Civil defense door performance detection system based on image analysis
Through the human defense door performance detection system based on image analysis, combined with static test and pressure difference method, multi-parameter detection of the human defense door is realized, solving the problem of low detection efficiency in the prior art, and improving detection efficiency and convenience.
Patent Information
- Application Number
- CN202411252024.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-09
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-09-09
AI Technical Summary
The existing civil defense door performance detection system has low detection efficiency and cannot uniformly detect performance parameters such as airtightness and strength, resulting in low overall performance detection efficiency.
Using a human defense door performance detection system based on image analysis, the data management subsystem and detection and analysis subsystem, including image acquisition module, image processing module, feature extraction module and feature comparison module, realize multi-parameter detection of the human defense door, combined with Static test and pressure difference method, automatic performance detection.
It realizes unified detection of multiple parameters such as airtightness and strength of the civil defense door, reduces the complexity of the detection process, improves the detection efficiency, simplifies the data acquisition and processing process, reduces the probability of damage, and improves the detection convenience.
Smart Images

Figure CN118762025B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of civil air defense door detection, and relates to an image analysis technology for civil air defense doors, and specifically is a civil air defense door performance detection system based on image analysis. Background Art
[0002] Civil air defense doors are doors used for entrances and exits of various security protection projects. They are clearly classified into ordinary single-leaf and double-leaf protective closed doors, single-leaf and double-leaf protective closed doors with movable thresholds, etc. Each type of closed door has different uses and different operating methods, and its specific performance parameter requirements are slightly different. Once the performance parameters do not meet the standards, its function may not be realized. Therefore, the performance testing of civil air defense doors is very strict and the testing procedures are also very numerous.
[0003] The performance parameters of civil air defense doors mainly include sealing performance, air tightness, strength, etc., and performance testing is mainly carried out on these parameters. At present, the existing civil air defense door performance testing system can only test each performance parameter of the civil air defense door separately through static tests, pressure difference methods, etc., with many processes and low efficiency. In the performance testing system of civil air defense doors, image detection technology can only realize surface quality analysis, and cannot uniformly detect and analyze the performance parameters of civil air defense doors such as air tightness and strength, resulting in low overall performance testing efficiency of civil air defense doors. Summary of the invention
[0004] The purpose of the present invention is to provide a civil air defense door performance detection system based on image analysis, so as to solve the problem of low detection efficiency of the civil air defense door performance detection system in the prior art.
[0005] The technical solution of the present invention is as follows:
[0006] The civil air defense door performance detection system based on image analysis includes a data management subsystem, a detection and analysis subsystem, and a database;
[0007] A number of personal air defense doors are randomly selected from the finished products of each production batch as test objects, and the data management subsystem is used to perform statistical management on the test data of the test objects; the data management subsystem includes an image acquisition module, an image processing module and a performance testing module;
[0008] Mark all civil air defense doors except the test object in each production batch of civil air defense door products as detection objects; the detection and analysis subsystem is used to perform performance detection and analysis on the detection objects; the detection and analysis subsystem includes a feature extraction module and a feature comparison module;
[0009] The image acquisition module is used to capture images of each surface of the test object, mark the front image and the back image of the test object as front view images, and mark all side images of the test object as side view images;
[0010] The image processing module is used to process the front view image and the side view image of the test object and obtain the front view parameters and the side view parameters;
[0011] The performance test module is used to perform performance test analysis on the test object and obtain the strength test performance value, the sealing coverage interval and the sealing gray deviation interval;
[0012] The feature extraction module is used to capture images of the detection object and extract image features;
[0013] The feature comparison module is used to perform performance detection on the detection object according to the image features of the detection object.
[0014] Furthermore, the process of acquiring the front parameters of the test object includes: enlarging the front view image of the test object into a pixel grid image and performing a grayscale transformation, then retrieving the front view image of the standard performance civil air defense door and marking it as a calibrated image, and overlapping and comparing the front view image of the test object with the calibrated image: marking the absolute value of the difference between the grayscale value of the pixel grid at the same position in the front view image and the grayscale value in the calibrated image as the deviation value of the pixel grid, and marking the pixel grid as a conforming grid or a deviation grid according to the deviation value; then all the deviation grids that touch each other in the front view image constitute a deviation area, and marking the number of deviation grids in the deviation area and the average value of the grayscale values of the deviation grids as the coverage value and gray deviation value of the deviation area respectively; the coverage values and gray deviation values of all deviation areas constitute the front parameters of the test object.
[0015] Furthermore, the specific process of marking a pixel grid as a conforming grid or a deviation grid includes: retrieving the deviation threshold through the database, and comparing the deviation value with the deviation threshold: if the deviation value is less than the deviation threshold, the corresponding pixel grid is marked as a conforming grid; if the deviation value is greater than or equal to the deviation threshold, the corresponding pixel grid is marked as a deviation grid.
[0016] Furthermore, the process of obtaining the side parameters includes: using a binary method to perform edge segmentation on the side view image of the test object to obtain edge parameters of the civil air defense door in the side view image, the edge parameters include length value, width value and thickness value, and retrieving the length standard value, width standard value and thickness standard value of the standard performance civil air defense door through the database; obtaining the length error value CW, width error value KW and thickness error value HW of the test object and performing numerical calculation to obtain the error coefficient WC of the side view image; the side parameters of the side view image are formed by the error coefficient WC of the side view image;
[0017] The calculation formula of the error coefficient WC of the side view image is: WC=k1×CW+k2×KW+k3×HW, where k1, k2 and k3 are all proportional coefficients, and k1>k2>k3>1.
[0018] Furthermore, the process of obtaining the length error value CW, the width error value KW and the thickness error value HW includes: marking the absolute value of the difference between the length value in the edge parameter of the test object and the length standard value as the length error value CW, marking the absolute value of the difference between the width value in the edge parameter of the test object and the width standard value as the width error value KW; marking the absolute value of the difference between the thickness value in the edge parameter of the test object and the thickness standard value as the thickness error value HW.
[0019] Furthermore, the process of obtaining the strength test performance value includes: testing the impact resistance and bearing capacity of the door leaf through a static test, marking the test object that passes the static test as a qualified strength object, and marking the test object that fails the static test as an unqualified strength object; then retrieving the error coefficient WC of the test objects of all unqualified strength objects and establishing an error set; performing variance calculation on all elements in the error set to obtain the precision coefficient of the error set, and comparing the precision coefficient with a preset precision threshold: if the precision coefficient is less than the precision threshold, marking the minimum element in the error set as the strength test performance value; if the precision coefficient is greater than or equal to the precision threshold, eliminating the maximum element and the minimum element in the error set, and then recalculating the precision coefficient of the error set, and again comparing the precision coefficient with the precision threshold, and so on, until the precision coefficient is less than the precision threshold. The precision threshold is a numerical constant directly retrieved from the database, and the specific value and value range of the precision threshold are set by the management personnel.
[0020] Furthermore, the process of obtaining the sealing coverage interval and the sealing gray deviation interval includes: using the pressure difference method to detect the sealing performance of the test object by applying a certain air pressure difference inside and outside the door, and marking the test object whose sealing performance meets the requirements as a sealed qualified object; marking the test object whose sealing performance does not meet the requirements as a sealed unqualified object, and the front parameters of all the sealed unqualified objects constitute a sealing set, and the sealing set includes a covering subset composed of the coverage values in the front parameters of all the sealed unqualified objects and a gray deviation subset composed of the gray deviation values in the front parameters of all the sealed unqualified objects; the maximum element and the minimum element in the covering subset constitute a covering range, and the covering range is divided into several covering intervals, and the number of elements in the covering subset whose covering values are within the covering interval is marked as the prominent value of the covering interval; the M1 covering intervals with the largest prominent value are marked as sealed covering intervals; the maximum element and the minimum element in the gray deviation subset constitute a gray deviation range, and the gray deviation range is divided into several gray deviation intervals, and the number of elements in the gray deviation subset whose gray deviation values are within the gray deviation interval is marked as the influence value of the gray deviation interval, and the M2 gray deviation intervals with the largest influence value are marked as sealed gray deviation intervals.
[0021] Furthermore, the image features of the detection object include: the image features include the coverage value and gray deviation value of the front view image corresponding to the detection object relative to the front view image of the standard performance civil air defense door; the image features also include the error coefficient WC of the side view image corresponding to the detection object relative to the side view image of the standard performance civil air defense door.
[0022] Furthermore, the specific process of performing performance testing on the test object includes:
[0023] Determine whether the coverage value and the gray deviation value are both in the sealing coverage interval and the sealing gray deviation interval: if so, determine that the sealing performance of the test object is unqualified, generate a sealing re-inspection signal and send the sealing re-inspection signal to the performance test module; if not, determine that the sealing performance of the test object is qualified;
[0024] Determine whether the error coefficient WC is greater than the strength test performance value: if so, the strength performance of the test object is determined to be unqualified, a strength re-test signal is generated and the strength re-test signal is sent to the performance test module; if not, the strength performance of the test object is determined to be qualified.
[0025] Compared with the prior art, the present invention has the following beneficial effects:
[0026] 1. The present invention uses image analysis technology to achieve unified detection of multiple parameters such as air tightness and strength of civil air defense doors, and the whole process is completed automatically by software. It not only solves the problem that each parameter needs to be detected separately for data collection and sorting, as well as separate detection procedures such as pressure difference method and static force test, but also greatly reduces the complexity of the overall detection process, reduces the detection workload, and improves the detection efficiency.
[0027] 2. The present invention can process the front view image and the side view image of the test object through the image processing module and obtain the front parameters and the side parameters. The front parameters are the features extracted from the front view image of the test object, and the side parameters are the features extracted from the side image of the test object. The front parameters and the side parameters provide data support for the performance test analysis process.
[0028] 3. The present invention can perform performance test analysis on the test object through the performance test module. After performing performance test on the test object in a traditional manner, the performance test result is compared with the front parameters and side parameters, and the rules of the test objects with unqualified performance in the front parameter set and the side parameter set are extracted to generate image detection basis, and performance detection is performed based on the image detection basis, and the multi-performance multi-process detection process is integrated into one image detection process, thereby effectively improving the overall detection efficiency of the civil defense door.
[0029] 4. The feature extraction module and the feature comparison module can be used to extract and compare the image features of the detection object, and the various performances of the detection object can be evaluated based on the comparison results. This breaks away from the technical limitations of traditional image detection that can only detect surface quality, simplifies the data collection and processing process during performance testing, and reduces the probability of damage to the civil air defense door during performance testing, thereby improving the convenience of performance testing of the civil air defense door. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0031] Figure 1 is the overall system block diagram of the present invention;
[0032] Figure 2 is a system block diagram of Embodiment 1 of the present invention;
[0033] Figure 3 This is a system block diagram of Embodiment 2 of the present invention. DETAILED DESCRIPTION
[0034] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0035] like Figure 1 As shown, the civil defense door performance detection system based on image analysis includes a data management subsystem, a detection and analysis subsystem and a database.
[0036] Embodiment 1: Figure 2 As shown, several personal air defense doors are randomly selected from each production batch of finished personal air defense doors as test objects, and the data management subsystem is used to perform statistical management on the test data of the test objects; the data management subsystem includes an image acquisition module, an image processing module and a performance testing module.
[0037] The image acquisition module is used to capture images of each surface of the test object, mark the front image and the back image of the test object as front view images, and mark all side images of the test object as side view images; and send the front view image and the side view image of the test object to the image processing module.
[0038] The image processing module is used to process the front view image and the side view image of the test object and obtain the front parameters and the side parameters: the front view image of the test object is enlarged into a pixel grid image and grayscale transformation is performed, then the front view image of the standard performance civil air defense door is retrieved and marked as the standard image, and the front view image of the test object is overlapped with the standard image: the absolute value of the difference between the grayscale value of the pixel grid at the same position in the front view image and the grayscale value in the standard image is marked as the deviation value of the pixel grid, the deviation threshold is retrieved through the database, and the deviation value is compared with the deviation threshold: if the deviation value is less than the deviation threshold, the corresponding pixel grid is marked as a conforming grid; if the deviation value is greater than or equal to the deviation threshold, the corresponding pixel grid is marked as a deviation grid; then all the deviation grids that contact each other in the front view image constitute a deviation area, and the number of deviation grids in the deviation area and the average value of the deviation grid grayscale value are marked as the coverage value and gray deviation value of the deviation area respectively; the coverage value and gray deviation value of all deviation areas constitute the front parameters of the test object; the side view image of the test object is edge segmented by the binary method to obtain the side view image. The edge parameters of the civil air defense door, the edge parameters include length value, width value and thickness value, the length standard value, width standard value and thickness standard value of the standard performance civil air defense door are retrieved through the database; the absolute value of the difference between the length value and the length standard value in the edge parameter of the test object is marked as the length error value CW, and the absolute value of the difference between the width value and the width standard value in the edge parameter of the test object is marked as the width error value KW; the absolute value of the difference between the thickness value and the thickness standard value in the edge parameter of the test object is marked as the thickness error value HW; the error coefficient WC of the side view image is obtained by the formula WC=k1×CW+k2×KW+k3×HW, wherein k1, k2 and k3 are all proportional coefficients, and k1>k2>k3>1; the side parameters of the side view image are constituted by the error coefficient WC of the side view image; the front view image and the side view image of the test object are processed to obtain the front parameters and the side parameters, the front parameters are the front view image extraction features of the test object, and the side parameters are the side image extraction features of the test object, and the front parameters and the side parameters are used to provide data support for the performance test analysis process.
[0039] The performance test module is used to perform performance test analysis on the test object: test the impact resistance and load-bearing capacity of the door leaf through a static test, mark the test object that passes the static test as a qualified strength object, and mark the test object that fails the static test as an unqualified strength object; then retrieve the error coefficient WC of all test objects with unqualified strength and establish an error set; perform variance calculation on all elements in the error set to obtain the precision coefficient of the error set, and compare the precision coefficient with the preset precision threshold: if the precision coefficient is less than the precision threshold, mark the minimum element in the error set as the strength test performance value; if the precision coefficient is greater than or equal to the precision threshold, mark the error set as the strength test performance value. The maximum and minimum elements in the difference set are removed, and then the precision coefficient of the error set is recalculated, and the precision coefficient is compared with the precision threshold again, and so on, until the precision coefficient is less than the precision threshold; the precision threshold is a numerical constant directly retrieved from the database, and the precision coefficient is a numerical value that reflects the discrete degree of all error coefficients WC in the error set. Therefore, this process is a data cleaning process for the error set, so that the numerical distribution of the error coefficients WC retained in the error set tends to be stable. The specific value and value range of the precision threshold can be set by the management personnel according to the accuracy of the feature comparison results; the strength test performance value is sent to the detection and analysis subsystem;
[0040] Then, the pressure difference method is used to detect the sealing performance of the test object by applying a certain air pressure difference inside and outside the door, and the test objects whose sealing performance meets the requirements are marked as qualified sealing objects; the test objects whose sealing performance does not meet the requirements are marked as unqualified sealing objects. The front parameters of all unqualified sealing objects constitute a sealing set, and the sealing set includes a covering subset composed of the covering values in the front parameters of all unqualified sealing objects and a gray bias subset composed of the gray bias values in the front parameters of all unqualified sealing objects; the maximum element and the minimum element in the covering subset constitute a covering range, and the covering range is divided into several covering intervals, and the number of elements in the covering subset whose covering values are in the covering interval is marked as the prominent value of the covering interval; the M1 covering intervals with the largest prominent value are marked as sealed Covering interval; the gray bias range is composed of the maximum element and the minimum element in the gray bias subset, and the gray bias range is divided into several gray bias intervals. The number of elements in the gray bias subset whose gray bias values are in the gray bias interval is marked as the influence value of the gray bias interval, and the M2 gray bias intervals with the largest influence value are marked as sealed gray bias intervals; the sealed coverage interval and the sealed gray bias interval are sent to the detection and analysis subsystem; after the performance test of the test object is carried out in the traditional way, the performance test result is compared with the front parameter and the side parameter, and the rules of the test objects with unqualified performance in the front parameter set and the side parameter set are extracted, so as to generate an image detection basis, and perform performance detection according to the image detection basis, and integrate the multi-performance and multi-process detection process into the image detection process to improve the overall detection efficiency of the civil air defense door.
[0041] Embodiment 2: Figure 3 As shown, all civil air defense doors except the test objects in each production batch of civil air defense door products are marked as detection objects; the detection subsystem is used to perform performance detection and analysis on the detection objects; the detection and analysis subsystem includes a feature extraction module and a feature comparison module.
[0042] The feature extraction module is used to capture images of the detection object and extract image features. The image features include the coverage value and gray deviation value of the front view image corresponding to the detection object relative to the front view image of the standard performance civil air defense door; the image features also include the error coefficient WC of the side view image corresponding to the detection object relative to the side view image of the standard performance civil air defense door; the image features of the detection image are sent to the feature comparison module.
[0043] The feature comparison module is used to perform performance testing on the test object according to the image features of the test object: determine whether the coverage value and the gray deviation value are simultaneously in the sealing coverage interval and the sealing gray deviation interval: if so, the sealing performance of the test object is determined to be unqualified, a sealing re-inspection signal is generated and sent to the performance test module; if not, the sealing performance of the test object is determined to be qualified;
[0044] Determine whether the error coefficient WC is greater than the strength test performance value: if so, determine that the strength performance of the test object is unqualified, generate a strength re-test signal and send the strength re-test signal to the performance test module; if not, determine that the strength performance of the test object is qualified; perform image feature extraction and comparison on the test object, and evaluate the various performances of the test object based on the comparison results, breaking away from the technical limitations of traditional technology that image detection can only detect surface quality, simplifying the data collection and processing process in the performance testing process, and reducing the probability of damage to the civil air defense door during the performance testing process, thereby improving the convenience of performance testing of the civil air defense door.
[0045] The performance detection system for human air defense doors based on image analysis, when working, takes images of various surfaces of the test object, marks the front image and the back image of the test object as front view images, and marks all side images of the test object as side view images; processes the front view image and the side view image of the test object and obtains the front parameter and the side parameter; performs performance test analysis on the test object and marks the test object differently; retrieves the error coefficient WC of the test object of all objects with unqualified strength and establishes an error set, and processes the error set to obtain a strength test performance value; forms a sealing set from the front parameters of all objects with unqualified sealing, and the sealing set includes a covering subset composed of the covering values in the front parameters of all objects with unqualified sealing and a gray deviation subset composed of the gray deviation values in the front parameters of all objects with unqualified sealing; processes the covering subset and the gray deviation subset to obtain a sealing covering interval and a sealing gray deviation interval; evaluates the strength of subsequent detection objects through the strength test performance value; and evaluates the airtightness of subsequent detection objects through the sealing covering interval and the sealing gray deviation interval.
[0046] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.
[0047] The above formulas are obtained by collecting a large amount of data for software simulation and selecting a formula close to the true value. The coefficients in the formula are set by technicians in this field according to actual conditions; for example: formula WC=k1×CW+k2×KW+k3×HW; technicians in this field collect multiple groups of sample data and set corresponding error coefficients for each group of sample data; substitute the set error coefficients and the collected sample data into the formula, any three formulas constitute a three-variable linear equation group, screen the calculated coefficients and take the average, and obtain the values of k1, k2 and k3 as 3.65, 2.83 and 2.12 respectively;
[0048] The size of the coefficient is to quantify each parameter to obtain a specific value for subsequent comparison. The size of the coefficient depends on the amount of sample data and the initial setting of the corresponding error coefficient for each set of sample data by technical personnel in this field; as long as it does not affect the proportional relationship between the parameter and the quantized value, such as the error coefficient is proportional to the length error value.
[0049] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.
[0050] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.
Claims
1. The civil air defense door performance detection system based on image analysis is characterized by: Including data management subsystem, detection and analysis subsystem and database; A number of personal air defense doors are randomly selected from the finished products of each production batch as test objects, and the data management subsystem is used to perform statistical management on the test data of the test objects; the data management subsystem includes an image acquisition module, an image processing module and a performance testing module; Mark all civil air defense doors except the test object in each production batch of civil air defense door products as detection objects; the detection and analysis subsystem is used to perform performance detection and analysis on the detection objects; the detection and analysis subsystem includes a feature extraction module and a feature comparison module; The image acquisition module is used to capture images of each surface of the test object, mark the front image and the back image of the test object as front view images, and mark all side images of the test object as side view images; The image processing module is used to process the front view image and the side view image of the test object and obtain the front view parameters and the side view parameters; The performance test module is used to perform performance test analysis on the test object and obtain the strength test performance value, the sealing coverage interval and the sealing gray deviation interval; The feature extraction module is used to capture images of the detection object and extract image features; The feature comparison module is used to perform performance detection on the detection object according to the image features of the detection object: Determine whether the coverage value and the gray deviation value are both in the sealing coverage interval and the sealing gray deviation interval: if so, determine that the sealing performance of the test object is unqualified, generate a sealing re-inspection signal and send the sealing re-inspection signal to the performance test module; if not, determine that the sealing performance of the test object is qualified; Determine whether the error coefficient WC is greater than the strength test performance value: if so, the strength performance of the test object is determined to be unqualified, a strength re-test signal is generated and the strength re-test signal is sent to the performance test module; if not, the strength performance of the test object is determined to be qualified.
2. The civil air defense door performance detection system based on image analysis according to claim 1 is characterized in that: The process of obtaining the front parameters of the test object includes: enlarging the front view image of the test object into a pixel grid image and performing grayscale transformation, then retrieving the front view image of the standard performance civil air defense door and marking it as the standard image, and overlapping and comparing the front view image of the test object with the standard image: marking the absolute value of the difference between the grayscale value of the pixel grid at the same position in the front view image and the grayscale value in the standard image as the deviation value of the pixel grid, and marking the pixel grid as a conforming grid or a deviation grid according to the deviation value; then all the deviation grids that touch each other in the front view image constitute a deviation area, and marking the number of deviation grids in the deviation area and the average value of the deviation grid grayscale value as the coverage value and gray deviation value of the deviation area respectively; the coverage values and gray deviation values of all deviation areas constitute the front parameters of the test object.
3. The civil air defense door performance detection system based on image analysis according to claim 2 is characterized in that: The specific process of marking a pixel grid as a conforming grid or a deviation grid includes: retrieving the deviation threshold through the database, and comparing the deviation value with the deviation threshold: if the deviation value is less than the deviation threshold, the corresponding pixel grid is marked as a conforming grid; if the deviation value is greater than or equal to the deviation threshold, the corresponding pixel grid is marked as a deviation grid.
4. The civil air defense door performance detection system based on image analysis according to claim 3 is characterized in that: The process of obtaining the side parameters includes: using a binary method to perform edge segmentation on the side image of the test object to obtain edge parameters of the civil air defense door in the side image, the edge parameters include length value, width value and thickness value, and retrieving the standard length value, width value and thickness value of the civil air defense door with standard performance through the database; obtaining the length error value CW, width error value KW and thickness error value HW of the test object and performing numerical calculation to obtain the error coefficient WC of the side image; the error coefficient WC of the side image constitutes the side parameters of the side image; The calculation formula of the error coefficient WC of the side view image is: WC=k1×CW+k2×KW+k3×HW, where k1, k2 and k3 are all proportional coefficients, and k1>k2>k3>1.
5. The civil air defense door performance detection system based on image analysis according to claim 4 is characterized in that: The process of obtaining the length error value CW, the width error value KW and the thickness error value HW includes: marking the absolute value of the difference between the length value in the edge parameter of the test object and the length standard value as the length error value CW, marking the absolute value of the difference between the width value in the edge parameter of the test object and the width standard value as the width error value KW; marking the absolute value of the difference between the thickness value in the edge parameter of the test object and the thickness standard value as the thickness error value HW.
6. The civil air defense door performance detection system based on image analysis according to claim 5 is characterized in that: The process of obtaining the strength test performance value includes: testing the impact resistance and bearing capacity of the door leaf through a static test, marking the test object that passed the static test as a qualified strength object, and marking the test object that failed the static test as an unqualified strength object; then retrieving the error coefficient WC of the test objects of all unqualified strength objects and establishing an error set; performing variance calculation on all elements in the error set to obtain the precision coefficient of the error set, and comparing the precision coefficient with the preset precision threshold: if the precision coefficient is less than the precision threshold, then marking the minimum element in the error set as the strength test performance value; if the precision coefficient is greater than or equal to the precision threshold, then eliminating the maximum element and the minimum element in the error set, and then recalculating the precision coefficient of the error set, and again comparing the precision coefficient with the precision threshold, and so on, until the precision coefficient is less than the precision threshold; the precision threshold is a numerical constant directly retrieved from the database, and the specific value and value range of the precision threshold are set by the management personnel.
7. The civil air defense door performance detection system based on image analysis according to claim 6 is characterized in that: The process of obtaining the sealing coverage interval and the sealing gray deviation interval includes: using the pressure difference method to detect the sealing performance of the test object by applying a certain air pressure difference inside and outside the door, and marking the test object whose sealing performance meets the requirements as a sealed qualified object; marking the test object whose sealing performance does not meet the requirements as a sealed unqualified object, and forming a sealing set by the front parameters of all unqualified sealed objects, and the sealing set includes a covering subset composed of the coverage values in the front parameters of all unqualified sealed objects and a gray deviation subset composed of the gray deviation values in the front parameters of all unqualified sealed objects; forming a covering range by the maximum element and the minimum element in the covering subset, dividing the covering range into several covering intervals, and marking the number of elements in the covering subset whose covering values are within the covering interval as the prominent values of the covering interval; marking the M1 covering intervals with the largest prominent value as sealed covering intervals; forming a gray deviation range by the maximum element and the minimum element in the gray deviation subset, dividing the gray deviation range into several gray deviation intervals, marking the number of elements in the gray deviation subset whose gray deviation values are within the gray deviation interval as the influence value of the gray deviation interval, and marking the M2 gray deviation intervals with the largest influence value as sealed gray deviation intervals.
8. The civil air defense door performance detection system based on image analysis according to claim 7 is characterized in that: The image features of the detection object include: the image features include the coverage value and gray deviation value of the front view image corresponding to the detection object relative to the front view image of the standard performance civil air defense door; the image features also include the error coefficient WC of the side view image corresponding to the detection object relative to the side view image of the standard performance civil air defense door.
