes obtained from the dishwasher under test, cold water was added to the contaminated tableware so that the cold water covered the contaminants on the inner surface of the tableware. The drying device was set according to the temperature and time parameters of the pre-wash, main wash, and rinse modes. The contaminants and the contaminated tableware with added cold water were placed into the drying devices of different settings in sequence for settling. After settling, the contaminated tableware was taken out, the water in the contaminated tableware was poured out, and it was set to stand together with the contaminants at the temperature and time parameters of drying. After settling, the washed contaminants and washed tableware were obtained and regarded as washed products.

4. The method for detecting the cleaning performance of a dishwasher based on fluorescence imaging according to claim 1, characterized in that, The method for obtaining the three-dimensional fluorescence spectrum includes: The surface layer of the washed product was extracted to remove contaminants. The sample surface layer was then filled into the sample cell of a fluorescence spectrophotometer and placed inside the front support. The incident angle was set to 30 degrees. The excitation wavelength range of the three-dimensional fluorescence spectrum was 200-600 nm with a step size of 10 nm, and the emission wavelength range was 200-700 nm with a step size of 10 nm. The slit width of the excitation and emission monochromators was 5 nm, the detector voltage was 700 V, and the scan rate was 30000 nm / min. The three-dimensional fluorescence spectrum of the sample surface layer and the maximum fluorophore in the fluorescence intensity were obtained. The optimal excitation wavelength and the optimal emission wavelength were obtained based on the maximum fluorophore and used as the optimal wavelength parameters.

5. The method for detecting the cleaning performance of a dishwasher based on fluorescence imaging according to claim 1, characterized in that, The method for constructing the fluorescence imaging system includes: A fluorescence imaging system was constructed based on the optimal wavelength parameters of three-dimensional fluorescence spectroscopy. The system includes an industrial color camera, an LED ultraviolet light source, a black sample plate, tableware, a light source controller, and a computer. The tableware is placed on the upper surface of the black sample plate. An LED ultraviolet light source is set above the black sample plate. An industrial color camera is vertically arranged above the LED ultraviolet light source. The lens of the industrial color camera faces the tableware through the top opening of the LED ultraviolet light source. The industrial color camera and the computer are connected by a data cable. The LED ultraviolet light source is driven by the light source controller 5. The LED ultraviolet light source is a ring light source with an outer diameter of 120mm and an inner diameter of 50mm. The luminous area of ​​the LED ultraviolet light source is larger than the maximum diameter of the tableware. The industrial color camera is embedded with a bandpass filter with a full width at half maximum (FWHM) of 170nm.

6. The method for detecting the cleaning performance of a dishwasher based on fluorescence imaging according to claim 5, characterized in that, A method for obtaining the fluorescence image includes: Based on the washed products, the washed tableware is used as a sample of tableware. The sample is placed on the upper surface of the black sample plate of the fluorescence imaging system with the sample opening facing upward and aligned coaxially with the LED ultraviolet light source. The distance between the lower end of the LED ultraviolet light source and the upper surface of the black sample plate is adjusted to be greater than the height of the sample. The fluorescence image of the sample is acquired using an industrial color camera to obtain the fluorescence image of the washed tableware.

7. The method for detecting the cleaning performance of a dishwasher based on fluorescence imaging according to claim 1, characterized in that, A method for calculating a cleaning performance score based on the fluorescence image includes: The pixel area of ​​the corresponding calibrated area in the washed tableware is obtained based on different temperature and time parameters. The pixel area is converted into physical area using a preset pinhole camera model to obtain the total area of ​​contaminants. The physical area scale is millimeters. The total area of ​​contaminants is scored according to the dishwasher performance test standard. If the total area of ​​contaminants is 0 square millimeters, the score is 5. If the total area of ​​contaminants is greater than 0 and less than or equal to 4, the score is 4. If the total area of ​​contaminants is greater than 4 and less than or equal to 20, the score is 3. If the total area of ​​contaminants is greater than 20 and less than or equal to 50, the score is 2. If the total area of ​​contaminants is greater than 50 and less than or equal to 200, the score is 1. If the total area of ​​contaminants is greater than 200, the score is 0.

8. A dishwasher cleaning performance testing system based on fluorescence imaging, used to execute the dishwasher cleaning performance testing method based on fluorescence imaging as described in any one of claims 1 to 7, characterized in that, The system includes: Washing simulation module: used to obtain temperature and time parameters of the washing mode based on the dishwasher under test, and to perform spray-free simulated washing of contaminants and contaminated tableware by static baking according to the temperature and time parameters, to obtain the washed products; Fluorescence image acquisition module: used to determine the optimal wavelength parameters for fluorescence imaging based on the three-dimensional fluorescence spectrum of washed contaminants, build a fluorescence imaging system based on the optimal wavelength parameters, and use the fluorescence imaging system to acquire fluorescence images of washed tableware; The contamination data acquisition module is used to model the contamination contact surface based on the physical parameters of the washed tableware, and use the surface model to correct the contamination area in the fluorescence image to obtain the corrected area. Cleaning performance calculation module: used to extract the total pixel area of ​​the contaminated area in the washed dishes based on the correction area, and calculate the cleaning performance score according to the total pixel area and the dishwasher performance test standard.

Images