Port electromechanical equipment detection method and system based on artificial intelligence

Abstract

The invention is suitable for the technical field of X-ray machine imaging optimization, and provides a port electromechanical equipment detection method and system based on artificial intelligence, and the method comprises the steps: obtaining original images and compensation images of a preset number of detected objects shot by a target X-ray imaging machine, the compensation image is collected based on a preset exposure delay parameter. According to the invention, by determining the complexity of the reference structure and dynamically adjusting the exposure delay parameter in the subsequent detection, the problem of imaging quality reduction caused by equipment aging can be effectively solved. The extraction of the complexity of the reference structure enables the equipment to clearly determine the adaptation range of the originally set exposure delay parameter when detecting a complex object; the dynamic adjustment mechanism further optimizes the exposure delay parameter according to the degree of deviation between the object to be measured and the reference structure complexity, thereby ensuring the imaging quality of the complex object.

Description

Technical Field

[0001] The present invention belongs to the technical field of X-ray machine imaging optimization, and in particular relates to a port electromechanical equipment detection method and system based on artificial intelligence. Background Art

[0002] At present, X-ray imaging equipment is widely used in the inspection of electromechanical equipment at ports to achieve rapid non-destructive inspection of complex items such as cargo and luggage. This type of equipment analyzes the internal structure and characteristics through the image generated by the rays after they penetrate the object. However, during the long-term operation of the equipment, due to the aging of the ray source, the intensity and stability of its ray output gradually decrease, resulting in a significant degradation of the imaging quality. To this end, the existing technology usually compensates for the impact of equipment aging by extending the exposure time (i.e., exposure delay parameters) or increasing the detection sensitivity. This method can improve the imaging effect to a certain extent, but it mostly relies on fixed or empirically set compensation parameters, and it is difficult to adapt to the diverse characteristics of the objects being tested. In particular, for objects with complex structures, the effect is often unsatisfactory.

[0003] A significant problem with existing technologies is that the compensation strategy of aging equipment lacks a dynamic response mechanism to the complexity of the internal structure of objects, resulting in the inability to accurately adjust imaging parameters to meet the needs of different objects. Complex objects have higher requirements for imaging clarity due to their multi-level and high-density structural characteristics, while the imaging of simple objects does not require too much compensation. This lack of adaptability not only wastes resources, but may also cause blurred imaging of complex objects or loss of key details. In addition, most existing technologies are based on global parameter optimization of equipment and lack real-time dynamic adjustment methods for the characteristics of a single object, which greatly restricts detection accuracy and efficiency. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for detecting electromechanical equipment at a port based on artificial intelligence, aiming to solve the problems raised in the background technology.

[0005] The present invention is implemented as follows: a method for detecting electromechanical equipment at a port based on artificial intelligence, the method comprising:

[0006] Acquire original images and compensated images taken by the target X-ray imaging machine of a preset number of objects to be tested, wherein the original images are acquired based on original exposure parameters, and the compensated images are acquired based on preset exposure delay parameters;

[0007] Processing the original image of each object to be tested, calculating its internal structure complexity, arranging a preset number of objects to be tested according to the size of the internal structure complexity, and generating a sorting sequence of the internal structure complexity;

[0008] Compare and analyze the original image and the compensated image of each object in the sorting sequence, calculate the difference in edge clarity between the two, match it with the internal structure complexity, and draw a relationship curve between the internal structure complexity and the edge clarity difference;

[0009] Analyze the relationship curve, identify specific turning points and extract the internal structure complexity corresponding to the specific turning points, which is defined as the benchmark structure complexity;

[0010] In subsequent inspections, the internal structure complexity of the object to be tested entering the target X-ray imaging machine is calculated in real time and compared with the baseline structure complexity. If it exceeds the baseline structure complexity, the preset exposure delay parameters are dynamically adjusted according to the degree of deviation to generate an optimized imaging image.

[0011] As a further limitation of the technical solution of the embodiment of the present invention, the steps of processing the original image of each tested object, calculating the internal structure complexity thereof, arranging a preset number of tested objects according to the size of the internal structure complexity, and generating a sorting sequence of the internal structure complexity include:

[0012] Processing the original image of each object under test, using an image analysis algorithm to extract feature data reflecting the complexity of the object structure, the feature data including edge density, texture distribution or other image indicators related to the object structure;

[0013] Based on the extracted feature data, these features are converted into a numerical internal structure complexity index through weighted analysis of comprehensive features, where the weighted analysis comprehensively evaluates the contribution of the complexity according to the importance of the features;

[0014] The complexity indicators of all tested objects are compared and arranged from small to large according to the complexity values ​​to generate a sequence sorted by complexity.

[0015] As a further limitation of the technical solution of the embodiment of the present invention, the steps of comparing and analyzing the original image and the compensated image of each tested object in the sorting sequence in turn, calculating the edge clarity difference value between the two, and matching it with the internal structure complexity, and drawing a relationship curve between the internal structure complexity and the edge clarity difference include:

[0016] Determine the edge clarity of the original image and the compensated image of each object under test based on edge detection algorithm and image processing technology, and subtract the edge clarity of the original image from the edge clarity of the compensated image to obtain an edge clarity difference value;

[0017] According to the internal structure complexity of each tested object and the calculated edge definition difference value, a corresponding relationship is established between the two to form a data pair with the internal structure complexity as the independent variable and the edge definition difference value as the dependent variable;

[0018] All matching data pairs are plotted into scatter plots in order of internal structure complexity from small to large, and based on the scatter plots, a relationship curve between internal structure complexity and edge clarity difference is generated through a fitting algorithm.

[0019] As a further limitation of the technical solution of the embodiment of the present invention, the steps of analyzing the relationship curve, identifying a specific turning point and extracting the internal structure complexity corresponding to the specific turning point, and defining it as the benchmark structure complexity include:

[0020] Observe the overall trend of the relationship curve, identify the transition point from a flat state to a clear downward trend, and determine the starting inflection point of this position as a specific turning point;

[0021] The internal structural complexity corresponding to a specific turning point is extracted and defined as the benchmark structural complexity.

[0022] As a further limitation of the technical solution of the embodiment of the present invention, in subsequent detection, the internal structure complexity of the object to be tested entering the target X-ray imaging machine is calculated in real time and compared with the reference structure complexity. If the reference structure complexity is exceeded, the preset exposure delay parameter is dynamically adjusted according to the degree of deviation to generate an optimized imaging image. The steps include:

[0023] When the object to be tested enters the target X-ray imaging machine, its original image is collected and its internal structure complexity is determined based on the image analysis algorithm;

[0024] Compare the internal structure complexity of the object to be tested with the reference structure complexity. If the internal structure complexity is greater than the reference structure complexity, call the exposure delay parameter optimization formula to calculate the optimized exposure delay parameter.

[0025] The image of the object to be tested is recaptured using the optimized exposure delay parameters to generate an imaging image with optimized quality.

[0026] As a further limitation of the technical solution of the embodiment of the present invention, the exposure delay parameter optimization formula is:

[0027]

[0028] Among them, T opt is the optimized exposure delay parameter, T preset is the preset exposure delay parameter, C real is the internal structure complexity of the object to be tested, C base is the baseline structural complexity, is the deviation degree of the internal structure complexity of the object to be tested compared with the reference structure complexity, and K is the adjustment coefficient.

[0029] A port electromechanical equipment detection system based on artificial intelligence, the system comprises: an image acquisition module, a sequence generation module, a relationship curve generation module, a reference structure complexity determination module and an exposure delay parameter adjustment module, wherein:

[0030] An image acquisition module, used to acquire original images and compensated images taken by a target X-ray imaging machine of a preset number of objects under test, wherein the original images are acquired based on original exposure parameters, and the compensated images are acquired based on preset exposure delay parameters;

[0031] A sequence generation module is used to process the original image of each object to be tested, calculate its internal structure complexity, arrange a preset number of objects to be tested according to the size of the internal structure complexity, and generate a sorting sequence of the internal structure complexity;

[0032] A relationship curve generation module is used to compare and analyze the original image and the compensated image of each tested object in the sorting sequence in turn, calculate the edge clarity difference value between the two, match it with the internal structure complexity, and draw a relationship curve between the internal structure complexity and the edge clarity difference;

[0033] A benchmark structural complexity determination module is used to analyze the relationship curve, identify specific turning points and extract the internal structural complexity corresponding to the specific turning points, which is defined as the benchmark structural complexity;

[0034] The exposure delay parameter adjustment module is used to calculate the internal structure complexity of the object to be tested entering the target X-ray imaging machine in real time during subsequent inspections, and compare it with the baseline structure complexity. If the baseline structure complexity is exceeded, the preset exposure delay parameters are dynamically adjusted according to the degree of deviation to generate an optimized imaging image.

[0035] As a further limitation of the technical solution of the embodiment of the present invention, the sequence generation module specifically includes:

[0036] An image processing unit is used to process the original image of each object under test and extract feature data reflecting the complexity of the object structure by using an image analysis algorithm. The feature data includes edge density, texture distribution or other image indicators related to the object structure.

[0037] A feature data processing unit, for converting the extracted feature data into a numerical internal structure complexity index through weighted analysis of the comprehensive features, wherein the weighted analysis comprehensively evaluates the contribution of the feature to the complexity according to its importance;

[0038] The sequence generation unit is used to compare the complexity indicators of all the tested objects and arrange them from small to large according to the complexity values ​​to generate a sequence sorted by complexity.

[0039] As a further limitation of the technical solution of the embodiment of the present invention, the relationship curve generation module specifically includes:

[0040] An edge definition difference value calculation unit is used to determine the edge definition of the original image and the compensated image of each object under test based on an edge detection algorithm and an image processing technology, and to obtain an edge definition difference value by subtracting the edge definition of the original image from the edge definition of the compensated image;

[0041] A data pair generating unit is used to establish a corresponding relationship between the internal structure complexity of each tested object and the calculated edge definition difference value, so as to form a data pair with the internal structure complexity as the independent variable and the edge definition difference value as the dependent variable;

[0042] The relationship curve generating unit is used to draw all matching data pairs into a scatter plot in the order of internal structure complexity from small to large, and based on the scatter plot, generate a relationship curve between the internal structure complexity and the edge clarity difference through a fitting algorithm.

[0043] As a further limitation of the technical solution of the embodiment of the present invention, the reference structure complexity determination module specifically includes:

[0044] A specific turning point determination unit is used to observe the overall trend of the relationship curve, identify the position where the trend transitions from a flat state to an obvious downward trend, and determine the starting inflection point of the position as a specific turning point;

[0045] The benchmark structural complexity determination unit is used to extract the internal structural complexity corresponding to a specific turning point and define it as the benchmark structural complexity.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] The present invention proposes an optimization method based on the correlation between internal structure complexity and exposure delay by analyzing the imaging characteristics of X-ray imaging equipment under the conditions of originally set exposure delay parameters based on aging. By determining the baseline structure complexity and dynamically adjusting the exposure delay parameters in subsequent detections, the present invention can effectively address the problem of image quality degradation caused by equipment aging. The extraction of the baseline structure complexity enables the equipment to clearly define the adaptation range of the originally set exposure delay parameters when detecting complex objects; the dynamic adjustment mechanism further optimizes the exposure delay parameters according to the degree of deviation between the object to be tested and the baseline structure complexity, thereby ensuring the imaging quality of complex objects.

[0048] The present invention effectively improves the adaptability and imaging reliability of aging equipment in complex detection scenarios, while reducing imaging blur and detail loss caused by aging problems. Through the dynamic adjustment mechanism, it can also optimize resource utilization while meeting imaging requirements, extend the service life of the equipment, and provide an effective solution for the intelligent upgrade of the imaging system. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A flowchart of a method provided by an embodiment of the present invention;

[0050] Figure 2 A flow chart of generating a sorting sequence related to internal structure complexity in a method provided in an embodiment of the present invention;

[0051] Figure 3 A flow chart of drawing a relationship curve between internal structure complexity and edge definition difference in the method provided by an embodiment of the present invention;

[0052] Figure 4 A flow chart of determining the complexity of a reference structure in a method provided in an embodiment of the present invention;

[0053] Figure 5 A flow chart of optimizing preset exposure delay parameters in the method provided in an embodiment of the present invention;

[0054] Figure 6 An application architecture diagram of a system provided by an embodiment of the present invention;

[0055] Figure 7 A structural block diagram of a sequence generation module in a system provided by an embodiment of the present invention;

[0056] Figure 8 A structural block diagram of a relationship curve generating module in a system provided by an embodiment of the present invention;

[0057] Fig. 9 This is a structural block diagram of a reference structure complexity determination module in a system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0059] Figure 1 A flow chart of a method provided by an embodiment of the present invention is shown.

[0060] Specifically, a port electromechanical equipment detection method based on artificial intelligence, the method specifically comprises the following steps:

[0061] Step S100, obtaining original images and compensated images taken by a target X-ray imaging machine of a preset number of objects to be tested, wherein the original images are collected based on original exposure parameters, and the compensated images are collected based on preset exposure delay parameters.

[0062] In the embodiment of the present invention, the object to which the present invention is applied is an X-ray imaging machine that is in an aging state, which means that after a long period of use, the output performance of the radiation source of the device decreases, resulting in a certain degree of degradation of the imaging quality. Therefore, compensation means are needed to optimize the imaging effect of the device.

[0063] The basis for the preset number is to cover a sufficient variety of objects to be tested, to ensure that the collected data is representative, and thus accurately reflect the imaging characteristics of the X-ray machine under different object complexities. Usually, the selection of the preset number will take into account the wide range of materials, shapes, densities, and structural complexity of the objects to be tested, so that the subsequent analysis is statistically significant.

[0064] The original exposure parameters refer to the default exposure settings of the X-ray machine under normal working conditions, which are usually pre-set by the equipment manufacturer based on imaging requirements and X-ray source performance. These parameters usually include exposure time, X-ray intensity, etc., which are used to ensure imaging quality under standard conditions.

[0065] The preset exposure delay parameter is an adjustment parameter set to compensate for the attenuation of the X-ray output when the X-ray machine is aged. It mainly compensates for the lack of X-ray energy caused by aging by extending the exposure time. This parameter is set based on the evaluation of the aging degree of the X-ray source and the results of optimizing the imaging effect under experimental conditions. Through appropriate delay parameters, the imaging quality of the aging X-ray machine in complex object inspection can be improved to approach or reach the initial performance standard of the equipment.

[0066] Furthermore, the artificial intelligence-based port electromechanical equipment detection method further includes the following steps:

[0067] Step S200 , processing the original image of each object to be tested, calculating the internal structure complexity thereof, arranging a preset number of objects to be tested according to the size of the internal structure complexity, and generating a sorting sequence of the internal structure complexity.

[0068] Specifically, Figure 2 A flow chart for generating a sorted sequence related to the internal structure complexity is shown.

[0069] The steps of processing the original image of each object to be tested, calculating the internal structure complexity thereof, arranging a preset number of objects to be tested according to the size of the internal structure complexity, and generating a sorting sequence of the internal structure complexity specifically include the following steps:

[0070] Step S201, processing the original image of each object to be tested, and using an image analysis algorithm to extract feature data reflecting the complexity of the object structure, wherein the feature data includes edge density, texture distribution or other image indicators related to the object structure;

[0071] Step S202, based on the extracted feature data, these features are converted into a numerical internal structure complexity index through weighted analysis of the comprehensive features, wherein the weighted analysis comprehensively evaluates the contribution of the complexity according to the importance of the features;

[0072] Step S203, compare the complexity indexes of all the tested objects, and arrange them from small to large according to the complexity values ​​to generate a sequence sorted by complexity.

[0073] In an embodiment of the present invention, the calculation and sorting of the internal structure complexity can be completed with the help of existing image processing technology and algorithms. For each original image of the tested object, an edge detection algorithm, such as the Canny algorithm or the Sobel algorithm, can be used to extract the edge features of the object by identifying the grayscale changes of the pixels, and further statistically analyze the density and distribution of the edges to quantify their edge complexity. At the same time, texture analysis methods can be used to mine the texture characteristics in the image, such as analyzing the directionality, uniformity or roughness of the texture through the gray level co-occurrence matrix (GLCM), thereby enriching the description of the structural complexity.

[0074] When comprehensively analyzing the extracted multidimensional features, a weight model can be combined to evaluate the contribution of each feature to complexity. This feature weighted analysis can normalize the features according to their importance, combining indicators such as edge density and texture distribution into a numerical internal structure complexity indicator. This step can use existing algorithm tool libraries, such as Python's Scikit-learn, OpenCV, or Matlab toolboxes, to ensure the accuracy and consistency of feature calculations.

[0075] After the calculation of the complexity index is completed, all the tested items can be sorted by numerical comparison to generate a complexity sorting sequence from small to large. The sorting process relies on standard sorting algorithms or existing programming language libraries, which can efficiently process and generate the final sorting results. These methods are all derived from mature technical systems and have high operability and implementation efficiency.

[0076] The above method relies on technical means widely used in the fields of image processing and data analysis. Its implementation logic and operation process are highly consistent with existing practices and have strong engineering adaptability.

[0077] Furthermore, the artificial intelligence-based port electromechanical equipment detection method further includes the following steps:

[0078] Step S300, compare and analyze the original image and the compensated image of each tested object in the sorting sequence in turn, calculate the edge clarity difference value between the two, match it with the internal structure complexity, and draw a relationship curve between the internal structure complexity and the edge clarity difference.

[0079] Specifically, Figure 3 A flow chart for plotting the relationship between internal structure complexity and edge sharpness difference is shown.

[0080] The original image and the compensated image of each object under test in the sorting sequence are compared and analyzed in turn, the edge definition difference value between the two is calculated, and the difference is matched with the internal structure complexity, and the relationship curve between the internal structure complexity and the edge definition difference is drawn, which specifically includes the following steps:

[0081] Step S301, determining edge clarity of the original image and the compensated image of each object under test based on an edge detection algorithm and image processing technology, and subtracting the edge clarity of the original image from the edge clarity of the compensated image to obtain an edge clarity difference value;

[0082] Step S302, establishing a corresponding relationship between the internal structure complexity of each tested object and the calculated edge definition difference value, to form a data pair with the internal structure complexity as the independent variable and the edge definition difference value as the dependent variable;

[0083] Step S303, plotting all matched data pairs into a scatter plot in order of internal structure complexity from small to large, and generating a relationship curve between internal structure complexity and edge definition difference through a fitting algorithm based on the scatter plot.

[0084] In an embodiment of the present invention, the edge clarity of the original image and the compensated image of each object under test is determined based on the edge detection algorithm and the image processing technology, which can be completed by preprocessing the image, extracting the edge and calculating the clarity. First, the original image and the compensated image are subjected to noise suppression and grayscale standardization processing to reduce the noise interference that may exist in the image and improve the processing consistency. Subsequently, an edge detection algorithm, such as the Canny algorithm or the Sobel algorithm, is applied to extract the edge features in the image, and the number of edge pixels, the edge distribution density or the edge gradient amplitude are counted as indicators of clarity. These indicators reflect the sharpness and clarity of the image edge and can be used for comparative analysis.

[0085] The logic of subtracting the edge clarity of the compensated image from the edge clarity of the original image to obtain the edge clarity difference value is that the compensated image should be clearer than the original image after optimization. This difference can reflect the improvement of the image clarity caused by the compensation effect and quantify the degree of clarity improvement.

[0086] According to the internal structure complexity of each tested object and the calculated edge clarity difference value, a corresponding relationship is established between the two. A pair of data points can be formed by matching the structure complexity value of each tested object with its clarity difference value. All data points are organized with the internal structure complexity as the independent variable and the edge clarity difference value as the dependent variable, and are sorted from small to large in order of internal structure complexity to ensure the logic of the relationship and the continuity of the curve. By plotting these data points into a scatter plot and applying a fitting algorithm (such as linear regression or polynomial regression), a relationship curve between internal structure complexity and edge clarity difference is generated, which can intuitively reflect the influence of internal structure complexity on the compensation effect.

[0087] Furthermore, the artificial intelligence-based port electromechanical equipment detection method further includes the following steps:

[0088] Step S400 , analyzing the relationship curve, identifying a specific turning point and extracting the internal structure complexity corresponding to the specific turning point, and defining it as a reference structure complexity.

[0089] Specifically, Figure 4 A flow chart for determining a baseline structural complexity is shown.

[0090] Among them, analyzing the relationship curve, identifying the specific turning point and extracting the internal structure complexity corresponding to the specific turning point, and defining it as the benchmark structure complexity specifically includes the following steps:

[0091] Step S401, observing the overall trend of the relationship curve, identifying the position where the trend transitions from a flat state to an obvious downward trend, and determining the starting inflection point of the position as a specific turning point;

[0092] Step S402: extract the internal structure complexity corresponding to the specific turning point and define it as the benchmark structure complexity.

[0093] In an embodiment of the present invention, to achieve "observing the overall trend of the relationship curve, identifying the position of transition from a gentle state to a clear downward trend, and determining the starting inflection point of the position as a specific turning point", it can be accomplished by analyzing the data characteristics and change rules of the curve. First, the relationship curve between the internal structure complexity and the edge clarity difference value is mathematically modeled, and a fitting algorithm is usually used to generate a continuous curve (such as polynomial fitting). Next, the first-order derivative of the curve is used to calculate the rate of change, or the position of the morphological change of the curve is determined by curvature analysis to identify the turning point from a gentle state to a clear downward trend. Specifically, the derivative value of the curve at the turning point will change significantly, such as gradually changing from close to zero to a negative value, or the curvature will have a maximum value, which can be used as a basis for judging the turning point.

[0094] Determining the "specific turning point" is of great significance because it reflects the critical point of the compensation optimization effect on image clarity improvement. In other words, the turning point indicates that the internal structure complexity of the object has reached a certain level, at which point the effect of delay compensation begins to significantly weaken, and even cannot compensate for the image quality degradation caused by aging. This critical information helps define the optimization capability range of the equipment and provides a basis for dynamically adjusting the exposure delay parameters.

[0095] The internal structure complexity corresponding to the "specific turning point" is defined as the "benchmark structure complexity", which can provide a standard reference value for subsequent testing. The benchmark structure complexity represents the upper limit of complexity that the original exposure parameters can effectively adapt to. When the complexity of the subsequent objects to be tested exceeds this value, the delayed exposure parameters can be adjusted according to the degree of deviation to ensure that the imaging quality of complex objects remains within an acceptable range, thereby optimizing the reliability and accuracy of the test results.

[0096] Furthermore, the artificial intelligence-based port electromechanical equipment detection method further includes the following steps:

[0097] Step S500, in subsequent detection, the internal structure complexity of the object to be tested entering the target X-ray imaging machine is calculated in real time and compared with the baseline structure complexity. If it exceeds the baseline structure complexity, the preset exposure delay parameters are dynamically adjusted according to the degree of deviation to generate an optimized imaging image.

[0098] Specifically, Figure 5 A flow chart for optimizing preset exposure delay parameters is shown.

[0099] Among them, in the subsequent detection, the internal structure complexity of the object to be tested entering the target X-ray imaging machine is calculated in real time and compared with the reference structure complexity. If the reference structure complexity is exceeded, the preset exposure delay parameters are dynamically adjusted according to the degree of deviation to generate an optimized imaging image. Specifically, the following steps are included:

[0100] Step S501, when the object to be tested enters the target X-ray imaging machine, its original image is collected and its internal structure complexity is determined based on the image analysis algorithm;

[0101] Step S502, comparing the internal structure complexity of the object to be tested with the reference structure complexity, if the internal structure complexity is greater than the reference structure complexity, calling the exposure delay parameter optimization formula to calculate the optimized exposure delay parameter;

[0102] Step S503: re-capture the image of the object to be tested using the optimized exposure delay parameters to generate an imaging image with optimized quality.

[0103] The exposure delay parameter optimization formula is:

[0104] Among them, T opt is the optimized exposure delay parameter, T preset is the preset exposure delay parameter, C real is the internal structure complexity of the object to be tested, C base is the baseline structural complexity, is the deviation degree of the internal structure complexity of the object to be tested compared with the reference structure complexity, and K is the adjustment coefficient.

[0105] In the embodiment of the present invention, after determining the previous reference structure complexity value, the subsequent real-time detection of the object to be tested is entered, and the entire process revolves around real-time calculation, comparison of reference values, and dynamic adjustment of parameters. The following is a detailed example description:

[0106] Assume that in the previous analysis, the baseline structural complexity has been determined to be 50, the preset exposure delay parameter is 10 milliseconds, and the adjustment factor k is set to 1.2. Now a new object to be tested enters the X-ray imaging machine.

[0107] The first step is to collect the original image of the object to be tested, and use the image analysis algorithm to extract its edge density and texture distribution and other features. Through comprehensive analysis, it is concluded that the internal structure complexity of the object is 70. Then, the system compares this complexity value with the benchmark structure complexity of 50 and calculates the degree of deviation of the complexity. The degree of deviation is calculated by first finding the difference between the complexity of the object to be tested and the benchmark complexity, that is, 70 minus 50, the result is 20; then divide this difference by the benchmark complexity of 50, and get a deviation ratio of 0.4. This shows that the complexity of the object is 40% higher than the benchmark value.

[0108] In the second step, the system calculates the delay ratio that needs to be adjusted based on this deviation ratio and the adjustment coefficient k=1.2. Specifically, the system adds 1 to the deviation ratio 0.4 to get an adjustment base of 1.4. Then, this adjustment base is multiplied by the adjustment coefficient k to get a final adjustment factor of 1.68. This means that the original preset exposure delay parameters need to be increased by 68% to accommodate the higher complexity of the object to be tested.

[0109] In the third step, the system multiplies the calculated adjustment factor 1.68 by the preset exposure delay parameter 10 milliseconds, and the optimized exposure delay parameter is 16.8 milliseconds. This value represents the compensated exposure time, which is used to ensure that complex objects can obtain higher quality imaging effects.

[0110] The fourth step is to re-image the object under test using the optimized exposure delay parameter of 16.8 milliseconds. The re-captured image shows that the edge clarity of the object is significantly improved, and the internal structure details are clearer, meeting the imaging quality requirements. The system then records this optimization result for further adjustment reference in subsequent inspections.

[0111] The specific solution introduced in the present invention is based on the connection between "internal structure complexity" and "compensation for exposure delay" to optimize the problem of reduced imaging quality of aging equipment. Similar ideas can also be applied to other directions. For example, in medical imaging equipment, imaging parameters can be optimized based on the connection between tissue density and X-ray exposure time to improve the detail clarity of diagnostic images. It can also be applied to industrial detection to adjust the detection signal strength based on the relationship between material thickness and ultrasonic signal attenuation to enhance the detection effect of complex structures. The above method fully reflects the flexible application of the connection between parameter adjustment and target characteristics.

[0112] In summary, the present invention combines the relationship between "internal structure complexity" and "compensation for exposure delay" to propose a dynamic optimization method for aging X-ray imaging equipment at ports, which effectively improves the imaging quality and system adaptability. Based on the image analysis and characteristic modeling of preset objects, the baseline structural complexity of the equipment under aging conditions is clarified, and the exposure delay parameters are dynamically adjusted through real-time detection to achieve accurate compensation for complex objects. This method can solve the problem of poor imaging effect of traditional equipment on highly complex objects in an aging state, avoid the effects of blurred imaging, loss of details, etc., and reduce unnecessary energy waste.

[0113] The benefits of this solution also lie in its wide applicability and efficient implementation. Through image processing algorithms and relationship curve analysis, a benchmark structural complexity model was constructed, and based on this, the exposure parameters were optimized in real time, taking into account both detection accuracy and operational efficiency. Especially in complex dynamic scenes, the solution can quickly identify the characteristics of the object to be tested, flexibly adjust the imaging settings, and ensure the stability and reliability of the imaging results. In addition, the technical framework is also highly scalable and can be applied to other fields, such as medical imaging and industrial inspection, providing new ideas and implementation paths for the intelligent and precise development of imaging systems. In summary, the present invention not only improves the service life and imaging capabilities of aging equipment, but also provides an efficient solution for complex detection tasks, and has important practical value.

[0114] Furthermore, Figure 6 The application architecture diagram of the system provided by the embodiment of the present invention is shown.

[0115] Among them, in another preferred embodiment provided by the present invention, a port electromechanical equipment detection system based on artificial intelligence includes:

[0116] The image acquisition module 100 is used to acquire original images and compensated images taken by a target X-ray imaging machine of a preset number of objects under test, wherein the original images are acquired based on original exposure parameters, and the compensated images are acquired based on preset exposure delay parameters.

[0117] In the embodiment of the present invention, the object to which the present invention is applied is an X-ray imaging machine that is in an aging state, which means that after a long period of use, the output performance of the radiation source of the device decreases, resulting in a certain degree of degradation of the imaging quality. Therefore, compensation means are needed to optimize the imaging effect of the device.

[0118] The basis for the preset number is to cover a sufficient variety of objects to be tested, to ensure that the collected data is representative, and thus accurately reflect the imaging characteristics of the X-ray machine under different object complexities. Usually, the selection of the preset number will take into account the wide range of materials, shapes, densities, and structural complexity of the objects to be tested, so that the subsequent analysis is statistically significant.

[0119] The original exposure parameters refer to the default exposure settings of the X-ray machine under normal working conditions, which are usually pre-set by the equipment manufacturer based on imaging requirements and X-ray source performance. These parameters usually include exposure time, X-ray intensity, etc., which are used to ensure imaging quality under standard conditions.

[0120] The preset exposure delay parameter is an adjustment parameter set to compensate for the attenuation of the X-ray output when the X-ray machine is aged. It mainly compensates for the lack of X-ray energy caused by aging by extending the exposure time. This parameter is set based on the evaluation of the aging degree of the X-ray source and the results of optimizing the imaging effect under experimental conditions. Through appropriate delay parameters, the imaging quality of the aging X-ray machine in complex object inspection can be improved to approach or reach the initial performance standard of the equipment.

[0121] Furthermore, the artificial intelligence-based port electromechanical equipment detection system also includes:

[0122] The sequence generation module 200 is used to process the original image of each tested object, calculate its internal structure complexity, arrange a preset number of tested objects according to the size of the internal structure complexity, and generate a sorted sequence of the internal structure complexity.

[0123] Specifically, Figure 7 It shows a structural block diagram of the sequence generation module 200 in the system provided by the embodiment of the present invention.

[0124] Among them, in the preferred embodiment provided by the present invention, the sequence generation module 200 specifically includes:

[0125] The image processing unit 201 is used to process the original image of each object under test and extract feature data reflecting the complexity of the structure of the object using an image analysis algorithm. The feature data includes edge density, texture distribution or other image indicators related to the structure of the object.

[0126] A feature data processing unit 202 is used to convert the extracted feature data into a numerical internal structure complexity index through weighted analysis of the comprehensive features, wherein the weighted analysis comprehensively evaluates the contribution of the feature to the complexity according to its importance;

[0127] The sequence generation unit 203 is used to compare the complexity indicators of all the tested objects, and arrange them from small to large according to the complexity values ​​to generate a sequence sorted by complexity.

[0128] In an embodiment of the present invention, the calculation and sorting of the internal structure complexity can be completed with the help of existing image processing technology and algorithms. For each original image of the tested object, an edge detection algorithm, such as the Canny algorithm or the Sobel algorithm, can be used to extract the edge features of the object by identifying the grayscale changes of the pixels, and further statistically analyze the density and distribution of the edges to quantify their edge complexity. At the same time, texture analysis methods can be used to mine the texture characteristics in the image, such as analyzing the directionality, uniformity or roughness of the texture through the gray level co-occurrence matrix (GLCM), thereby enriching the description of the structural complexity.

[0129] When comprehensively analyzing the extracted multidimensional features, a weight model can be combined to evaluate the contribution of each feature to complexity. This feature weighted analysis can normalize the features according to their importance, combining indicators such as edge density and texture distribution into a numerical internal structure complexity indicator. This step can use existing algorithm tool libraries, such as Python's Scikit-learn, OpenCV, or Matlab toolboxes, to ensure the accuracy and consistency of feature calculations.

[0130] After the calculation of the complexity index is completed, all the tested items can be sorted by numerical comparison to generate a complexity sorting sequence from small to large. The sorting process relies on standard sorting algorithms or existing programming language libraries, which can efficiently process and generate the final sorting results. These methods are all derived from mature technical systems and have high operability and implementation efficiency.

[0131] The above method relies on technical means widely used in the fields of image processing and data analysis. Its implementation logic and operation process are highly consistent with existing practices and have strong engineering adaptability.

[0132] Furthermore, the artificial intelligence-based port electromechanical equipment detection system also includes:

[0133] The relationship curve generation module 300 is used to compare and analyze the original image and the compensated image of each tested object in the sorting sequence in turn, calculate the edge clarity difference value between the two, match it with the internal structure complexity, and draw a relationship curve between the internal structure complexity and the edge clarity difference.

[0134] Specifically, Figure 8 It shows a structural block diagram of the relationship curve generating module 300 in the system provided by the embodiment of the present invention.

[0135] In a preferred embodiment of the present invention, the relationship curve generating module 300 specifically includes:

[0136] The edge definition difference value calculation unit 301 is used to determine the edge definition of the original image and the compensated image of each detected object based on the edge detection algorithm and the image processing technology, and subtract the edge definition of the original image from the edge definition of the compensated image to obtain the edge definition difference value;

[0137] A data pair generating unit 302 is used to establish a corresponding relationship between the internal structure complexity of each tested object and the calculated edge definition difference value, so as to form a data pair with the internal structure complexity as the independent variable and the edge definition difference value as the dependent variable;

[0138] The relationship curve generating unit 303 is used to plot all matching data pairs into a scatter plot in the order of internal structure complexity from small to large, and generate a relationship curve between internal structure complexity and edge definition difference through a fitting algorithm based on the scatter plot.

[0139] In an embodiment of the present invention, the edge clarity of the original image and the compensated image of each object under test is determined based on the edge detection algorithm and the image processing technology, which can be completed by preprocessing the image, extracting the edge and calculating the clarity. First, the original image and the compensated image are subjected to noise suppression and grayscale standardization processing to reduce the noise interference that may exist in the image and improve the processing consistency. Subsequently, an edge detection algorithm, such as the Canny algorithm or the Sobel algorithm, is applied to extract the edge features in the image, and the number of edge pixels, the edge distribution density or the edge gradient amplitude are counted as indicators of clarity. These indicators reflect the sharpness and clarity of the image edge and can be used for comparative analysis.

[0140] The logic of subtracting the edge clarity of the compensated image from the edge clarity of the original image to obtain the edge clarity difference value is that the compensated image should be clearer than the original image after optimization. This difference can reflect the improvement of the image clarity caused by the compensation effect and quantify the degree of clarity improvement.

[0141] According to the internal structure complexity of each tested object and the calculated edge clarity difference value, a corresponding relationship is established between the two. A pair of data points can be formed by matching the structure complexity value of each tested object with its clarity difference value. All data points are organized with the internal structure complexity as the independent variable and the edge clarity difference value as the dependent variable, and are sorted from small to large in order of internal structure complexity to ensure the logic of the relationship and the continuity of the curve. By plotting these data points into a scatter plot and applying a fitting algorithm (such as linear regression or polynomial regression), a relationship curve between internal structure complexity and edge clarity difference is generated, which can intuitively reflect the influence of internal structure complexity on the compensation effect.

[0142] Furthermore, the artificial intelligence-based port electromechanical equipment detection system also includes:

[0143] The reference structural complexity determination module 400 is used to analyze the relationship curve, identify specific turning points and extract the internal structural complexity corresponding to the specific turning points, and define it as the reference structural complexity.

[0144] Specifically, Fig. 9 It shows a structural block diagram of the reference structure complexity determination module 400 in the system provided by an embodiment of the present invention.

[0145] Among them, in the preferred implementation manner provided by the present invention, the reference structure complexity determination module 400 specifically includes:

[0146] The specific turning point determination unit 401 is used to observe the overall trend of the relationship curve, identify the position where the trend transitions from a flat state to a significant downward trend, and determine the starting inflection point of the position as the specific turning point;

[0147] The reference structural complexity determination unit 402 is used to extract the internal structural complexity corresponding to a specific turning point and define it as the reference structural complexity.

[0148] In an embodiment of the present invention, to achieve "observing the overall trend of the relationship curve, identifying the position of transition from a gentle state to a clear downward trend, and determining the starting inflection point of the position as a specific turning point", it can be accomplished by analyzing the data characteristics and change rules of the curve. First, the relationship curve between the internal structure complexity and the edge clarity difference value is mathematically modeled, and a fitting algorithm is usually used to generate a continuous curve (such as polynomial fitting). Next, the first-order derivative of the curve is used to calculate the rate of change, or the position of the morphological change of the curve is determined by curvature analysis to identify the turning point from a gentle state to a clear downward trend. Specifically, the derivative value of the curve at the turning point will change significantly, such as gradually changing from close to zero to a negative value, or the curvature will have a maximum value, which can be used as a basis for judging the turning point.

[0149] Determining the "specific turning point" is of great significance because it reflects the critical point of the compensation optimization effect on image clarity improvement. In other words, the turning point indicates that the internal structure complexity of the object has reached a certain level, at which point the effect of delay compensation begins to significantly weaken, and even cannot compensate for the image quality degradation caused by aging. This critical information helps define the optimization capability range of the equipment and provides a basis for dynamically adjusting the exposure delay parameters.

[0150] The internal structure complexity corresponding to the "specific turning point" is defined as the "benchmark structure complexity", which can provide a standard reference value for subsequent testing. The benchmark structure complexity represents the upper limit of complexity that the original exposure parameters can effectively adapt to. When the complexity of the subsequent objects to be tested exceeds this value, the delayed exposure parameters can be adjusted according to the degree of deviation to ensure that the imaging quality of complex objects remains within an acceptable range, thereby optimizing the reliability and accuracy of the test results.

[0151] Furthermore, the artificial intelligence-based port electromechanical equipment detection system also includes:

[0152] The exposure delay parameter adjustment module 500 is used to calculate the internal structure complexity of the object to be tested entering the target X-ray imaging machine in real time during subsequent detection, and compare it with the reference structure complexity. If the reference structure complexity is exceeded, the preset exposure delay parameters are dynamically adjusted according to the degree of deviation to generate an optimized imaging image.

[0153] In the embodiment of the present invention, after determining the previous reference structure complexity value, the subsequent real-time detection of the object to be tested is entered, and the entire process revolves around real-time calculation, comparison of reference values, and dynamic adjustment of parameters. The following is a detailed example description:

[0154] Assume that in the previous analysis, the baseline structural complexity has been determined to be 50, the preset exposure delay parameter is 10 milliseconds, and the adjustment factor k is set to 1.2. Now a new object to be tested enters the X-ray imaging machine.

[0155] The first step is to collect the original image of the object to be tested, and use the image analysis algorithm to extract its edge density and texture distribution and other features. Through comprehensive analysis, it is concluded that the internal structure complexity of the object is 70. Then, the system compares this complexity value with the benchmark structure complexity of 50 and calculates the degree of deviation of the complexity. The degree of deviation is calculated by first finding the difference between the complexity of the object to be tested and the benchmark complexity, that is, 70 minus 50, the result is 20; then divide this difference by the benchmark complexity of 50, and get a deviation ratio of 0.4. This shows that the complexity of the object is 40% higher than the benchmark value.

[0156] In the second step, the system calculates the delay ratio that needs to be adjusted based on this deviation ratio and the adjustment coefficient k=1.2. Specifically, the system adds 1 to the deviation ratio 0.4 to get an adjustment base of 1.4. Then, this adjustment base is multiplied by the adjustment coefficient k to get a final adjustment factor of 1.68. This means that the original preset exposure delay parameters need to be increased by 68% to accommodate the higher complexity of the object to be tested.

[0157] In the third step, the system multiplies the calculated adjustment factor 1.68 by the preset exposure delay parameter 10 milliseconds, and the optimized exposure delay parameter is 16.8 milliseconds. This value represents the compensated exposure time, which is used to ensure that complex objects can obtain higher quality imaging effects.

[0158] The fourth step is to re-image the object under test using the optimized exposure delay parameter of 16.8 milliseconds. The re-captured image shows that the edge clarity of the object is significantly improved, and the internal structure details are clearer, meeting the imaging quality requirements. The system then records this optimization result for further adjustment reference in subsequent inspections.

[0159] The specific solution introduced in the present invention is based on the connection between "internal structure complexity" and "compensation for exposure delay" to optimize the problem of reduced imaging quality of aging equipment. Similar ideas can also be applied to other directions. For example, in medical imaging equipment, imaging parameters can be optimized based on the connection between tissue density and X-ray exposure time to improve the detail clarity of diagnostic images. It can also be applied to industrial detection to adjust the detection signal strength based on the relationship between material thickness and ultrasonic signal attenuation to enhance the detection effect of complex structures. The above method fully reflects the flexible application of the connection between parameter adjustment and target characteristics.

[0160] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0161] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0162] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0163] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0164] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for detecting electromechanical equipment at a port based on artificial intelligence, characterized in that: The method comprises: Acquire original images and compensated images taken by the target X-ray imaging machine of a preset number of objects to be tested, wherein the original images are acquired based on original exposure parameters, and the compensated images are acquired based on preset exposure delay parameters; Processing the original image of each object to be tested, calculating its internal structure complexity, arranging a preset number of objects to be tested according to the size of the internal structure complexity, and generating a sorting sequence of the internal structure complexity; Compare and analyze the original image and the compensated image of each object in the sorting sequence, calculate the difference in edge clarity between the two, match it with the internal structure complexity, and draw a relationship curve between the internal structure complexity and the edge clarity difference; Analyze the relationship curve, identify specific turning points and extract the internal structure complexity corresponding to the specific turning points, which is defined as the benchmark structure complexity; In subsequent inspections, the internal structure complexity of the object to be tested entering the target X-ray imaging machine is calculated in real time and compared with the baseline structure complexity. If it exceeds the baseline structure complexity, the preset exposure delay parameters are dynamically adjusted according to the degree of deviation to generate an optimized imaging image.

2. The artificial intelligence-based port electromechanical equipment detection method according to claim 1 is characterized in that: The steps of processing the original image of each object to be tested, calculating the internal structure complexity thereof, arranging a preset number of objects to be tested according to the size of the internal structure complexity, and generating a sorting sequence of the internal structure complexity include: Processing the original image of each object under test, using an image analysis algorithm to extract feature data reflecting the complexity of the object structure, the feature data including edge density, texture distribution or other image indicators related to the object structure; Based on the extracted feature data, these features are converted into a numerical internal structure complexity index through weighted analysis of comprehensive features, where the weighted analysis comprehensively evaluates the contribution of the complexity according to the importance of the features; The complexity indicators of all tested objects are compared and arranged from small to large according to the complexity values ​​to generate a sequence sorted by complexity.

3. The method for detecting port electromechanical equipment based on artificial intelligence according to claim 1 is characterized in that: The steps of sequentially comparing and analyzing the original image and the compensated image of each object under test in the sorting sequence, calculating the edge definition difference value between the two, matching the difference with the internal structure complexity, and drawing a relationship curve between the internal structure complexity and the edge definition difference include: Determine the edge clarity of the original image and the compensated image of each object under test based on edge detection algorithm and image processing technology, and subtract the edge clarity of the original image from the edge clarity of the compensated image to obtain an edge clarity difference value; According to the internal structure complexity of each tested object and the calculated edge definition difference value, a corresponding relationship is established between the two to form a data pair with the internal structure complexity as the independent variable and the edge definition difference value as the dependent variable; All matching data pairs are plotted into scatter plots in order of internal structure complexity from small to large, and based on the scatter plots, a relationship curve between internal structure complexity and edge clarity difference is generated through a fitting algorithm.

4. The artificial intelligence-based port electromechanical equipment detection method according to claim 1 is characterized in that: The steps of analyzing the relationship curve, identifying a specific turning point, extracting the internal structure complexity corresponding to the specific turning point, and defining it as the baseline structure complexity include: Observe the overall trend of the relationship curve, identify the transition point from a flat state to a clear downward trend, and determine the starting inflection point of this position as a specific turning point; The internal structural complexity corresponding to a specific turning point is extracted and defined as the benchmark structural complexity.

5. The artificial intelligence-based port electromechanical equipment detection method according to claim 1 is characterized in that: In subsequent testing, the internal structure complexity of the object to be tested entering the target X-ray imaging machine is calculated in real time and compared with the reference structure complexity. If the reference structure complexity is exceeded, the preset exposure delay parameters are dynamically adjusted according to the degree of deviation to generate an optimized imaging image. The steps include: When the object to be tested enters the target X-ray imaging machine, its original image is collected and its internal structure complexity is determined based on the image analysis algorithm; Compare the internal structure complexity of the object to be tested with the reference structure complexity. If the internal structure complexity is greater than the reference structure complexity, call the exposure delay parameter optimization formula to calculate the optimized exposure delay parameter. The image of the object to be tested is recaptured using the optimized exposure delay parameters to generate an imaging image with optimized quality.

6. The artificial intelligence-based port electromechanical equipment detection method according to claim 5 is characterized in that: The exposure delay parameter optimization formula is: Among them, T opt is the optimized exposure delay parameter, T preset is the preset exposure delay parameter, C real is the internal structure complexity of the object to be tested, C base is the baseline structural complexity, is the deviation degree of the internal structure complexity of the object to be tested compared with the reference structure complexity, and K is the adjustment coefficient.

7. A port electromechanical equipment detection system based on artificial intelligence, characterized in that: The system comprises: an image acquisition module, a sequence generation module, a relationship curve generation module, a reference structure complexity determination module and an exposure delay parameter adjustment module, wherein: An image acquisition module, used to acquire original images and compensated images taken by a target X-ray imaging machine of a preset number of objects under test, wherein the original images are acquired based on original exposure parameters, and the compensated images are acquired based on preset exposure delay parameters; A sequence generation module is used to process the original image of each object to be tested, calculate its internal structure complexity, arrange a preset number of objects to be tested according to the size of the internal structure complexity, and generate a sorting sequence of the internal structure complexity; A relationship curve generation module is used to compare and analyze the original image and the compensated image of each tested object in the sorting sequence in turn, calculate the edge clarity difference value between the two, match it with the internal structure complexity, and draw a relationship curve between the internal structure complexity and the edge clarity difference; A benchmark structural complexity determination module is used to analyze the relationship curve, identify specific turning points and extract the internal structural complexity corresponding to the specific turning points, which is defined as the benchmark structural complexity; The exposure delay parameter adjustment module is used to calculate the internal structure complexity of the object to be tested entering the target X-ray imaging machine in real time during subsequent inspections, and compare it with the baseline structure complexity. If the baseline structure complexity is exceeded, the preset exposure delay parameters are dynamically adjusted according to the degree of deviation to generate an optimized imaging image.

8. The artificial intelligence-based port electromechanical equipment detection system according to claim 7 is characterized in that: The sequence generation module specifically includes: An image processing unit is used to process the original image of each object under test and extract feature data reflecting the complexity of the object structure by using an image analysis algorithm. The feature data includes edge density, texture distribution or other image indicators related to the object structure. A feature data processing unit, for converting the extracted feature data into a numerical internal structure complexity index through weighted analysis of the comprehensive features, wherein the weighted analysis comprehensively evaluates the contribution of the feature to the complexity according to its importance; The sequence generation unit is used to compare the complexity indicators of all the tested objects and arrange them from small to large according to the complexity values ​​to generate a sequence sorted by complexity.

9. The artificial intelligence-based port electromechanical equipment detection system according to claim 8 is characterized in that: The relationship curve generation module specifically includes: An edge definition difference value calculation unit is used to determine the edge definition of the original image and the compensated image of each object under test based on an edge detection algorithm and an image processing technology, and to obtain an edge definition difference value by subtracting the edge definition of the original image from the edge definition of the compensated image; A data pair generating unit is used to establish a corresponding relationship between the internal structure complexity of each tested object and the calculated edge definition difference value, so as to form a data pair with the internal structure complexity as the independent variable and the edge definition difference value as the dependent variable; The relationship curve generating unit is used to draw all matching data pairs into a scatter plot in the order of internal structure complexity from small to large, and based on the scatter plot, generate a relationship curve between the internal structure complexity and the edge clarity difference through a fitting algorithm.

10. The artificial intelligence-based port electromechanical equipment detection system according to claim 9 is characterized in that: The benchmark structure complexity determination module specifically includes: A specific turning point determination unit is used to observe the overall trend of the relationship curve, identify the position where the trend transitions from a flat state to an obvious downward trend, and determine the starting inflection point of the position as a specific turning point; The benchmark structural complexity determination unit is used to extract the internal structural complexity corresponding to a specific turning point and define it as the benchmark structural complexity.

Images