Combustion chamber inner wall carbon deposition detection method and system based on machine vision

Abstract

The invention belongs to the technical field of image anomaly detection, and particularly relates to a combustion chamber inner wall carbon deposition detection method and system based on machine vision so as to solve the technical problem that an existing method is prone to interference and misjudgment due to neglect of texture directivity and lack of adaptivity. The detection method comprises the following steps: S1, calculating a kurtosis coefficient of a global brightness histogram of an RGB image to obtain a first chrominance component and a second chrominance component; s2, acquiring respective high-frequency sub-bands in horizontal, vertical and diagonal directions under at least two scales; calculating a directivity index representing the anisotropy degree of the chrominance noise; s3, determining a judgment threshold value of the directivity index; and S4, performing weighted summation on the chrominance noise energy obtained under each scale in combination with the center frequency corresponding to each scale to obtain a comprehensive noise index representing the carbon deposition degree. The comprehensive noise index obtained through multi-scale information fusion can accurately represent the carbon deposition degree.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of image anomaly detection, and particularly relates to a combustion chamber inner wall carbon deposition detection method and system based on machine vision. BACKGROUND

[0002] Carbon deposition in the combustion chamber of an engine is an important factor affecting the power, economy and emission of the engine, and long-term accumulation can cause engine knock, unstable idling, increased fuel consumption and a series of problems. The traditional visual detection method mainly relies on color and brightness information of the image. However, the internal working environment of the combustion chamber is poor, the lighting conditions are usually uneven, and there are various interference factors such as strong metal reflection, oil stains and mechanical scratches on the inner wall surface. These interference factors will affect the stability of color and brightness features, and are easy to misjudge oil stains or shadows as carbon deposition, or miss the carbon deposition in the reflection area.

[0003] Carbon deposition usually appears as rough and irregular texture in microcosm, therefore, some methods extract texture features of the image by using gray level co-occurrence matrix, local binary pattern or wavelet transform, etc., in order to distinguish carbon deposition from smooth metal inner wall. However, the texture features of carbon deposition are different at different scales, and are similar to the texture features of oil stains, corrosion spots and other defects. Most of the existing texture analysis methods focus on the energy or statistical features of the texture, and ignore the directional information. The scratches and processing marks commonly seen on the inner wall of the combustion chamber have obvious directional texture, and this kind of structural texture is easy to interfere with the random texture of carbon deposition, leading to misjudgment. In addition, when the overall brightness and contrast of the photographed image change due to working conditions or equipment differences, the detection performance often decreases, and there is a lack of adaptive adjustment ability to the global state of the image. SUMMARY

[0004] The application provides a combustion chamber inner wall carbon deposition detection method and system based on machine vision, to solve the technical problem that the existing methods are easy to be disturbed and misjudged due to ignoring the texture directionality and lacking adaptability.

[0005] In a first aspect, the application provides a combustion chamber inner wall carbon deposition detection method based on machine vision, comprising the following steps: S1, acquiring an RGB image of the inner wall of the combustion chamber, calculating the kurtosis coefficient of the global brightness histogram of the RGB image, and performing nonlinear function weighting on the R, G and B components of each pixel point based on the ratio of the neighborhood brightness mean value of each pixel point in the RGB image to the preset brightness reference, to obtain a first chroma component and a second chroma component; S2, perform two-dimensional discrete wavelet transform on the first chromaticity component and the second chromaticity component respectively to obtain high-frequency sub-bands in the horizontal, vertical and diagonal directions at at least two scales; at the same scale, summarize the high-frequency sub-band energy of the first chromaticity component and the second chromaticity component in the horizontal, vertical and diagonal directions to obtain the energy distribution in the three directions at the corresponding scale, and calculate the directional index representing the degree of anisotropy of chromaticity noise based on the energy distribution. S3, based on the arctangent function of the difference between the kurtosis coefficient and the preset reference coefficient, determine the judgment threshold of the directional index; when the directional index is less than the judgment threshold, the sum of the high-frequency subband energies in the three directions at the corresponding scale is taken as the chromaticity noise energy; otherwise, the chromaticity noise energy is set to a preset low value. S4, the chromatic noise energy obtained at each scale is weighted and summed with the center frequency corresponding to each scale to obtain the comprehensive noise index representing the degree of carbon deposition.

[0006] Furthermore, in S1, for any pixel in the RGB image, a 5×5 neighborhood window is constructed, and the arithmetic mean of the brightness of all pixels within this neighborhood window is calculated as the neighborhood brightness mean. ; Calculate the weighting coefficients ,in, The preset brightness reference; The R, G, and B components of the pixel are multiplied by the weighting coefficients respectively. The weighted components are obtained. , , ; The first chromaticity component is calculated based on the weighted components. Second chromaticity component .

[0007] Furthermore, in S2, the Daubechies4 wavelet basis is used to perform three-level two-dimensional discrete wavelet decomposition on the first and second chromaticity components respectively to obtain the horizontal high-frequency sub-band, vertical high-frequency sub-band and diagonal high-frequency sub-band at the first, second and third scales.

[0008] Furthermore, in S2, for any scale and direction Calculate the first chromaticity component in the direction The average of the absolute values ​​of all wavelet coefficients in the high-frequency subband is used to obtain the high-frequency subband energy of the first chromaticity component. ; Calculate the second chromaticity component in the direction The average of the absolute values ​​of all wavelet coefficients in the high-frequency subband is used to obtain the high-frequency subband energy of the second chromaticity component. ; The high-frequency subband energy of the first chromaticity component High-frequency subband energy of the second chromaticity component By adding them together, we obtain the total high-frequency subband energy at that scale and in that direction. ,satisfy .

[0009] Furthermore, in S2, for any scale Calculate the high-frequency subband energy in the horizontal, vertical, and diagonal directions respectively. , , ; Calculate the arithmetic mean of the high-frequency subband energies in three directions. and standard deviation ; Calculate the corresponding scale Directional indicators below Let be the ratio of the standard deviation to the arithmetic mean, satisfying . .

[0010] Furthermore, in S3, the decision threshold T is calculated using the following formula: ,in, The kurtosis coefficient is the kurtosis coefficient of the global luminance histogram of the RGB image. As a preset baseline coefficient, This is the scaling factor. This is the bias coefficient.

[0011] Furthermore, the preset low value is 0.

[0012] Furthermore, in S4, the chromaticity noise energy obtained at the 1st, 2nd, and 3rd scales is denoted as follows: , , ; Set weighting coefficients for each scale. , , The weighting coefficients are positively correlated with the center frequency of the corresponding scale; Calculate the overall noise index ,satisfy .

[0013] Furthermore, , , .

[0014] In a second aspect, the present application provides a machine vision-based combustion chamber inner wall carbon deposition detection system, comprising a memory and a processor, wherein the memory stores computer program instructions which, when executed by the processor, implement the machine vision-based combustion chamber inner wall carbon deposition detection method described above.

[0015] The beneficial effect is that by performing nonlinear function weighting on the image pixels based on the ratio of the neighborhood brightness mean value to the preset brightness reference, the interference caused by uneven illumination and metal reflection on the combustion chamber inner wall is suppressed, and the stability of the first chroma component and the second chroma component is enhanced. The present application obtains high-frequency subbands in three directions of horizontal, vertical and diagonal at least two scales by performing two-dimensional discrete wavelet transform on the two chroma components, and constructs a directionality index representing the anisotropy degree of chroma noise based on the energy distribution of the high-frequency subbands in three directions at each scale, thereby distinguishing the irregular, isotropic texture of carbon deposition from the structural texture with obvious directionality such as scratches and processing marks, and accordingly excluding the energy of interference texture, thereby reducing the misjudgment rate. In addition, the determination threshold of the directionality index is associated with the kurtosis coefficient of the global brightness histogram of the image, so that the detection standard can be adjusted according to the overall contrast of the image, improving the consistency and reliability of the detection results of the method under different shooting conditions. The comprehensive noise index obtained by multi-scale information fusion can accurately represent the carbon deposition degree. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 Flowchart of the machine vision-based combustion chamber inner wall carbon deposition detection method; Figure 2 Flowchart of the chroma noise energy determination based on the directionality index; Figure 3 Schematic diagram of the machine vision-based combustion chamber inner wall carbon deposition detection system. DETAILED DESCRIPTION

[0017] Embodiments of the machine vision-based combustion chamber inner wall carbon deposition detection method provided by the present application: As shown in Figure 1 The machine vision-based combustion chamber inner wall carbon deposition detection method comprises the following steps: S1, obtaining an RGB image of the combustion chamber inner wall, calculating the kurtosis coefficient of the global brightness histogram of the RGB image, performing nonlinear function weighting on the R, G and B components of each pixel point based on the ratio of the neighborhood brightness mean value of each pixel point in the RGB image to the preset brightness reference, to obtain a first chroma component and a second chroma component.

[0018] An RGB image of MxN pixels is obtained by taking a picture of the inner wall of the combustion chamber through an endoscope, the RGB image is converted into a gray scale image, the brightness component L of the gray scale image is calculated, and the number of pixels of each brightness level from 0 to 255 is counted to obtain a brightness histogram H. The kurtosis coefficient K is calculated based on the brightness histogram H, and the formula is the ratio of the fourth central moment of the image to the fourth power of the standard deviation.

[0019] In an optional embodiment, in S1, for any pixel point in the RGB image, a 5x5 neighborhood window is constructed, the arithmetic mean value of the brightness of all pixels in the neighborhood window is calculated as the neighborhood brightness mean value ; The weighted coefficient is calculated as , wherein is the preset brightness reference; The R, G and B components of the pixel point are multiplied by the weighted coefficient , respectively, to obtain the weighted components , , ; The first chroma component and the second chroma component are calculated based on the weighted components.

[0020] For example, a pixel point with coordinates (100, 150) in the RGB image is processed, and the original R, G and B values of the pixel point are 160, 120 and 80, respectively. A 5x5 pixel window from coordinates (98, 148) to (102, 152) is constructed with the pixel point as the center. The brightness of all 25 pixels in the window is calculated, and the arithmetic mean value of the brightness is obtained, which is assumed to be 110.

[0021] The preset brightness reference is set to 128, the weighted coefficient is calculated by substituting the formula, and the weighted coefficient is about 0.933. The weighted coefficient is multiplied by the original R, G and B components of the pixel point, respectively, to obtain the new weighted components, i.e. is 149.28, is 111.96, is 74.64. The chroma components are calculated based on the weighted components, the first chroma component = 37.32, and the second chroma component = 55.98.

[0022] The preset brightness reference can be selected by the following steps: (1) Select a set of representative combustion chamber inner wall images as the reference set, which should cover different lighting conditions, combustion states, and carbon deposition levels to ensure the generalization ability of the model.

[0023] (2) Perform global brightness histogram analysis on each image to obtain the frequency distribution of pixels with brightness values from 0 to 255; calculate the cumulative probability density function of each brightness level and determine the brightness value corresponding to a certain percentage of cumulative probability as a candidate reference value. The specific percentage can be 50%, for example.

[0024] (3) Based on experimental or empirical data, select a brightness value that can balance the performance under different lighting conditions as the final preset brightness reference . For example, the average or median of all candidate reference values can be selected.

[0025] (4) Use the selected preset brightness reference to process multiple test images and evaluate whether the results meet expectations. If significant deviations are found, re-evaluate and adjust the value of the preset brightness reference .

[0026] S2, perform two-dimensional discrete wavelet transform on the first and second chroma components respectively to obtain high-frequency subbands in horizontal, vertical, and diagonal directions at least two scales; in the same scale, aggregate the high-frequency subband energies of the first and second chroma components in horizontal, vertical, and diagonal directions to obtain energy distributions in the three directions at the corresponding scale, and calculate a directionality index representing the anisotropy degree of chroma noise based on the energy distributions.

[0027] In an optional embodiment, in S2, Daubechies4 wavelet basis is used to perform 3-layer two-dimensional discrete wavelet decomposition on the first and second chroma components respectively to obtain horizontal, vertical, and diagonal high-frequency subbands at the 1st, 2nd, and 3rd scales.

[0028] Specifically, for the two-dimensional data matrix composed of the first chroma component calculated by all pixel points, Daubechies4 wavelet basis is applied to perform first-layer two-dimensional discrete wavelet decomposition, and the original two-dimensional data matrix is decomposed into a low-frequency approximation subband and three high-frequency subbands, which are horizontal , vertical , and diagonal , respectively, and the three high-frequency subbands constitute the detail information at the 1st scale.

[0029] The low-frequency approximation subband obtained by the first-layer decomposition is Perform the same two-dimensional discrete wavelet decomposition again to generate a new low-frequency subband. And the three high-frequency sub-bands at the second scale, namely the horizontal ones. Vertical direction and diagonal direction The decomposition process continues to repeat, and the low-frequency approximate subbands obtained from the second-level decomposition are... Further decomposition yields the horizontal high-frequency subband at the third scale. Vertical high-frequency subband and diagonal high-frequency subband After completing the three-layer decomposition, the first chromaticity component was obtained. Nine high-frequency subbands at three different scales. The same complete three-layer decomposition process is also applied to the second chromaticity component. The constructed two-dimensional data matrix yields the second chromaticity component. Nine additional high-frequency subbands across three scales.

[0030] In an optional embodiment, in S2, for any scale and direction Calculate the first chromaticity component in the direction The average of the absolute values ​​of all wavelet coefficients in the high-frequency subband is used to obtain the high-frequency subband energy of the first chromaticity component. ; Calculate the second chromaticity component in the direction The average of the absolute values ​​of all wavelet coefficients in the high-frequency subband is used to obtain the high-frequency subband energy of the second chromaticity component. ; The high-frequency subband energy of the first chromaticity component High-frequency subband energy of the second chromaticity component By adding them together, we obtain the total high-frequency subband energy at that scale and in that direction. ,satisfy .

[0031] Specifically, taking the calculation of the total high-frequency subband energy in the horizontal direction of the second scale as an example, it is necessary to locate the high-frequency subband in the horizontal direction of the second scale, that is, the first chromaticity component obtained above. Quantity Subband and second chromaticity component Quantity Sub-band. Assume of The subband is a matrix containing 4096 wavelet coefficients. During calculation, the absolute value of each of the 4096 wavelet coefficients is calculated, all absolute values ​​are summed, and the sum is then divided by the total number of coefficients, 4096.

[0032] For example, if the sum of the absolute values of all coefficients in the high frequency subband is 81920, then the high frequency subband energy for the 2nd scale is 20. Similarly, perform the same calculation for the high frequency subband of the 1st scale, and assume the calculated high frequency subband energy is 16. Add the two energy values together to obtain the total high frequency subband energy for the 2nd scale in the horizontal direction, which is 36. Repeat this calculation process for all three scales and all three directions, i.e. horizontal, vertical, diagonal, for a total of nine pairs of subbands, to obtain nine total high frequency subband energies. In an alternative embodiment, in S2, for any scale , calculate the high frequency subband energy

[0033] , , for the horizontal, vertical, and diagonal directions, respectively; calculate the arithmetic mean and the standard deviation of the high frequency subband energies for the three directions; calculate the directionality index for the corresponding scale as the ratio of the standard deviation to the arithmetic mean, satisfying . Specifically, the directionality index is used to determine whether the chroma noise at a particular scale is uniformly distributed or has a significant directionality. Using the example from the previous step, focus on the 2nd scale, and assume that the total high frequency subband energies for the three directions at this scale have been calculated to be: 36 for the horizontal direction , 38 for the vertical direction

[0034] , and 34 for the diagonal direction . Calculate the arithmetic mean of the three high frequency subband energies , which is 36. Calculate the squared differences of the three high frequency subband energies from the arithmetic mean, which are 0, 4, and 4, respectively. Take the average of the three squared differences (i.e. the variance), which is approximately 2.67. Take the square root to obtain the standard deviation , which is approximately 1.63. Calculate the directionality index D(2) for the 2nd scale by dividing the standard deviation by the arithmetic mean, which is approximately 0.045. A lower directionality index indicates that the noise energy is more uniformly distributed across the three directions at the 2nd scale, and the noise is isotropic. Similarly, perform the above calculations for the 1st scale and the 3rd scale to obtain the corresponding directionality indices D(l) and D(3), respectively.

[0035]

[0036] ​​​It should be noted that before the calculation, the arithmetic mean value is judged, and if its value is less than a very small positive number (such as ), the value of is directly set to 0.

[0037] S3, based on the arctangent function of the difference between the kurtosis coefficient and the preset reference coefficient, determines the determination threshold of the directionality index; when the directionality index is less than the determination threshold, the sum of the high-frequency sub-band energies of the three directions under the corresponding scale is taken as the chroma noise energy; otherwise, the chroma noise energy is set to a preset low value.

[0038] At the k scale, the calculated directionality index D(k) is compared with the determination threshold T: if D(k) is less than T, it indicates that the noise energy at this scale is more uniformly distributed in each direction, and it is determined to be the noise caused by carbon deposition, and the chroma noise energy at this scale is equal to the sum of the high-frequency sub-band energies of the three directions at this scale. If D(k) ≥ T, it indicates that the noise energy has obvious directionality, and it is determined to be an interference such as a scratch, and the chroma noise energy at this scale is set to a very small positive number, such as Figure 2 .

[0039] In an optional embodiment, in S3, the determination threshold T is calculated by the following formula: , wherein is the kurtosis coefficient of the global brightness histogram of the RGB image, is the preset reference coefficient, is the scaling coefficient, is the bias coefficient.

[0040] Specifically, the kurtosis coefficient K of the global brightness histogram of the input image needs to be calculated, and the kurtosis represents the sharpness of the data distribution. Assuming that for an input image, by analyzing the brightness value distribution of all pixels of the image, the calculated kurtosis coefficient K is 4.2. The preset reference coefficient is set to 3.0, the scaling coefficient is 0.15, and the bias coefficient is 0.25. Substituting into the formula, the determination threshold T is calculated to be 0.381. The determination threshold T is compared with the calculated directionality index D(k) of each scale to make a judgment.

[0041] , wherein the preset reference coefficient can be determined in the following way: a plurality of images of the inner wall of the combustion chamber under the condition of no carbon deposition or light carbon deposition are collected, the kurtosis coefficient of the global brightness histogram of each image is calculated, and the statistical median is taken as the preset reference coefficient .

[0042] In an alternative embodiment, the preset low value is 0.

[0043] Specifically, in the foregoing steps, a directionality indicator D(k) is calculated for each scale k, and a global decision threshold T is calculated. D(k) is compared with T, and if D(k) < T, the noise at the scale is considered to be isotropic, i.e. uniform, non-structured noise, and the noise energy is considered to be chrominance noise energy.

[0044] Suppose the directionality indicator D(2) at the 2nd scale is less than the decision threshold T. It is determined that the noise at the 2nd scale is chrominance noise, and at this time, the chrominance noise energy corresponding to the 2nd scale is set to a preset low value, and the arithmetic mean value calculated at the 2nd scale is forcibly cleared to zero. The purpose of this operation is to ignore or weaken the contribution of the chrominance noise identified as uniform and non-directional in the noise evaluation, because such noise usually has less impact on visual quality or is irrelevant to image texture.

[0045] In other embodiments, the preset low value can be set to a small positive number close to zero, for example, the preset low value can be .

[0046] S4, the chrominance noise energies obtained at each scale are weighted and summed in combination with the center frequencies corresponding to each scale to obtain a comprehensive noise index representing the degree of carbon deposition.

[0047] Specifically, the chrominance noise energies calculated at the three scales , , are weighted and summed. Since carbon deposition texture is more pronounced in the medium and high frequency bands, and the low frequency band is mostly background light or smooth area, the weights of each scale are positively correlated with the center frequencies of the corresponding scales. Assuming that the center frequencies of the 1st, 2nd and 3rd scales are , the weights decrease in turn from high to low frequencies. The comprehensive noise index I is calculated by the formula . The greater the value of the comprehensive noise index I, the more serious the degree of carbon deposition on the inner wall of the combustion chamber.

[0048] In an alternative embodiment, in S4, the chrominance noise energies obtained at the 1st, 2nd and 3rd scales are obtained, respectively denoted as , , ; According to the order from high to low of the center frequencies corresponding to each scale, the weighting coefficients are set for the corresponding scales , , ; Calculate the overall noise index ,satisfy .

[0049] Specifically, the chromaticity noise energy is collected after judgment and processing at all scales: assuming that after comparison with the judgment threshold, the noise at scale 1 is judged to be anisotropic, the chromaticity noise energy... The average energy value of the original calculation is retained as 50; the noise at the second scale is determined to be isotropic, and the chromaticity noise energy... Set to 0; the noise at the third scale is also determined to be anisotropic, and the chromaticity noise energy... Retain the original value of 15.

[0050] The calculation is performed based on preset weighting coefficients, which reflect the differences in the impact of noise of different frequencies on human vision. High-frequency noise is usually more easily perceived and therefore has the highest weight. Weighting of the first scale. The weight of the second scale is 0.5. The weight of the third scale is 0.3. The value is 0.2. Multiplying the chromaticity noise energy at each scale by the corresponding weighting coefficient and summing all the results, the overall noise index I is 28.

[0051] An embodiment of the machine vision-based combustion chamber wall carbon deposit detection system provided by the present invention: like Figure 3 As shown, the machine vision-based combustion chamber wall carbon deposit detection system includes a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement the aforementioned machine vision-based combustion chamber wall carbon deposit detection method.

[0052] The machine vision-based combustion chamber wall carbon deposit detection system also includes other components well known to those skilled in the art, such as communication interfaces. Their settings and functions are known in the art and will not be described in detail here.

[0053] In addition, in the description of this specification, "multiple" means at least two, such as two, three or more, etc., unless otherwise expressly and specifically defined.

Claims

1. A machine vision-based method for detecting carbon deposits on the inner walls of a combustion chamber, characterized in that, Includes the following steps: S1. Obtain the RGB image of the inner wall of the combustion chamber, calculate the kurtosis coefficient of the global brightness histogram of the RGB image, and perform nonlinear function weighting on the R, G, and B components of each pixel based on the ratio of the neighborhood brightness mean of each pixel in the RGB image to the preset brightness benchmark to obtain the first chromaticity component and the second chromaticity component. S2, perform two-dimensional discrete wavelet transform on the first chromaticity component and the second chromaticity component respectively to obtain high-frequency sub-bands in the horizontal, vertical and diagonal directions at at least two scales; at the same scale, summarize the high-frequency sub-band energy of the first chromaticity component and the second chromaticity component in the horizontal, vertical and diagonal directions to obtain the energy distribution in the three directions at the corresponding scale, and calculate the directional index representing the degree of anisotropy of chromaticity noise based on the energy distribution. S3, determine the judgment threshold of the directional index based on the arctangent function of the difference between the kurtosis coefficient and the preset benchmark coefficient; When the directionality index is less than the judgment threshold, the sum of the high-frequency subband energies in the three directions at the corresponding scale is taken as the chromaticity noise energy. Conversely, the chromaticity noise energy is set to a preset low value; S4, the chromatic noise energy obtained at each scale is weighted and summed with the center frequency corresponding to each scale to obtain the comprehensive noise index representing the degree of carbon deposition.

2. The method for detecting carbon deposits on the inner wall of a combustion chamber based on machine vision according to claim 1, characterized in that, In S1, for any pixel in the RGB image, a 5×5 neighborhood window is constructed, and the arithmetic mean of the brightness of all pixels within this neighborhood window is calculated as the neighborhood brightness mean. ; Calculate the weighting coefficients ,in, The preset brightness reference; The R, G, and B components of the pixel are multiplied by the weighting coefficients respectively. The weighted components are obtained. , , ; The first chromaticity component is calculated based on the weighted components. Second chromaticity component .

3. The method for detecting carbon deposits on the inner wall of a combustion chamber based on machine vision according to claim 1, characterized in that, In S2, the Daubechies4 wavelet basis is used to perform three-level two-dimensional discrete wavelet decomposition on the first and second chromaticity components to obtain the horizontal high-frequency subband, vertical high-frequency subband and diagonal high-frequency subband at the first, second and third scales respectively.

4. The method for detecting carbon deposits on the inner wall of a combustion chamber based on machine vision according to claim 3, characterized in that, In S2, for any scale and direction Calculate the first chromaticity component in the direction The average of the absolute values ​​of all wavelet coefficients in the high-frequency subband is used to obtain the high-frequency subband energy of the first chromaticity component. ; Calculate the second chromaticity component in the direction The average of the absolute values ​​of all wavelet coefficients in the high-frequency subband is used to obtain the high-frequency subband energy of the second chromaticity component. ; The high-frequency subband energy of the first chromaticity component High-frequency subband energy of the second chromaticity component By adding them together, we obtain the total high-frequency subband energy at that scale and in that direction. ,satisfy .

5. The method for detecting carbon deposits on the inner wall of a combustion chamber based on machine vision according to claim 4, characterized in that, In S2, for any scale Calculate the high-frequency subband energy in the horizontal, vertical, and diagonal directions respectively. , , ; Calculate the arithmetic mean of the high-frequency subband energies in three directions. and standard deviation ; Calculate the corresponding scale Directional indicators below Let be the ratio of the standard deviation to the arithmetic mean, satisfying . .

6. The method for detecting carbon deposits on the inner wall of a combustion chamber based on machine vision according to claim 1, characterized in that, In S3, the decision threshold T is calculated using the following formula: ,in, The kurtosis coefficient is the kurtosis coefficient of the global luminance histogram of the RGB image. As a preset baseline coefficient, This is the scaling factor. This is the bias coefficient.

7. The method for detecting carbon deposits on the inner wall of a combustion chamber based on machine vision according to claim 6, characterized in that, The preset low value is 0.

8. The method for detecting carbon deposits on the inner wall of a combustion chamber based on machine vision according to any one of claims 1-7, characterized in that, In S4, the chromaticity noise energy obtained at the 1st, 2nd, and 3rd scales is denoted as follows: , , ; Set weighting coefficients for each scale. , , The weighting coefficients are positively correlated with the center frequency of the corresponding scale; Calculate the overall noise index ,satisfy .

9. The method for detecting carbon deposits on the inner wall of a combustion chamber based on machine vision according to claim 8, characterized in that, 、 、 。 10. A machine vision-based combustion chamber wall carbon deposit detection system, characterized in that, It includes a memory and a processor, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the machine vision-based method for detecting carbon deposits on the inner wall of a combustion chamber as described in any one of claims 1-9 is implemented.

Images