A Machine Vision-Based Method and System for Testing the Fire Resistance Performance of Cables

Abstract

This invention provides a machine vision-based method and system for testing the fire resistance performance of cables, relating to the field of cable performance testing technology. The method includes: acquiring a time-series image sequence during the cable combustion process; performing spatiotemporal convolution filtering on each image in the time-series image sequence to obtain a spatiotemporal gradient vector describing the grayscale change rate of each pixel in the image; determining candidate flame fronts based on the spatiotemporal gradient vectors; determining the evolution velocity constraints of the candidate flame fronts; updating the candidate flame fronts based on the evolution velocity constraints to obtain the target flame front; projecting it onto the axial centerline of the cable and calculating the flame spread velocity; outputting the cable as qualified if the flame spread velocity is less than a preset flame spread velocity, otherwise outputting the cable as unqualified. By fusing spatiotemporal multidimensional features and constraints, the stability of flame front tracking is significantly improved, especially adapting to complex combustion scenarios with irregular flame shapes, achieving objective quantification and automated evaluation of flame spread velocity.

Description

Technical Field

[0001] This invention relates to the field of cable performance testing technology, and in particular to a method and system for testing the fire resistance performance of cables based on machine vision. Background Technology

[0002] Cables are conductors used to transmit electricity or signals, typically consisting of metal wires, insulation materials, and a protective layer. Cables are widely used in power, communications, construction, and many other fields. Machine vision is a technology that uses computer and image processing techniques to enable machines to understand images or videos and perform automatic analysis, judgment, and decision-making based on visual information. Its goal is to achieve object detection by simulating human visual perception and understanding.

[0003] Fire resistance testing of cables is crucial for ensuring the safety of power systems and buildings. The performance of cables during a fire directly impacts the speed of fire spread and the time available for evacuation. Fire resistance testing assesses the stability and resilience of cables in high-temperature environments, ensuring they remain flame-retardant and slow the spread of fire even in extreme conditions.

[0004] However, existing technologies for testing the fire resistance performance of cables typically rely on manual observation or edge detection algorithms to track the flame front, which suffers from poor real-time performance and low flame front tracking accuracy. They are also difficult to adapt to the complexity of dynamic combustion processes and have weak adaptability to irregular flame patterns, resulting in highly subjective test results with large deviations. Summary of the Invention

[0005] In view of the shortcomings of the prior art, the purpose of this invention is to provide a machine vision-based method for testing the fire resistance performance of cables. This method can solve the problems of poor real-time performance and low accuracy of flame front tracking, which are common in the prior art for testing the fire resistance performance of cables. These problems include reliance on manual observation or edge detection algorithms to track the flame front, difficulty in adapting to the complexity of dynamic combustion processes, and weak adaptability to irregular flame shapes, resulting in highly subjective test results and large deviations.

[0006] A first aspect of this invention provides a machine vision-based method for testing the fire resistance performance of cables, comprising:

[0007] S1: Acquire a time-series image sequence during the cable combustion process;

[0008] S2: Perform spatiotemporal convolution filtering on each image in the time-series image sequence to obtain the spatiotemporal gradient vector describing the gray-level change rate of each pixel in the image;

[0009] S3: Determine the candidate flame front based on the spatiotemporal gradient vector;

[0010] S4: Determine the evolution rate constraint of the candidate flame front based on the spatiotemporal gradient vector;

[0011] S5: Update the candidate flame front based on the evolution rate constraint to obtain the target flame front;

[0012] S6: Project the target flame front onto the axial centerline of the cable and calculate the flame spread rate;

[0013] S7: If the flame spread rate is less than the preset flame spread rate, the output cable is qualified; otherwise, the output cable is unqualified.

[0014] A second aspect of this invention provides a machine vision-based cable fire resistance performance testing system, comprising: a processor and a memory;

[0015] The memory stores programs or instructions that can run on the processor, which, when executed by the processor, implement the steps of the machine vision-based cable fire resistance testing method as described in the first aspect.

[0016] A third aspect of the present invention provides a readable storage medium on which a program or instructions are stored, which, when executed by a processor, implement the steps of the machine vision-based cable fire resistance testing method described in the first aspect.

[0017] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0018] In this embodiment of the invention, the pixel grayscale change rate is extracted by spatiotemporal convolution filtering, and a spatiotemporal gradient vector is constructed to accurately capture the dynamic features of the flame. Combined with the evolution speed constraint model, adaptive tracking of the irregular flame front is achieved. In addition, the flame spread speed quantification model based on axial projection calculation effectively solves the problems of strong subjectivity of manual observation and high misjudgment rate of edge detection algorithm, and significantly improves the accuracy and reliability of fire resistance performance test results. Attached Figure Description

[0019] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts. Obviously, the drawings described below are merely some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0020] Figure 1 This is a flowchart illustrating a method for testing the fire resistance performance of cables based on machine vision, provided in an embodiment of the present invention.

[0021] Figure 2This is a schematic diagram of a cable fire resistance performance testing system based on machine vision provided in an embodiment of the present invention. Detailed Implementation

[0022] To enable those skilled in the art to better understand the technical solutions in the embodiments of the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. It should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0023] The following description, in conjunction with the accompanying drawings, details the machine vision-based cable fire resistance testing method provided by the present invention through specific embodiments and application scenarios.

[0024] Reference manual attached Figure 1 The diagram shows a flowchart of a cable fire resistance performance testing method based on machine vision provided by an embodiment of the present invention.

[0025] This invention provides a machine vision-based method for testing the fire resistance performance of cables, which may include the following steps:

[0026] S1: Acquire a time-series image sequence during the cable combustion process.

[0027] The time-series image sequence refers to multiple frames of images continuously acquired in chronological order during the cable combustion experiment, used to record the dynamic spread of the flame on the cable surface. These images need to cover the entire combustion cycle and include detailed changes in flame morphology (such as color, brightness, and outline).

[0028] In one possible implementation, the cable combustion process is specifically an axial cable combustion process. S1 specifically refers to:

[0029] A time-series image sequence was obtained by vertically capturing the axial combustion process of the cable using an industrial camera.

[0030] The axial combustion process of cables refers to the combustion phenomenon in which the flame spreads vertically along the length (axial direction) of the cable during the fire resistance test. This combustion process involves suspending the cable and igniting it, causing the flame to spread axially. An industrial camera, perpendicular to the cable, captures changes in the position of the flame front, providing crucial data for subsequent calculations of the flame spread rate and ensuring that the test results are consistent with actual fire scenarios.

[0031] S2: Perform spatiotemporal convolution filtering on each image in the time-series image sequence to obtain the spatiotemporal gradient vector describing the rate of change of gray level of each pixel in the image.

[0032] Spatiotemporal convolutional filtering is a convolutional operation that combines the temporal dimension (inter-frame variations) and the spatial dimension (inter-pixel grayscale differences). The spatiotemporal gradient vector is the gradient vector of each pixel in the temporal-spatial domain, composed of the temporal gradient (the rate of change of grayscale over time) and the spatial gradient (the direction of change of grayscale in the image space). This vector can quantify the dynamic characteristics of the flame front. By fusing temporal and spatial information, spatiotemporal convolutional filtering can simultaneously capture the dynamic rate of change and spatial distribution characteristics of the flame, significantly improving the accuracy and robustness of flame front detection, and is particularly suitable for the real-time tracking needs of irregular flame shapes in complex combustion scenarios.

[0033] In one possible implementation, the spatiotemporal gradient vector includes both spatial and temporal gradients.

[0034] In one possible implementation, S2 specifically includes:

[0035] S201: Calculate the spatial gradient using the Sobel operator, where the spatial gradient includes the horizontal gradient and the vertical gradient.

[0036] The horizontal gradient is the axial gradient of the cable, and the vertical gradient is the radial gradient of the cable.

[0037] The formulas for calculating the horizontal and vertical gradients are as follows:

[0038] , .

[0039] in, This represents the horizontal gradient of image I. This represents the gradient in the vertical direction of image I. This represents the convolution operation.

[0040] S202: The temporal gradient is calculated using the adjacent frame difference method.

[0041] The formula for calculating the time gradient is as follows:

[0042] .

[0043] in, and These represent the pixel intensities of the same pixel in the images at time t+1 and time t-1 in the time-series image sequence, respectively. This indicates the acquisition interval between adjacent image frames in a time-series image sequence. This represents the temporal gradient of image I.

[0044] S203: Combine spatial gradient and temporal gradient at each pixel to obtain a spatiotemporal gradient vector.

[0045] The specific formula for the spatiotemporal gradient vector is as follows:

[0046] .

[0047] in, This represents the spatiotemporal gradient vector, and the subscript T indicates transpose.

[0048] Specifically, this process constructs a spatiotemporal gradient vector by combining spatial and temporal gradients, achieving a multidimensional representation of the dynamic characteristics of flames. The spatial gradient utilizes the Sobel operator to capture the edge contours (axial and radial variations) of the flame in the image, while the temporal gradient reflects the dynamic evolution rate of the flame through the difference between adjacent frames. The fusion of these two gradients forms a spatiotemporal gradient vector, which can accurately describe the grayscale variation patterns of the flame in both spatial and temporal dimensions. By simultaneously considering the spatial morphology and temporal evolution characteristics of the flame, this method significantly improves the accuracy of flame front detection, particularly adapting to the real-time tracking needs of irregular flames in complex combustion scenarios. It effectively overcomes the dependence of traditional methods on static edges or single-dimensional features, thereby improving the objectivity and accuracy of the test results.

[0049] S3: Determine the candidate flame front based on the spatiotemporal gradient vector.

[0050] Among them, the candidate flame front refers to the pixel region or contour that may constitute the flame boundary, which is initially screened by the spatiotemporal gradient vector.

[0051] In one possible implementation, S3 specifically includes:

[0052] S301: Calculate the magnitude of the spatiotemporal gradient vector.

[0053] S302: Output each pixel point corresponding to the maximum spatiotemporal gradient vector magnitude in the axial direction of the cable as a candidate pixel group for the flame front, wherein each candidate pixel group for the flame front constitutes a candidate flame front.

[0054] It should be noted that this process calculates the magnitude of the spatiotemporal gradient vector and selects the pixels with the most dramatic axial grayscale changes from each row of pixels as candidate flame fronts, accurately capturing high-gradient regions of dynamic flame spread. By combining spatial morphology and temporal evolution characteristics, it effectively suppresses background noise interference and significantly improves the accuracy and robustness of flame front identification in complex combustion scenarios.

[0055] S4: Determine the evolution rate constraint of the candidate flame front based on the spatiotemporal gradient vector.

[0056] The evolution rate constraint is a dynamic evolution rate threshold set for candidate flame fronts based on the gradient components in the spatiotemporal gradient vector and the combustion pattern. By combining the physical characteristics of flame combustion with spatiotemporal gradient information, the evolution rate constraint dynamically limits the evolution range of candidate flame fronts, effectively suppressing noise interference and transient anomalies, improving the stability and rationality of flame front tracking, and significantly enhancing the algorithm's adaptability to irregular flame morphologies in complex combustion scenarios.

[0057] In one possible implementation, S4 specifically includes:

[0058] S401: Defines the level set function for the candidate flame front.

[0059] S402: Calculate the local curvature of the candidate flame front using the level set function.

[0060] S403: Establish evolution velocity constraints by combining local curvature.

[0061] It should be noted that this process describes the geometry of the candidate flame front using a level set function. By combining local curvature constraints with the dynamic characteristics of the spatiotemporal gradient vector, a comprehensive evolution velocity constraint is established, ensuring that the evolution of the flame front conforms to both geometric smoothness and the laws of physical combustion. Through multi-dimensional feature fusion, noise interference is effectively suppressed, enhancing the adaptability to irregular flame shapes and significantly improving the stability and physical plausibility of flame front tracking.

[0062] The specific formula for the evolution rate constraint is as follows:

[0063] .

[0064] = .

[0065] .

[0066] .

[0067] in, This indicates taking the minimum function value. As the target evolution speed , Describes the level set function. Indicates local curvature. Represents the spatial divergence operator. Represents the gradient of the level set function space. This represents the magnitude of the spatial gradient of the level set function. Represents the spatiotemporal gradient vector. Represents the candidate flame front normal vector. Denotes the square of the L2 norm. This represents the normal diffusion velocity to be solved, which is the diffusion velocity of the candidate flame front along the n direction. This represents the weighting coefficient.

[0068] It should be noted that this process achieves precise tracking of the flame front by optimizing its evolution rate and combining spatiotemporal gradient vectors and level set functions. By introducing constraints on the normal vector and local curvature, the evolution rate of the flame front is effectively controlled, avoiding errors under irregular morphologies and enhancing adaptability to complex combustion processes. This method improves tracking accuracy and robustness, ensuring the accuracy and consistency of test results.

[0069] Optionally, the weighting factor can be set to 0.3.

[0070] S5: Update the candidate flame front based on the evolution rate constraint to obtain the target flame front.

[0071] The target flame front is the optimized result of the candidate flame front under the constraint of evolution rate. By combining the dynamic characteristics of the spatiotemporal gradient vector and the combustion law, the candidate front is iteratively corrected, and finally a continuous curve that highly matches the actual flame boundary is formed. This front can accurately reflect the real-time spread path of the flame on the cable surface and is the accurate data for subsequent flame spread rate calculation.

[0072] In one possible implementation, S5 specifically includes:

[0073] S501: The gradient descent method is used to iteratively solve the evolution rate constraint to obtain the target evolution rate.

[0074] Specifically, the process involves the following steps: First, setting an initial value for the evolution rate and the learning rate. Then, calculating the gradient of the current evolution rate based on the objective function. Next, updating the evolution rate along the negative gradient direction. Repeating gradient calculation and parameter updates until the gradient approaches zero or the preset number of iterations is reached. The final converged evolution rate is the target evolution rate that minimizes the objective function.

[0075] S502: Update the level set function according to the target evolution rate.

[0076] The updated formula is as follows:

[0077] .

[0078] in, Level set function The partial derivative with respect to time t.

[0079] S503: Extract the zero contour lines of the updated level set function to obtain the target flame front.

[0080] The zero contour line represents the frontal line of the target flame.

[0081] It should be noted that this process uses gradient descent to iteratively optimize the evolution rate constraint, driving the dynamic evolution of the level set function, and finally extracting the zero contour line as the target flame front. Combined with implicit curve evolution and constraints, this ensures the flame front maintains geometric smoothness and dynamic continuity in complex combustion scenarios, effectively adapting to irregular morphological changes and significantly improving tracking accuracy.

[0082] S6: Project the front edge of the target flame onto the axial centerline of the cable and calculate the flame spread rate.

[0083] The cable axial centerline refers to the geometric centerline along the cable's length (i.e., the direction of combustion spread). Flame spread rate is the distance the flame travels along the cable axial centerline per unit time. By projecting the target flame front onto the cable axial centerline, the flame spread path along the combustion direction can be accurately extracted, avoiding misjudgments of irregular flame shapes by traditional manual observation or edge detection algorithms. This achieves objective quantification of flame spread rate, significantly improving the automation level and data accuracy of fire resistance testing.

[0084] In one possible implementation, S6 specifically includes:

[0085] S601: Collect the maximum pixel movement distance among all pixels corresponding to the front edge of the target flame within a preset time period.

[0086] It should be noted that those skilled in the art can set the preset duration according to actual needs, and this invention does not limit this.

[0087] S602: Project the maximum pixel movement distance onto the axial direction of the cable and calculate the target flame front displacement per unit time, i.e., the flame spread rate.

[0088] Specifically, this process accurately calculates the actual flame propagation speed along the cable axis by collecting the maximum pixel movement distance of the target flame front within a preset time period and combining axial projection and tangent angle correction. Through pixel equivalent conversion and geometric projection compensation, interference from non-axial movement is effectively eliminated, significantly improving the accuracy and physical consistency of flame velocity measurement in complex combustion scenarios.

[0089] The specific formula for calculating the flame spread rate is as follows:

[0090] .

[0091] in, Indicates the instantaneous speed of flame spread. Represents pixel equivalent. Indicates the maximum pixel movement distance. The sampling time interval represents the maximum pixel movement distance. This represents the angle between the tangent at the pixel corresponding to the maximum pixel movement distance of the target flame front and the axial direction of the cable. express The cosine value.

[0092] Pixel equivalent is a scaling factor that converts pixel distances in an image into actual physical distances.

[0093] It should be noted that this process calculates the instantaneous flame spread velocity along the cable axis by combining the pixel movement distance of the flame front with the time interval. By introducing pixel equivalent conversion and tangent angle correction, the physical accuracy of the measurement results is ensured, and interference from non-axial movement is avoided. This method can provide high-precision flame spread velocity measurement in complex combustion scenarios, improving the reliability of the test.

[0094] S7: If the flame spread rate is less than the preset flame spread rate, the output cable is qualified; otherwise, the output cable is unqualified.

[0095] It should be noted that those skilled in the art can set the preset flame spread rate according to actual needs, and this invention does not limit it.

[0096] In practical applications, a time-series image sequence of the cable combustion process is first acquired using an industrial camera. Spatiotemporal convolution filtering is then used to extract dynamic flame features (such as grayscale change rate and spatial gradient) to construct a spatiotemporal gradient vector, accurately capturing the dynamic evolution of the flame front. Subsequently, an evolution velocity model is established by combining level set functions and local curvature constraints. Iterative optimization drives the dynamic update of the flame front, ensuring its geometric smoothness and physical rationality. Finally, the target flame front is projected onto the cable's axial centerline. Pixel equivalent conversion and geometric projection correction are used to calculate the flame spread rate, which is then compared with a preset threshold to determine cable qualification. This method, by fusing spatiotemporal multidimensional features and constraints, significantly improves the real-time performance and robustness of flame front tracking, particularly adapting to complex combustion scenarios with irregular flame morphologies. It effectively overcomes the subjectivity of traditional manual observation and the misjudgment problems of edge detection algorithms, achieving objective quantification of flame spread rate and high-precision automated evaluation of fire resistance performance.

[0097] In this embodiment of the invention, the pixel grayscale change rate is extracted by spatiotemporal convolution filtering, and a spatiotemporal gradient vector is constructed to accurately capture the dynamic features of the flame. Combined with the evolution velocity constraint model, adaptive tracking of irregular flame fronts is achieved. In addition, the flame spread velocity quantification model based on axial projection calculation effectively solves the problems of strong subjectivity in manual observation and high misjudgment rate of edge detection algorithms, significantly improving the accuracy and reliability of fire resistance performance test results.

[0098] Reference manual attached Figure 2 The diagram shows a schematic of a machine vision-based cable fire resistance testing system provided in an embodiment of the present invention.

[0099] This invention provides a machine vision-based cable fire resistance performance testing system 20, comprising: a processor 201 and a memory 202;

[0100] The memory 202 stores programs or instructions that can run on the processor 201. When the program or instructions are executed by the processor 201, they implement the steps of the above-described machine vision-based cable fire resistance performance testing method and achieve the same technical effect. To avoid repetition, the present invention will not elaborate further.

[0101] It should be understood that the processor 201 in this embodiment of the invention may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0102] It should also be understood that the memory 202 in the embodiments of the present invention can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus RAM (DR RAM).

[0103] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0104] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0105] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0106] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0107] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0108] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0109] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0110] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0111] This invention provides a readable storage medium comprising: storing a program or instructions on the readable storage medium, wherein when the program or instructions are executed by a processor, the program or instructions implement the steps of the above-described machine vision-based cable fire resistance performance testing method and achieve the same technical effect. To avoid repetition, this invention will not elaborate further.

[0112] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the embodiments of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the protection scope of the present invention.

Claims

1. A machine vision-based method of testing the fire resistance performance of an electrical cable, characterized in that, The method comprises: S1: collecting a time sequence image sequence in a cable combustion process; S2: performing space-time convolution filtering on each image in the time sequence image sequence to obtain a space-time gradient vector describing a gray scale change rate of each pixel in the image; S3: determining a candidate flame front according to the space-time gradient vector; S4: determining an evolution speed constraint of the candidate flame front based on the space-time gradient vector; S5: updating the candidate flame front based on the evolution speed constraint to obtain a target flame front; S6: projecting the target flame front to a cable axial center line to calculate a flame propagation speed; S7: in a case where the flame propagation speed is less than a preset flame propagation speed, outputting that the cable is qualified, otherwise, outputting that the cable is unqualified. The S3 specifically comprises: S301: calculating a space-time gradient vector module length; S302: taking each pixel point corresponding to a maximum space-time gradient vector module length in a cable axial direction as a flame front candidate pixel group, wherein each flame front candidate pixel group constitutes the candidate flame front.

2. The machine vision-based cable fire endurance performance testing method according to claim 1, wherein, The cable combustion process is specifically a cable axial combustion process; and the S1 is specifically: vertically photographing the cable axial combustion process by an industrial camera to obtain the time sequence image sequence.

3. The machine vision-based cable fire endurance performance testing method according to claim 1, wherein, The space-time gradient vector comprises a space gradient and a time gradient.

4. The machine vision-based cable fire endurance performance testing method according to claim 3, wherein, The S2 specifically comprises: S201: calculating the space gradient by a Sobel operator, wherein the space gradient comprises a horizontal direction gradient and a vertical direction gradient; S202: calculating the time gradient by an adjacent frame difference method; S203: combining the space gradient and the time gradient by a pixel point to obtain the space-time gradient vector.

5. The machine vision-based cable fire endurance performance testing method according to claim 1, wherein, The S4 specifically comprises: S401: defining a level set function of the candidate flame front; S402: calculating a local curvature of the candidate flame front by the level set function; S403: combining the local curvature to establish the evolution speed constraint.

6. The machine vision-based cable fire endurance performance testing method according to claim 5, wherein, The S5 specifically comprises: S501: iteratively solving the evolution speed constraint by a gradient descent method to obtain a target evolution speed; S502: updating the level set function according to the target evolution speed; S503: extracting a zero contour line of the updated level set function to obtain the target flame front.

7. The machine vision-based cable fire endurance performance testing method according to claim 1, wherein, The S6 specifically comprises: S601: collecting a maximum pixel movement distance in all pixel points corresponding to the target flame front within a preset time length; S602: projecting the maximum pixel movement distance to a cable axial direction to calculate a target flame front displacement in a unit time, that is, the flame propagation speed.

8. A machine vision-based cable fire endurance performance testing system, characterized by, The method comprises: a processor and a memory; the memory stores a program or instructions executable on the processor, and the program or instructions are executed by the processor to implement the steps of the machine vision-based cable fire resistance performance test method according to any one of claims 1 to 7.

9. A readable storage medium, characterized by, The program or instructions are stored on the readable storage medium, and the program or instructions are executed by the processor to implement the steps of the machine vision-based cable fire resistance performance test method according to any one of claims 1 to 7.

Images