A method and related device for detecting oven performance based on image processing

Abstract

The present invention discloses an oven performance detection method and related device based on image processing, which relates to the field of image processing technology. The method comprises: preprocessing a plurality of first surface images and second surface images obtained; performing minimum circumscribed rectangle calculation on the first and second surface images to obtain the respective corresponding minimum circumscribed rectangles; extracting the corresponding first and second size parameters to calculate the food volume; performing color feature extraction on the first and second surface images based on connected domain analysis to obtain the corresponding surface color features; performing food maturity calculation based on the respective food volumes and surface color features; performing oven baking performance stability detection based on the first and second food maturity; performing surface color difference analysis on the first and second surface color features based on the color co-occurrence matrix to perform oven baking performance accuracy detection. The present invention can effectively improve the reliability and comprehensiveness of oven performance detection.

Description

Technical Field

[0001] The present invention mainly relates to the technical field of image processing, and in particular to an oven performance detection method based on image processing and a related device. Background Art

[0002] Ovens are common cooking equipment in modern life. In order to ensure the quality of oven use, performance testing is required. In the performance testing of ovens, baking performance stability testing is a very important test. In the current oven performance stability testing, the maturity of each food baked by the oven according to the same baking settings is usually obtained, and the maturity of the food is compared with the standard food maturity. The oven performance stability is determined based on the comparison results. However, this method is only for the maturity of food with the same baking settings, and the maturity of food with different baking settings is not added to the oven performance stability test. The lack of consideration for the maturity of food with different baking settings will result in the oven performance stability test being not comprehensive enough and unable to reflect the overall stability performance of the oven. In the method of obtaining food maturity, the color and volume of the baked food are usually determined by relevant personnel to obtain food maturity, but this method is too dependent on the professional quality of the relevant personnel, so that the accuracy of the obtained food maturity cannot be effectively guaranteed, which leads to the inability to accurately detect the oven performance stability of the oven. At the same time, the detection of the baking performance accuracy of the oven is also a very important part. The baking performance accuracy reflects the control accuracy of the oven on the baking of food. At present, the color difference changes of the baked food are usually obtained through the target detection frame to perform baking performance accuracy detection, but this method has a large error in color difference analysis, resulting in a large deviation between the obtained color difference analysis results and the actual situation, and it is impossible to obtain reliable baking performance accuracy detection results. Summary of the invention

[0003] The purpose of the present invention is to overcome the deficiencies of the prior art. The present invention provides an oven performance detection method and related devices based on image processing, which can effectively improve the reliability and comprehensiveness of oven performance detection and achieve a more ideal effect of oven performance detection.

[0004] In order to solve the above technical problems, the present invention provides an oven performance detection method based on image processing, the method comprising:

[0005] Acquire a plurality of first surface images of different surfaces of each food baked in the oven according to different baking settings and a plurality of second surface images of different surfaces of each food baked in the oven according to the same baking settings, and preprocess the first surface images and the second surface images to obtain preprocessed first surface images and second surface images;

[0006] Calculating minimum bounding rectangles for a plurality of preprocessed first surface images and a plurality of preprocessed second surface images respectively, to obtain a first minimum bounding rectangle corresponding to each preprocessed first surface image and a second minimum bounding rectangle corresponding to each preprocessed second surface image;

[0007] extracting first size parameters corresponding to each preprocessed first surface image based on the first minimum circumscribed rectangle, extracting second size parameters corresponding to each preprocessed second surface image based on the second minimum circumscribed rectangle, and calculating a first food volume using a correction coefficient based on the first size parameters, and calculating a second food volume using the correction coefficient based on the second size parameters;

[0008] Based on the connected domain analysis, color features are extracted from a plurality of preprocessed first surface images and a plurality of preprocessed second surface images to obtain first surface color features corresponding to each preprocessed first surface image and second surface color features corresponding to each preprocessed second surface image;

[0009] Calculating the degree of maturity of food based on the first food volume and the first surface color feature to obtain a first degree of maturity of each food baked according to different baking settings, and calculating the degree of maturity of food based on the second food volume and the second surface color feature to obtain a second degree of maturity of each food baked according to the same baking settings;

[0010] Performing a baking performance stability test on the oven based on the first food maturity and the second food maturity to obtain a baking performance stability test result;

[0011] Based on the color co-occurrence matrix, surface color difference analysis is performed on each first surface color feature and each second surface color feature to obtain a first surface color difference analysis result and a second surface color difference analysis result, and based on the first surface color difference analysis result and the second surface color difference analysis result, the baking performance accuracy of the oven is detected to obtain a baking performance accuracy detection result.

[0012] Optionally, the step of respectively calculating the minimum bounding rectangle of a plurality of preprocessed first surface images and a plurality of preprocessed second surface images to obtain a first minimum bounding rectangle corresponding to each preprocessed first surface image and a second minimum bounding rectangle corresponding to each preprocessed second surface image includes:

[0013] Respectively extracting first food surface contours in each preprocessed first surface image and second food surface contours in each preprocessed second surface image;

[0014] Determine a first search direction range of each first food surface contour and a second search direction range of each second food surface contour, obtain a first reference line of each preprocessed first surface image based on the first search direction range, and obtain a second reference line of each preprocessed second surface image based on the second search direction range;

[0015] Obtaining a plurality of first convex hull points of each first food surface contour and a plurality of second convex hull points of each second food surface contour, calculating a plurality of corresponding first circumscribed rectangles based on the plurality of first convex hull points and the first reference line using a rotation matrix, and calculating a plurality of corresponding second circumscribed rectangles based on the plurality of second convex hull points and the second reference line using a rotation matrix;

[0016] Based on the boundingRect function, a first minimum bounding rectangle corresponding to each preprocessed first surface image among a plurality of first bounding rectangles and a second minimum bounding rectangle corresponding to each preprocessed second surface image among a plurality of second bounding rectangles are calculated.

[0017] Optionally, the extracting first size parameters corresponding to each preprocessed first surface image based on the first minimum circumscribed rectangle, extracting second size parameters corresponding to each preprocessed second surface image based on the second minimum circumscribed rectangle, and calculating the first food volume using a correction coefficient based on the first size parameter, and calculating the second food volume using the correction coefficient based on the second size parameter, includes:

[0018] Performing size parameter extraction processing on each preprocessed first surface image based on the first minimum circumscribed rectangle to obtain first size parameters, where the first size parameters include first length information, first width information, and first height information;

[0019] Performing size parameter extraction processing on each preprocessed second surface image based on the second minimum circumscribed rectangle to obtain second size parameters, where the second size parameters include second length information, second width information, and second height information;

[0020] Determine a first correction coefficient in a correction coefficient library based on the first size parameter, and determine a second correction coefficient in a correction coefficient library based on the second size parameter, wherein the correction coefficients in the correction coefficient library are calculated based on past measured data and past simulation data using fitting equations and interval boundaries;

[0021] A first food volume is calculated using a first correction factor based on the first size parameter, and a second food volume is calculated using a second correction factor based on the second size parameter.

[0022] Optionally, the color feature extraction is performed on a plurality of preprocessed first surface images and a plurality of preprocessed second surface images based on the connected domain analysis to obtain a first surface color feature corresponding to each preprocessed first surface image and a second surface color feature corresponding to each preprocessed second surface image, including:

[0023] Performing region segmentation processing on a plurality of preprocessed first surface images and a plurality of preprocessed second surface images respectively, to obtain first surface images after the region segmentation processing and second surface images after the region segmentation processing;

[0024] Based on the rgb2hsv function, the first surface image after each region segmentation process and the second surface image after each region segmentation process are converted from the RGB color space to the HSV color space;

[0025] A clustering method based on a genetic algorithm is used to extract cluster color histograms from the first surface image after each region segmentation process and the second surface image after each region segmentation process in the HSV color space, so as to obtain a first cluster color histogram corresponding to the first surface image after each region segmentation process and a second cluster color histogram corresponding to the second surface image after each region segmentation process;

[0026] Performing connected domain extraction on the first surface image after each region segmentation process and the second surface image after each region segmentation process to obtain corresponding first connected domains and second connected domains;

[0027] Based on the first connected domain, color features of the first cluster color histograms corresponding to the first surface image after each region segmentation process are merged to obtain a first surface color feature;

[0028] Based on the second connected domain, the second cluster color histograms corresponding to the second surface images after the segmentation of each region are merged to obtain the second surface color features.

[0029] Optionally, the calculating the food maturity based on the first food volume and the first surface color feature to obtain the first food maturity of each food baked according to different baking settings includes:

[0030] The volume difference between the first food volume and the unbaked food volume is calculated, and the volume difference and the first surface color feature are input into a maturity model to calculate the food maturity, thereby obtaining the first food maturity of each food baked according to different baking settings.

[0031] Optionally, the performing a baking performance stability test of the oven based on the first food maturity and the second food maturity to obtain a baking performance stability test result includes:

[0032] comparing the first food maturity of each food baked at different baking settings with the standard food maturity of the corresponding baking setting to obtain a comparison result;

[0033] Calculating a food maturity mean square deviation based on second food maturity of the food baked under the same baking settings;

[0034] Based on the comparison result and the mean square error of food maturity, the baking performance stability test of the oven is performed to obtain a baking performance stability test result.

[0035] Optionally, performing surface color difference analysis on each first surface color feature and each second surface color feature based on the color co-occurrence matrix to obtain a first surface color difference analysis result and a second surface color difference analysis result, and performing a baking performance accuracy test of the oven based on the first surface color difference analysis result and the second surface color difference analysis result to obtain a baking performance accuracy test result, including:

[0036] Convert the color space where each first surface color feature and each second surface color feature are located into the CIELAB color space, and calculate the three channel parameters of each first surface color feature in the CIELAB color space and the three channel parameters of each second surface color feature in the CIELAB color space;

[0037] Obtaining a first color co-occurrence matrix for each first surface color feature and a second color co-occurrence matrix for each second surface color feature;

[0038] performing color difference analysis between different surfaces of each food baked with different baking settings based on the first color co-occurrence matrix of each first surface color feature and three channel parameters in the CIELAB color space to obtain a first surface color difference analysis result;

[0039] Performing color difference analysis between different surfaces of each food baked with the same baking settings on the second color co-occurrence matrix of each second surface color feature and three channel parameters in the CIELAB color space to obtain a second surface color difference analysis result;

[0040] Based on the first surface color difference analysis result and the second surface color difference analysis result combined with a preset color difference judgment standard, the baking performance accuracy test of the oven is performed to obtain a baking performance accuracy test result.

[0041] In addition, the present invention also provides an oven performance detection device based on image processing, the device comprising:

[0042] An image preprocessing module: used to obtain a plurality of first surface images of different surfaces of each food baked in the oven according to different baking settings and a plurality of second surface images of different surfaces of each food baked in the oven according to the same baking settings, and preprocess the first surface images and the second surface images to obtain the preprocessed first surface images and the second surface images;

[0043] Minimum bounding rectangle calculation module: used to calculate the minimum bounding rectangle of a plurality of preprocessed first surface images and a plurality of preprocessed second surface images, respectively, to obtain a first minimum bounding rectangle corresponding to each preprocessed first surface image and a second minimum bounding rectangle corresponding to each preprocessed second surface image;

[0044] A food volume calculation module: used for extracting a first size parameter corresponding to each preprocessed first surface image based on the first minimum circumscribed rectangle, extracting a second size parameter corresponding to each preprocessed second surface image based on the second minimum circumscribed rectangle, and calculating a first food volume based on the first size parameter using a correction coefficient, and calculating a second food volume based on the second size parameter using a correction coefficient;

[0045] Color feature extraction module: used for respectively extracting color features from a plurality of preprocessed first surface images and a plurality of preprocessed second surface images based on connected domain analysis, to obtain first surface color features corresponding to each preprocessed first surface image and second surface color features corresponding to each preprocessed second surface image;

[0046] a food maturity calculation module: configured to calculate the food maturity based on the first food volume and the first surface color feature to obtain the first food maturity of each food baked according to different baking settings, and to calculate the food maturity based on the second food volume and the second surface color feature to obtain the second food maturity of each food baked according to the same baking settings;

[0047] A baking performance stability detection module: used to detect the baking performance stability of the oven based on the first food maturity and the second food maturity, and obtain a baking performance stability detection result;

[0048] Baking performance accuracy detection module: used to perform surface color difference analysis on each first surface color feature and each second surface color feature based on the color co-occurrence matrix, obtain the first surface color difference analysis result and the second surface color difference analysis result, and perform baking performance accuracy detection of the oven based on the first surface color difference analysis result and the second surface color difference analysis result to obtain the baking performance accuracy detection result.

[0049] In addition, the present invention also provides an electronic device, which includes a processor and a memory, wherein the memory is used to store instructions, and the processor is used to call the instructions in the memory so that the electronic device executes the above-mentioned oven performance detection method based on image processing.

[0050] In addition, the present invention also provides a computer-readable storage medium, which stores computer instructions. When the computer instructions are executed on an electronic device, the electronic device executes the above-mentioned oven performance detection method based on image processing.

[0051] In an embodiment of the present invention, the minimum circumscribed rectangle is calculated for several preprocessed first surface images and several preprocessed second surface images to extract corresponding size parameters, and the food volume is calculated according to the corresponding size parameters, so that the obtained food volume can be more accurate. Based on the connected domain analysis, color features are extracted for several preprocessed first surface images and several preprocessed second surface images, so that more accurate and comprehensive color features can be obtained without increasing the calculation complexity. The food maturity is calculated based on the first food volume and the first surface color feature to obtain the first food maturity of each food baked according to different baking settings, and the food maturity is calculated based on the second food volume and the second surface color feature to obtain the second food maturity of each food baked according to the same baking setting, so as to perform the baking performance stability test of the oven, so that the obtained baking performance stability test result is more accurate and comprehensive, and can more reliably reflect the overall stability performance of the oven. Based on the color co-occurrence matrix, the color difference between different surfaces of food is analyzed for each first surface color feature and each second surface color feature, and the color difference analysis results are used to detect the baking performance accuracy of the oven, which greatly reduces the color difference analysis error and avoids a large deviation between the color difference analysis results and the actual situation. More reliable baking performance accuracy detection results can be obtained, so that the oven performance detection can achieve a more ideal effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0053] Figure 1 is a flow chart of an oven performance detection method based on image processing in an embodiment of the present invention;

[0054] Figure 2is a schematic diagram of the structure of an oven performance detection device based on image processing in an embodiment of the present invention;

[0055] Figure 3 It is a schematic diagram of the structure of an electronic device in an embodiment of the present invention. DETAILED DESCRIPTION

[0056] 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. Example

[0057] See also Figure 1 , Figure 1 : is a flow chart of an oven performance detection method based on image processing in an embodiment of the present invention, the method comprising:

[0058] S11: acquiring a plurality of first surface images of different surfaces of each food baked in an oven according to different baking settings and a plurality of second surface images of different surfaces of each food baked in an oven according to the same baking settings, and preprocessing the first surface images and the second surface images to obtain preprocessed first surface images and second surface images;

[0059] In the specific implementation process of the present invention, a plurality of first surface images of different surfaces of each food baked in an oven according to different baking settings are obtained, that is, the same food is put into the oven, the food can be selected as bread, etc., different baking settings are set to bake the food, such as setting different baking temperatures, and each surface of the food baked according to the different baking settings is photographed to obtain a plurality of corresponding first surface images, and each baked food is photographed for each surface. A plurality of second surface images of different surfaces of each food baked in an oven according to the same baking settings are obtained, the same food is put into the oven, each food is set with the same baking settings to bake, and each surface of the food baked according to the same baking settings is photographed to obtain a plurality of corresponding second surface images, and each baked food is photographed for each surface. The first surface image and the second surface image are preprocessed, and the preprocessing includes image enhancement, grayscale and binarization processing to obtain the preprocessed first surface image and the second surface image.

[0060] S12: Calculating minimum bounding rectangles for a plurality of preprocessed first surface images and a plurality of preprocessed second surface images, respectively, to obtain a first minimum bounding rectangle corresponding to each preprocessed first surface image and a second minimum bounding rectangle corresponding to each preprocessed second surface image;

[0061] In a specific implementation of the present invention, the method of respectively calculating the minimum bounding rectangle of a plurality of preprocessed first surface images and a plurality of preprocessed second surface images to obtain a first minimum bounding rectangle corresponding to each preprocessed first surface image and a second minimum bounding rectangle corresponding to each preprocessed second surface image includes: respectively extracting a first food surface contour in each preprocessed first surface image and a second food surface contour in each preprocessed second surface image; determining a first search direction range for each first food surface contour and a second search direction range for each second food surface contour, acquiring a first baseline for each preprocessed first surface image based on the first search direction range, and Based on the second search direction range, a second baseline of each preprocessed second surface image is obtained; a plurality of first convex hull points of each first food surface contour and a plurality of second convex hull points of each second food surface contour are obtained, and a plurality of corresponding first circumscribed rectangles are calculated using a rotation matrix based on the plurality of first convex hull points and the first baseline, and a plurality of corresponding second circumscribed rectangles are calculated using a rotation matrix based on the plurality of second convex hull points and the second baseline; based on the boundingRect function, a first minimum circumscribed rectangle corresponding to each preprocessed first surface image among the plurality of first circumscribed rectangles and a second minimum circumscribed rectangle corresponding to each preprocessed second surface image among the plurality of second circumscribed rectangles are calculated.

[0062] Specifically, the first food surface contour in each preprocessed first surface image and the second food surface contour in each preprocessed second surface image are extracted respectively, that is, contour extraction is performed on the images of different surfaces of the baked food, a reference image of the first surface image is input, a central pixel is selected in the image, and the central pixel is any pixel in the image that is not at the boundary position. A third-order pixel matrix is ​​constructed with the central pixel as the center, and contour extraction is performed based on the third-order pixel matrix combined with the canny operator to obtain the food surface contour corresponding to each surface image. The first search direction range of each first food surface contour and the second search direction range of each second food surface contour are determined, the first main direction of the image of the first food surface contour is calculated, and the search direction range of the first food surface contour is obtained by using the inverse tangent function and the direction angle function according to the first main direction, the second main direction of the image of the second food surface contour is calculated, and the search direction range of the second food surface contour is obtained by using the inverse tangent function and the direction angle function according to the second main direction. Based on the first search direction range, a first baseline of each preprocessed first surface image is obtained, and the preprocessed first surface image is divided into intervals according to the first search direction range to determine the first center point corresponding to the preprocessed first surface image. According to the first center point, the first surface image is divided into a first interval and a second interval according to the first search direction range, and the interval center point is determined according to the first interval and the second interval. According to the interval center point, a first baseline is generated, and based on the second search direction range, a second baseline of each preprocessed second surface image is obtained. The step of obtaining the second baseline is the same as the step of obtaining the first baseline, and will not be repeated again. A plurality of first convex hull points of each first food surface contour and a plurality of second convex hull points of each second food surface contour are obtained, wherein the convex hull points are nodes on the convex hull, and the convex hull is a convex polygon surrounding the contour, and the intersection points of two adjacent line segments on the convex polygon are nodes, and a plurality of corresponding first circumscribed rectangles are calculated based on the plurality of first convex hull points and the first reference line using a rotation matrix, and the first reference line is rotated using the rotation matrix to obtain a plurality of regional edge points, and each first convex hull point is rotated according to the center point of the first food surface contour using the rotation matrix to obtain a plurality of circumscribed points, and a plurality of first circumscribed rectangles are generated based on the plurality of circumscribed points and the regional edge points, and a plurality of corresponding second circumscribed rectangles are calculated based on the plurality of second convex hull points and the second reference line using the rotation matrix, and the generation method of the second circumscribed rectangle is the same as the generation method of the first circumscribed rectangle, which will not be described in detail here.Based on the boundingRect function, a first minimum bounding rectangle corresponding to each preprocessed first surface image among several first bounding rectangles and a second minimum bounding rectangle corresponding to each preprocessed second surface image among several second bounding rectangles are calculated. The boundingRect function is used to calculate the length and width of the bounding rectangle to obtain the area of ​​each bounding rectangle, and the bounding rectangle with the smallest area among the bounding rectangles is selected as the minimum bounding rectangle. The boundingRect function can more accurately determine the boundary range and configuration of the target without occupying too many computing resources, thereby obtaining the minimum bounding rectangles of different surface contours of the food.

[0063] S13: extracting first size parameters corresponding to each preprocessed first surface image based on the first minimum circumscribed rectangle, extracting second size parameters corresponding to each preprocessed second surface image based on the second minimum circumscribed rectangle, and calculating the first food volume using a correction coefficient based on the first size parameter, and calculating the second food volume using the correction coefficient based on the second size parameter;

[0064] In a specific implementation of the present invention, the first size parameter corresponding to each preprocessed first surface image is extracted based on the first minimum circumscribed rectangle, the second size parameter corresponding to each preprocessed second surface image is extracted based on the second minimum circumscribed rectangle, and the first food volume is calculated based on the first size parameter using a correction coefficient, and the second food volume is calculated based on the second size parameter using a correction coefficient, including: performing size parameter extraction processing on each preprocessed first surface image based on the first minimum circumscribed rectangle to obtain a first size parameter, the first size parameter including first length information, first width information and first height information; performing size parameter extraction processing on each preprocessed second surface image based on the second minimum circumscribed rectangle to obtain a second size parameter, the second size parameter including second length information, second width information and second height information; determining a first correction coefficient in a correction coefficient library based on the first size parameter, and determining a second correction coefficient in the correction coefficient library based on the second size parameter, the correction coefficients in the correction coefficient library being calculated based on past measured data and past simulation data using a fitting equation and an interval boundary; calculating the first food volume based on the first size parameter using the first correction coefficient, and calculating the second food volume based on the second size parameter using the second correction coefficient.

[0065] Specifically, based on the first minimum bounding rectangle, the size parameters of each preprocessed first surface image are extracted to obtain the first size parameters, which include first length information, first width information and first height information, that is, the corresponding size parameters are extracted according to the minimum bounding rectangle of each surface image of the food, such as extracting length information and height information according to the minimum bounding rectangle of the surface image of the front of the food, and extracting width information according to the minimum bounding rectangle of the surface image of the side of the food, and integrating the extracted parameters to obtain the first size parameters. Based on the second minimum bounding rectangle, the size parameters of each preprocessed second surface image are extracted to obtain the second size parameters, which include second length information, second width information and second height information. A first correction coefficient is determined in a correction coefficient library based on the first size parameter, and a second correction coefficient is determined in the correction coefficient library based on the second size parameter. The correction coefficients in the correction coefficient library are calculated based on past measured data and past simulation data using fitting equations and interval boundaries. The past measured data are measured volume data obtained by actually measuring the volume of food baked according to the baking settings in the past, and the past simulation data are simulated volume data obtained by performing parameter extraction processing of the minimum circumscribed rectangle of food baked according to the baking settings in the past, and then performing volume simulation calculation according to the extracted size parameters. A correction curve is generated according to the past measured data and the past simulation data, and a number of correction interval boundary points are determined on the correction curve according to each corresponding size parameter. The corresponding correction coefficient is determined by fitting the geometric correction interval decomposition point of the equation, and the correction coefficient is stored in the correction coefficient library. The corresponding correction coefficient can be matched in the correction coefficient library according to the size parameter. The first food volume is calculated based on the first size parameter using a first correction coefficient. The first food volume is obtained by multiplying the length information, width information and height information in the first size parameter with the first correction coefficient. The second food volume is calculated based on the second size parameter using a second correction coefficient. The calculation method of the second food volume is consistent with the calculation method of the first food volume. The food volume calculation is corrected by the correction coefficient to make the obtained food volume more in line with the actual situation.

[0066] S14: performing color feature extraction on a plurality of preprocessed first surface images and a plurality of preprocessed second surface images based on connected domain analysis to obtain first surface color features corresponding to each preprocessed first surface image and second surface color features corresponding to each preprocessed second surface image;

[0067] In the specific implementation process of the present invention, the method of extracting color features from a plurality of preprocessed first surface images and a plurality of preprocessed second surface images based on connected domain analysis to obtain first surface color features corresponding to each preprocessed first surface image and second surface color features corresponding to each preprocessed second surface image includes: performing region segmentation processing on a plurality of preprocessed first surface images and a plurality of preprocessed second surface images to obtain first surface images after region segmentation processing and second surface images after region segmentation processing; converting the first surface images after region segmentation processing and the second surface images after region segmentation processing from RGB color space to HSV color space based on rgb2hsv function; clustering the images in HSV color space using a genetic algorithm-based clustering method; The first surface image after each region segmentation processing and the second surface image after each region segmentation processing are subjected to cluster color histogram extraction to obtain a first cluster color histogram corresponding to the first surface image after each region segmentation processing and a second cluster color histogram corresponding to the second surface image after each region segmentation processing; connected domain extraction is performed on the first surface image after each region segmentation processing and the second surface image after each region segmentation processing to obtain the corresponding first connected domain and second connected domain; based on the first connected domain, the first cluster color histogram corresponding to the first surface image after each region segmentation processing is subjected to color feature merging to obtain a first surface color feature; based on the second connected domain, the second cluster color histogram corresponding to the second surface image after each region segmentation processing is subjected to color feature merging to obtain a second surface color feature.

[0068] Specifically, a plurality of preprocessed first surface images and a plurality of preprocessed second surface images are respectively subjected to region segmentation processing, a plurality of extreme points of the preprocessed first surface images are obtained, a plurality of first image matrices and a first heat map are generated according to the plurality of extreme points of the first surface images, and a first image segmentation region is obtained according to the plurality of first image matrices and the first heat map using an image segmentation model, the image segmentation model is a convergence model obtained by inputting a sample data set into a deep neural network for training, a second heat map is determined according to the first image segmentation region in combination with the annotation points, an image segmentation region is generated according to the second heat map using the image segmentation model to obtain a second image segmentation region, and a region segmentation processing is performed on each preprocessed first surface image according to the first image segmentation region and the second image segmentation region, the region segmentation processing of each preprocessed second surface image is consistent with the region segmentation processing method of the first surface image, and the first surface image after each region segmentation processing and the second surface image after each region segmentation processing are obtained, which can effectively improve the accuracy of image region segmentation. The first surface image after each region segmentation processing and the second surface image after each region segmentation processing are converted from the red-green-blue (RGB) color space to the hue-saturation-value (HSV) color space based on the rgb2hsv function. The rgb2hsv function is a function used to realize the conversion between the RGB color space and the HSV color space. A clustering method based on a genetic algorithm is used to extract clustered color histograms of the first surface image and the second surface image after each region segmentation processing in the HSV color space. The clustering method based on the genetic algorithm is used to extract color features of graphics, and the first HSV color histogram of the first surface image after each region segmentation processing and the second HSV color histogram of the second surface image after each region segmentation processing are calculated. A first initial neighborhood width corresponding to the first surface image after the region segmentation processing and a second initial neighborhood width corresponding to the second surface image after the region segmentation processing are set. The first HSV color histogram is clustered according to the set first initial neighborhood width combined with chromosome linear encoding, and the second HSV color histogram is clustered according to the set second initial neighborhood width combined with chromosome linear encoding. The chromosome linear encoding is to use the coordinates of the image center as the linear encoding of the chromosome, and the first cluster color histogram corresponding to the first surface image after each region segmentation processing and the second cluster color histogram corresponding to the second surface image after each region segmentation processing are obtained.Connected domains are extracted from the first surface image after each region segmentation process and the second surface image after each region segmentation process respectively, and connected domains are extracted by a connected domain extraction algorithm. Connected domain analysis can be performed on the first surface image after each region segmentation process and the second surface image after each region segmentation process by the connectedComponentsWithStats() function, and a result matrix including the number of connected domains and related statistical information is returned to detect the connected domains, and the corresponding first connected domains and second connected domains are obtained. Based on the first connected domain, the color features of the first cluster color histograms corresponding to the first surface images after each region segmentation process are merged, and the connected domain labels of the first connected domains in the first surface images after each region segmentation process are obtained. The color features are merged according to the connected domain labels to obtain the first surface color features, which are the color features of each surface of the baked food. Based on the second connected domain, the color features of the second cluster color histograms corresponding to the second surface images after each region segmentation process are merged to obtain the second surface color features, and the acquisition method of the second surface color features is consistent with the acquisition method of the first surface color features. The accuracy of color feature extraction is effectively improved, and the computational complexity is low, which can improve the processing efficiency of color feature extraction.

[0069] S15: Calculating the degree of maturity of food based on the first food volume and the first surface color feature to obtain a first degree of maturity of each food baked according to different baking settings, and calculating the degree of maturity of food based on the second food volume and the second surface color feature to obtain a second degree of maturity of each food baked according to the same baking settings;

[0070] In the specific implementation process of the present invention, the food maturity calculation is performed based on the first food volume and the first surface color feature to obtain the first food maturity of each food baked according to different baking settings, including: calculating the volume difference between the first food volume and the unbaked food volume, inputting the volume difference and the first surface color feature into the maturity model to calculate the food maturity, and obtaining the first food maturity of each food baked according to different baking settings.

[0071] Specifically, the volume difference between the first food volume and the unbaked food volume is calculated, the volume difference and the first surface color feature of each surface of the food are input into the maturity model to calculate the food maturity, the original volume data and baking settings corresponding to different categories of food are collected, the volume change data and color feature data of different categories of food after baking are recorded according to the original volume data, the association between the volume change, color feature data and baking settings and the preset maturity is recorded, and the maturity model is constructed according to the association, and the food maturity is calculated through the maturity model to obtain the second food maturity of each food baked according to the same baking settings. The volume difference between the second food volume and the unbaked food volume is calculated, and the volume difference and the second surface color feature are input into the maturity model to calculate the food maturity, and the second food maturity of each food baked according to the same baking settings is obtained, which not only makes the calculation information of food maturity more comprehensive, but also makes the obtained food maturity more accurate.

[0072] S16: performing a baking performance stability test of the oven based on the first food maturity and the second food maturity to obtain a baking performance stability test result;

[0073] In the specific implementation process of the present invention, the baking performance stability test of the oven is performed based on the first food maturity and the second food maturity to obtain the baking performance stability test result, including: comparing the first food maturity of each food baked with different baking settings with the standard food maturity of the corresponding baking settings to obtain a comparison result; calculating the food maturity mean square deviation based on the second food maturity of the food baked with the same baking settings; performing the baking performance stability test of the oven based on the comparison result and the food maturity mean square deviation to obtain the baking performance stability test result.

[0074] Specifically, the first food maturity of each food baked with different baking settings is compared with the standard food maturity of the corresponding baking settings to obtain a comparison result, that is, the first food maturity of the food baked according to each baking setting is compared with the standard food maturity of the baking setting, and the difference between the first food maturity and the standard food maturity is calculated. The standard food maturity of different baking settings is different. The food maturity mean square deviation is calculated based on the second food maturity of the food baked with the same baking settings. The second maturity of all foods baked with the same baking settings is calculated using the mean square deviation formula to obtain the food maturity mean square deviation, which can reflect the fluctuation of the data. The baking performance stability of the oven is detected based on the comparison result and the mean square error of the food maturity, that is, the mean square error is calculated according to the difference between each first food maturity and the corresponding standard food maturity, and it is judged whether a stable baking result output can be maintained under different baking settings. It can be intuitively understood that when the parameters of the baking settings are high, the food maturity is high, or when the parameters of the baking settings are low, the food maturity is low. The mean square error of the food maturity calculated according to the second maturity can reflect whether a stable food maturity can be maintained under the same baking settings. The baking performance stability level is matched by the mean square error calculated according to the difference between the first food maturity and the corresponding standard food maturity and the mean square error of the food maturity calculated according to the second maturity, which can be divided into high stability level, high stability level and medium stability level, that is, the baking performance stability detection result is obtained.

[0075] S17: Performing surface color difference analysis on each first surface color feature and each second surface color feature based on the color co-occurrence matrix to obtain a first surface color difference analysis result and a second surface color difference analysis result, and performing a baking performance accuracy test of the oven based on the first surface color difference analysis result and the second surface color difference analysis result to obtain a baking performance accuracy test result.

[0076] In the specific implementation process of the present invention, the surface color difference analysis is performed on each first surface color feature and each second surface color feature based on the color co-occurrence matrix to obtain the first surface color difference analysis result and the second surface color difference analysis result, and the baking performance accuracy detection of the oven is performed based on the first surface color difference analysis result and the second surface color difference analysis result to obtain the baking performance accuracy detection result, including: converting the color space where each first surface color feature and each second surface color feature are located into the CIELAB color space, calculating the three channel parameters of each first surface color feature in the CIELAB color space and the three channel parameters of each second surface color feature in the CIELAB color space; obtaining the first surface color feature A first color co-occurrence matrix and a second color co-occurrence matrix of each second surface color feature; based on the first color co-occurrence matrix of each first surface color feature and the three channel parameters in the CIELAB color space, a color difference analysis between different surfaces of each food baked with different baking settings is performed to obtain a first surface color difference analysis result; the second color co-occurrence matrix of each second surface color feature and the three channel parameters in the CIELAB color space are used to perform a color difference analysis between different surfaces of each food baked with the same baking settings to obtain a second surface color difference analysis result; based on the first surface color difference analysis result and the second surface color difference analysis result combined with a preset color difference judgment standard, the baking performance accuracy of the oven is tested to obtain a baking performance accuracy test result.

[0077] Specifically, the color space where each first surface color feature and each second surface color feature are located is converted into the CIELAB color space. The CIELAB color space is a uniform color space obtained by nonlinear transformation based on the chromaticity system, which consists of three channel parameters, namely brightness, hue and saturation. The three channel parameters of each first surface color feature in the CIELAB color space and the three channel parameters of each second surface color feature in the CIELAB color space are calculated, and the channel parameters are calculated by a linear transformation matrix. The first color co-occurrence matrix of each first surface color feature and the second color co-occurrence matrix of each second surface color feature are obtained, and the color co-occurrence matrix is ​​constructed according to the hue channel value of the surface color feature. Based on the first color co-occurrence matrix of each first surface color feature and the three channel parameters in the CIELAB color space, the color difference analysis between different surfaces of each food baked under different baking settings is performed, that is, whether there is a large difference in the color difference between the surfaces of each food baked under different baking settings is analyzed, the autocorrelation value of each surface is calculated according to the first color co-occurrence matrix of the first surface color features of different surfaces of the food, the difference between the autocorrelation values ​​of each surface is calculated, the color difference value between each surface is calculated according to the three channel parameters of the first surface color features of different surfaces of the food in the CIELAB color space using the color difference calculation formula, the target color difference between each surface is obtained according to the difference between the autocorrelation values ​​and the color difference value, and the color difference between different surfaces of the baked food is analyzed, that is, the first surface color difference analysis result is obtained. The second color co-occurrence matrix of each second surface color feature and the three channel parameters in the CIELAB color space are used to perform the color difference analysis between different surfaces of each food baked under the same baking settings, and the second surface color difference analysis result is obtained, that is, whether there is a large difference in the color difference between the surfaces of each food baked under the same baking settings is analyzed, and the analysis steps are consistent with the first surface color difference analysis result.Based on the first surface color difference analysis result and the second surface color difference analysis result combined with the preset color difference judgment standard, the baking performance accuracy of the oven is detected. According to the first surface color difference analysis result, it can be detected whether the color difference between the surfaces of the food baked in the oven is not large under different baking settings, that is, the color of each surface is uniform, and it can be obtained whether the oven can accurately control the baking effect of each surface of the food for different baking settings, which reflects whether the oven can maintain a high control accuracy under different baking settings. According to the second surface color difference analysis result, it can be detected whether the color between the surfaces of the food baked in the oven is uniform under the same baking settings, and it can be obtained whether the oven can accurately control the baking effect of each surface of the food for the same baking settings and maintain the stability of the accuracy, which reflects whether the oven can maintain a high and stable control accuracy under the same baking settings; the baking performance accuracy score evaluation of different baking settings is performed according to the first surface color difference analysis, and the baking performance accuracy score evaluation of the same baking setting is performed according to the second surface color difference analysis, and the two are comprehensively evaluated to obtain the baking performance accuracy detection result.

[0078] In an embodiment of the present invention, the minimum circumscribed rectangle is calculated for several preprocessed first surface images and several preprocessed second surface images to extract corresponding size parameters, and the food volume is calculated according to the corresponding size parameters, so that the obtained food volume can be more accurate. Based on the connected domain analysis, color features are extracted for several preprocessed first surface images and several preprocessed second surface images, so that more accurate and comprehensive color features can be obtained without increasing the calculation complexity. The food maturity is calculated based on the first food volume and the first surface color feature to obtain the first food maturity of each food baked according to different baking settings, and the food maturity is calculated based on the second food volume and the second surface color feature to obtain the second food maturity of each food baked according to the same baking setting, so as to perform the baking performance stability test of the oven, so that the obtained baking performance stability test result is more accurate and comprehensive, and can more reliably reflect the overall stability performance of the oven. Based on the color co-occurrence matrix, the color difference between different surfaces of food is analyzed for each first surface color feature and each second surface color feature, and the color difference analysis results are used to detect the baking performance accuracy of the oven, which greatly reduces the color difference analysis error and avoids a large deviation between the color difference analysis results and the actual situation. More reliable baking performance accuracy detection results can be obtained, so that the oven performance detection can achieve a more ideal effect. Example

[0079] See also Figure 2 , Figure 2: is a schematic diagram of the structure of an oven performance detection device based on image processing in an embodiment of the present invention, the device comprises:

[0080] The image preprocessing module 21 is used to obtain a plurality of first surface images of different surfaces of each food baked in the oven according to different baking settings and a plurality of second surface images of different surfaces of each food baked in the oven according to the same baking settings, and preprocess the first surface images and the second surface images to obtain the preprocessed first surface images and the second surface images;

[0081] Minimum bounding rectangle calculation module 22: used to calculate the minimum bounding rectangle of a plurality of preprocessed first surface images and a plurality of preprocessed second surface images, respectively, to obtain a first minimum bounding rectangle corresponding to each preprocessed first surface image and a second minimum bounding rectangle corresponding to each preprocessed second surface image;

[0082] A food volume calculation module 23: configured to extract a first size parameter corresponding to each preprocessed first surface image based on the first minimum circumscribed rectangle, extract a second size parameter corresponding to each preprocessed second surface image based on the second minimum circumscribed rectangle, calculate a first food volume based on the first size parameter using a correction coefficient, and calculate a second food volume based on the second size parameter using a correction coefficient;

[0083] Color feature extraction module 24: used for performing color feature extraction on a plurality of preprocessed first surface images and a plurality of preprocessed second surface images based on connected domain analysis, to obtain first surface color features corresponding to each preprocessed first surface image and second surface color features corresponding to each preprocessed second surface image;

[0084] The food maturity calculation module 25 is used to calculate the food maturity based on the first food volume and the first surface color feature to obtain the first food maturity of each food baked according to different baking settings, and to calculate the food maturity based on the second food volume and the second surface color feature to obtain the second food maturity of each food baked according to the same baking settings;

[0085] A baking performance stability detection module 26 is used to detect the baking performance stability of the oven based on the first food maturity and the second food maturity, and obtain a baking performance stability detection result;

[0086] Baking performance accuracy detection module 27: used to perform surface color difference analysis on each first surface color feature and each second surface color feature based on the color co-occurrence matrix, obtain the first surface color difference analysis result and the second surface color difference analysis result, and perform baking performance accuracy detection of the oven based on the first surface color difference analysis result and the second surface color difference analysis result to obtain the baking performance accuracy detection result.

[0087] In the specific implementation process of the present invention, the specific implementation method of the device item can refer to the implementation method of the above-mentioned method item, which will not be repeated here.

[0088] In an embodiment of the present invention, the minimum circumscribed rectangle is calculated for several preprocessed first surface images and several preprocessed second surface images to extract corresponding size parameters, and the food volume is calculated according to the corresponding size parameters, so that the obtained food volume can be more accurate. Based on the connected domain analysis, color features are extracted for several preprocessed first surface images and several preprocessed second surface images, so that more accurate and comprehensive color features can be obtained without increasing the calculation complexity. The food maturity is calculated based on the first food volume and the first surface color feature to obtain the first food maturity of each food baked according to different baking settings, and the food maturity is calculated based on the second food volume and the second surface color feature to obtain the second food maturity of each food baked according to the same baking setting, so as to perform the baking performance stability test of the oven, so that the obtained baking performance stability test result is more accurate and comprehensive, and can more reliably reflect the overall stability performance of the oven. Based on the color co-occurrence matrix, the color difference between different surfaces of food is analyzed for each first surface color feature and each second surface color feature, and the color difference analysis results are used to detect the baking performance accuracy of the oven, which greatly reduces the color difference analysis error and avoids a large deviation between the color difference analysis results and the actual situation. More reliable baking performance accuracy detection results can be obtained, so that the oven performance detection can achieve a more ideal effect.

[0089] An embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, an oven performance detection method based on image processing in any of the above embodiments is implemented. The computer-readable storage medium includes, but is not limited to, any type of disk (including floppy disk, hard disk, optical disk, CD-ROM, and magneto-optical disk), ROM (Read-Only Memory), RAM (Random Access Memory), EPROM (Erasable Programmable Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), flash memory, magnetic card or optical card. That is, the storage device includes any medium that can store or transmit information in a readable form by a device (for example, a computer, a mobile phone), which can be a read-only memory, a disk or an optical disk, etc. Example

[0090] See also Figure 3 , Figure 3 It is a schematic diagram of the structure of an electronic device in an embodiment of the present invention.

[0091] The embodiment of the present invention further provides an electronic device, such as Figure 3 As shown, the electronic device includes a memory 31, a processor 33, and a computer program 32 stored in the memory 31 and executable on the processor 33. Those skilled in the art will appreciate that Figure 3The electronic device shown does not constitute a limitation on all devices, and may include more or fewer components than shown, or combine certain components. The memory 31 can be used to store the computer program 32 and various functional modules, and the processor 33 runs the computer program 32 stored in the memory 31, thereby executing various functional applications and data processing of the device. The memory can be an internal memory or an external memory, or include both internal and external memories. The internal memory can include a read-only memory (ROM), a programmable ROM (PROM), an electrically programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a flash memory, or a random access memory. The external memory can include a hard disk, a floppy disk, a ZIP disk, a U disk, a magnetic tape, etc. The processor 33 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general processor may be a microprocessor, a single chip microcomputer or the processor 33 may also be any conventional processor, etc. The processor and memory disclosed in the present invention include but are not limited to these types of processors and memories. The processor and memory disclosed in the present invention are only examples and not limitations.

[0092] As an embodiment, the electronic device includes: one or more processors 33, a memory 31, and one or more computer programs 32, wherein the one or more computer programs 32 are stored in the memory 31 and are configured to be executed by the one or more processors 33, and the one or more computer programs 32 are configured to execute the oven performance detection method based on image processing in any of the above-mentioned embodiments. For the specific implementation process, please refer to the above-mentioned embodiments, which will not be repeated here.

[0093] In an embodiment of the present invention, the minimum circumscribed rectangle is calculated for several preprocessed first surface images and several preprocessed second surface images to extract corresponding size parameters, and the food volume is calculated according to the corresponding size parameters, so that the obtained food volume can be more accurate. Based on the connected domain analysis, color features are extracted for several preprocessed first surface images and several preprocessed second surface images, so that more accurate and comprehensive color features can be obtained without increasing the calculation complexity. The food maturity is calculated based on the first food volume and the first surface color feature to obtain the first food maturity of each food baked according to different baking settings, and the food maturity is calculated based on the second food volume and the second surface color feature to obtain the second food maturity of each food baked according to the same baking setting, so as to perform the baking performance stability test of the oven, so that the obtained baking performance stability test result is more accurate and comprehensive, and can more reliably reflect the overall stability performance of the oven. Based on the color co-occurrence matrix, the color difference between different surfaces of food is analyzed for each first surface color feature and each second surface color feature, and the color difference analysis results are used to detect the baking performance accuracy of the oven, which greatly reduces the color difference analysis error and avoids a large deviation between the color difference analysis results and the actual situation. More reliable baking performance accuracy detection results can be obtained, so that the oven performance detection can achieve a more ideal effect.

[0094] In addition, the above is a detailed introduction to an oven performance detection method based on image processing and related devices provided in an embodiment of the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for a general technician in this field, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. An oven performance detection method based on image processing, characterized in that: The method comprises: Acquire a plurality of first surface images of different surfaces of each food baked in the oven according to different baking settings and a plurality of second surface images of different surfaces of each food baked in the oven according to the same baking settings, and preprocess the first surface images and the second surface images to obtain preprocessed first surface images and second surface images; Calculating minimum bounding rectangles for a plurality of preprocessed first surface images and a plurality of preprocessed second surface images respectively, to obtain a first minimum bounding rectangle corresponding to each preprocessed first surface image and a second minimum bounding rectangle corresponding to each preprocessed second surface image; extracting first size parameters corresponding to each preprocessed first surface image based on the first minimum circumscribed rectangle, extracting second size parameters corresponding to each preprocessed second surface image based on the second minimum circumscribed rectangle, and calculating a first food volume using a correction coefficient based on the first size parameters, and calculating a second food volume using the correction coefficient based on the second size parameters; Based on the connected domain analysis, color features are extracted from a plurality of preprocessed first surface images and a plurality of preprocessed second surface images to obtain first surface color features corresponding to each preprocessed first surface image and second surface color features corresponding to each preprocessed second surface image; Calculating the degree of maturity of food based on the first food volume and the first surface color feature to obtain a first degree of maturity of each food baked according to different baking settings, and calculating the degree of maturity of food based on the second food volume and the second surface color feature to obtain a second degree of maturity of each food baked according to the same baking settings; Performing a baking performance stability test of the oven based on the first food maturity and the second food maturity to obtain a baking performance stability test result, including: comparing the first food maturity of each food baked with different baking settings with the standard food maturity of the corresponding baking settings to obtain a comparison result; calculating the food maturity mean square deviation based on the second food maturity of the food baked with the same baking settings; performing a baking performance stability test of the oven based on the comparison result and the food maturity mean square deviation to obtain a baking performance stability test result; Based on the color co-occurrence matrix, surface color difference analysis is performed on each first surface color feature and each second surface color feature to obtain a first surface color difference analysis result and a second surface color difference analysis result, and based on the first surface color difference analysis result and the second surface color difference analysis result, the baking performance accuracy detection of the oven is performed to obtain a baking performance accuracy detection result, including: converting the color space where each first surface color feature and each second surface color feature are located into a CIELAB color space, calculating three channel parameters of each first surface color feature in the CIELAB color space and three channel parameters of each second surface color feature in the CIELAB color space; obtaining a first color co-occurrence matrix of each first surface color feature The method comprises the steps of: obtaining a first color co-occurrence matrix of each first surface color feature and a second color co-occurrence matrix of each second surface color feature; performing a color difference analysis between different surfaces of each food baked with different baking settings based on the first color co-occurrence matrix of each first surface color feature and the three channel parameters in the CIELAB color space to obtain a first surface color difference analysis result; performing a color difference analysis between different surfaces of each food baked with the same baking settings based on the second color co-occurrence matrix of each second surface color feature and the three channel parameters in the CIELAB color space to obtain a second surface color difference analysis result; performing a baking performance accuracy test on the oven based on the first surface color difference analysis result and the second surface color difference analysis result in combination with a preset color difference judgment standard to obtain a baking performance accuracy test result.

2. The oven performance detection method based on image processing according to claim 1, characterized in that: The method of calculating the minimum bounding rectangles of the plurality of preprocessed first surface images and the plurality of preprocessed second surface images to obtain a first minimum bounding rectangle corresponding to each preprocessed first surface image and a second minimum bounding rectangle corresponding to each preprocessed second surface image comprises: Respectively extracting first food surface contours in each preprocessed first surface image and second food surface contours in each preprocessed second surface image; Determine a first search direction range of each first food surface contour and a second search direction range of each second food surface contour, obtain a first reference line of each preprocessed first surface image based on the first search direction range, and obtain a second reference line of each preprocessed second surface image based on the second search direction range; Obtaining a plurality of first convex hull points of each first food surface contour and a plurality of second convex hull points of each second food surface contour, calculating a plurality of corresponding first circumscribed rectangles based on the plurality of first convex hull points and the first reference line using a rotation matrix, and calculating a plurality of corresponding second circumscribed rectangles based on the plurality of second convex hull points and the second reference line using a rotation matrix; Based on the boundingRect function, a first minimum bounding rectangle corresponding to each preprocessed first surface image among a plurality of first bounding rectangles and a second minimum bounding rectangle corresponding to each preprocessed second surface image among a plurality of second bounding rectangles are calculated.

3. The oven performance detection method based on image processing according to claim 1, characterized in that: The method of extracting a first size parameter corresponding to each preprocessed first surface image based on the first minimum circumscribed rectangle, extracting a second size parameter corresponding to each preprocessed second surface image based on the second minimum circumscribed rectangle, and calculating a first food volume using a correction coefficient based on the first size parameter, and calculating a second food volume using a correction coefficient based on the second size parameter, comprises: Performing size parameter extraction processing on each preprocessed first surface image based on the first minimum circumscribed rectangle to obtain first size parameters, where the first size parameters include first length information, first width information, and first height information; Performing size parameter extraction processing on each preprocessed second surface image based on the second minimum circumscribed rectangle to obtain second size parameters, where the second size parameters include second length information, second width information, and second height information; Determine a first correction coefficient in a correction coefficient library based on the first size parameter, and determine a second correction coefficient in a correction coefficient library based on the second size parameter, wherein the correction coefficients in the correction coefficient library are calculated based on past measured data and past simulation data using fitting equations and interval boundaries; A first food volume is calculated using a first correction factor based on the first size parameter, and a second food volume is calculated using a second correction factor based on the second size parameter.

4. The oven performance detection method based on image processing according to claim 1, characterized in that: The method of extracting color features from a plurality of preprocessed first surface images and a plurality of preprocessed second surface images based on connected domain analysis to obtain first surface color features corresponding to each preprocessed first surface image and second surface color features corresponding to each preprocessed second surface image includes: Performing region segmentation processing on a plurality of preprocessed first surface images and a plurality of preprocessed second surface images respectively, to obtain first surface images after the region segmentation processing and second surface images after the region segmentation processing; Based on the rgb2hsv function, the first surface image after each region segmentation process and the second surface image after each region segmentation process are converted from the RGB color space to the HSV color space; A clustering method based on a genetic algorithm is used to extract cluster color histograms from the first surface image after each region segmentation process and the second surface image after each region segmentation process in the HSV color space, so as to obtain a first cluster color histogram corresponding to the first surface image after each region segmentation process and a second cluster color histogram corresponding to the second surface image after each region segmentation process; Performing connected domain extraction on the first surface image after each region segmentation process and the second surface image after each region segmentation process to obtain corresponding first connected domains and second connected domains; Based on the first connected domain, color features of the first cluster color histograms corresponding to the first surface image after each region segmentation process are merged to obtain a first surface color feature; Based on the second connected domain, the second cluster color histograms corresponding to the second surface images after the segmentation of each region are merged to obtain the second surface color features.

5. The oven performance detection method based on image processing according to claim 1, characterized in that: The calculating the food maturity based on the first food volume and the first surface color feature to obtain the first food maturity of each food baked according to different baking settings includes: The volume difference between the first food volume and the unbaked food volume is calculated, and the volume difference and the first surface color feature are input into a maturity model to calculate the food maturity, thereby obtaining the first food maturity of each food baked according to different baking settings.

6. An oven performance detection device based on image processing, using the oven performance detection method based on image processing according to any one of claims 1 to 5, characterized in that: The device comprises: An image preprocessing module: used to obtain a plurality of first surface images of different surfaces of each food baked in the oven according to different baking settings and a plurality of second surface images of different surfaces of each food baked in the oven according to the same baking settings, and preprocess the first surface images and the second surface images to obtain the preprocessed first surface images and the second surface images; Minimum bounding rectangle calculation module: used to calculate the minimum bounding rectangle of a plurality of preprocessed first surface images and a plurality of preprocessed second surface images, respectively, to obtain a first minimum bounding rectangle corresponding to each preprocessed first surface image and a second minimum bounding rectangle corresponding to each preprocessed second surface image; A food volume calculation module: used for extracting a first size parameter corresponding to each preprocessed first surface image based on the first minimum circumscribed rectangle, extracting a second size parameter corresponding to each preprocessed second surface image based on the second minimum circumscribed rectangle, and calculating a first food volume based on the first size parameter using a correction coefficient, and calculating a second food volume based on the second size parameter using a correction coefficient; Color feature extraction module: used for respectively extracting color features from a plurality of preprocessed first surface images and a plurality of preprocessed second surface images based on connected domain analysis, to obtain first surface color features corresponding to each preprocessed first surface image and second surface color features corresponding to each preprocessed second surface image; a food maturity calculation module: configured to calculate the food maturity based on the first food volume and the first surface color feature to obtain the first food maturity of each food baked according to different baking settings, and to calculate the food maturity based on the second food volume and the second surface color feature to obtain the second food maturity of each food baked according to the same baking settings; A baking performance stability detection module: used to detect the baking performance stability of the oven based on the first food maturity and the second food maturity, and obtain a baking performance stability detection result; Baking performance accuracy detection module: used to perform surface color difference analysis on each first surface color feature and each second surface color feature based on the color co-occurrence matrix, obtain the first surface color difference analysis result and the second surface color difference analysis result, and perform baking performance accuracy detection of the oven based on the first surface color difference analysis result and the second surface color difference analysis result to obtain the baking performance accuracy detection result.

7. An electronic device, comprising a processor and a memory, characterized in that: The memory is used to store instructions, and the processor is used to call the instructions in the memory, so that the electronic device executes the oven performance detection method based on image processing as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and when the computer instructions are executed on an electronic device, the electronic device executes the oven performance detection method based on image processing according to any one of claims 1 to 5.

Images