A multifunctional intelligent bag damage detection method and system

Abstract

The invention discloses a multifunctional intelligent detection method and system for luggage damage, which belongs to the technical field of luggage damage detection. The method comprises: acquiring information of each luggage in real time, screening the information of each luggage to obtain standard luggage information, and establishing a luggage information library according to the information of each standard luggage; setting an information module according to the luggage information library, and a user collects image data of the luggage to be inspected through the information module to obtain a collected image of the luggage to be inspected; acquiring the collected image and standard luggage information corresponding to the luggage to be inspected, setting corresponding damage standard data for each part in the collected image according to the standard luggage information, performing damage detection on the luggage to be inspected in the collected image according to the damage standard data, and determining a corresponding appearance damage detection result; establishing a quality library, collecting deformation analysis data of the luggage to be inspected, matching quality verification data according to the luggage to be inspected, performing quality damage detection according to the quality verification data and the deformation analysis data, and obtaining a quality damage detection result.

Description

Technical Field

[0001] The present invention belongs to the technical field of luggage damage detection, and in particular to a multifunctional luggage damage intelligent detection method and system. Background Art

[0002] With the improvement of living standards, bags are becoming an important tool for people's daily travel and storage, and their use frequency and importance are increasing day by day. However, during use, bags will inevitably suffer various forms of damage, such as scratches, dents, and breakages. These damages not only affect the appearance of bags, but also may reduce their service life and functionality.

[0003] At present, the detection and assessment of luggage damage mostly rely on manual visual inspection or the use of simple measuring tools. This method has many shortcomings: first, manual inspection is time-consuming and labor-intensive, and is easily affected by subjective factors, making it difficult to ensure the accuracy and consistency of the test results; second, simple measuring tools can often only detect the surface phenomenon of damage and cannot conduct an in-depth assessment of the degree of damage; finally, complex or hidden damage is often difficult to detect through manual inspection, which increases the risk of luggage use; especially for many individual users, it is difficult to accurately assess the damage of their own luggage.

[0004] Based on this, the present invention provides a multifunctional luggage damage intelligent detection method and system. Summary of the invention

[0005] In order to solve the problems existing in the above-mentioned solutions, the present invention provides a multifunctional luggage damage intelligent detection method and system.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A multifunctional intelligent bag damage detection method, the method comprising:

[0008] Step 1: Acquire the information of each bag in real time, screen the information of each bag, obtain standard bag information, and establish a bag information database according to the information of each standard bag;

[0009] Furthermore, the method for screening the information of each bag includes:

[0010] The management personnel select less than K pieces of standard bag information and no less than U pieces of reference bag information;

[0011] Identify the brand, price and recycling rate corresponding to each standard bag information and reference bag information;

[0012] Formula for setting identification value: DW = β1×PB+β2×BG+β3×e λ ;

[0013] Where: DW is the identification value; PB is the brand value; BG is the price; λ is the recovery rate; e is a constant; β1, β2, β3 are the corresponding adjustment coefficients, and the value range is (0, 1];

[0014] Optimize each adjustment coefficient according to the information of each bag;

[0015] The screening model is established by the identification value formula, and the expression of the screening model is: X1 and X2 are both thresholds; and X1>X2; output data Wie screening value;

[0016] Analyze the information of each bag through the screening model to obtain the screening value corresponding to the information of each bag;

[0017] Mark the luggage information with a filter value of 1 as standard luggage information; mark the luggage information with a filter value of 0 as retained information;

[0018] Integrate all retained information to establish a process library.

[0019] Furthermore, each adjustment coefficient is optimized through the optimization function, and the expression of the optimization function is:

[0020] δ is a constant. When μ=min{DW} and YP(DW)=0.5, σ is the distance from DW to min{DW}.

[0021] Step 2: An information module is set according to the luggage information library, and the user collects image data of the luggage to be inspected through the information module to obtain the collected image corresponding to the luggage to be inspected;

[0022] Furthermore, the information module is used for the user to input the luggage information to be inspected, and to identify the luggage information to be inspected, and to match the identified luggage information to be inspected with each standard luggage information in the luggage information library to determine whether it falls within the detection range;

[0023] When it is judged that it does not fall within the detection range, the analysis is stopped and no subsequent operations are performed;

[0024] When it is determined that it belongs to the detection range, the corresponding standard luggage information is output; when the user collects image data, the image of the luggage to be inspected is collected according to the preset image collection method.

[0025] Furthermore, the preset image acquisition method is:

[0026] Identify the user's mobile phone model, and obtain corresponding camera data according to the mobile phone model;

[0027] Instructing the user to collect a comparison image combination, wherein the comparison image combination includes a comparison image and a reference image;

[0028] Performing grayscale processing on the contrast image and the reference image in the contrast image combination to obtain a corresponding contrast grayscale image and a reference grayscale image; identifying the grayscale value of the reference grayscale image and marking it as a reference grayscale value;

[0029] Identify the grayscale value of each pixel in the contrast grayscale image and mark it as the contrast grayscale value; merge the pixels according to the contrast grayscale values ​​to obtain each unit area; mark the contrast grayscale value corresponding to each unit area as a unit value; segment the contrast grayscale image according to each unit value to obtain a grayscale segmentation image;

[0030] Determine whether the camera parameters are qualified based on the grayscale segmentation image and the reference grayscale value;

[0031] When it is determined that the camera acquisition is unqualified, the specific gravity value and grayscale difference of each unit area are calculated, the specific gravity value and grayscale difference of the unit area are integrated into adjustment points, and each adjustment point is integrated into an adjustment set;

[0032] The corresponding camera adjustment parameters are determined according to the adjustment set, and the camera is adjusted according to the obtained camera adjustment parameters; and the image is acquired through the adjusted camera.

[0033] Furthermore, the method of combining pixels according to the contrasting gray values ​​includes:

[0034] Step SA1: randomly select a number of non-adjacent pixels in non-unit areas as initial pixels, and identify the contrast grayscale value of each initial pixel; mark the pixels adjacent to the initial pixels as candidate pixels, and identify the contrast grayscale value of the candidate pixels;

[0035] When the contrast grayscale value of the initial pixel is the same as the contrast grayscale value of the corresponding pixel to be selected, the initial pixel and the pixel to be selected are merged to obtain the initial area;

[0036] Otherwise, no merger will be performed;

[0037] Step SA2: Mark the pixels adjacent to the initial area as candidate pixels, and identify the contrast grayscale values ​​of the candidate pixels;

[0038] When the contrast grayscale value of the initial region is the same as the contrast grayscale value of the corresponding pixel point to be selected, the initial region and the pixel point to be selected are merged to obtain a new initial region;

[0039] Otherwise, no merger will be performed;

[0040] Step SA3: looping step SA2 until there are no pixels to be merged in the initial area, marking the current initial area as a unit area;

[0041] Step SA4: looping steps SA1 to SA3 until there is no initial pixel point. A single initial pixel point is also a unit area.

[0042] Furthermore, the method for segmenting the contrast grayscale image according to each unit value includes:

[0043] Calculate the grayscale difference between each unit area and the reference grayscale value; establish a segmentation model, the expression of the segmentation model is

[0044] Where: s is the grayscale difference, s′ is the standard deviation; the output data is the segmentation value;

[0045] The grayscale difference of each unit area is analyzed through the segmentation model to obtain the segmentation value corresponding to each unit area, and each unit area with a segmentation value of 1 is eliminated to obtain a grayscale segmentation image.

[0046] Furthermore, the method of determining whether the camera parameters are qualified according to the grayscale segmentation image and the reference grayscale value includes:

[0047] Identify the area of ​​each unit region, marked as unit area Ai, i represents the corresponding unit region, i = 1, 2, ..., n, n is a positive integer;

[0048] According to the formula Calculate image deviation value;

[0049] Where: PC is the image deviation value; h′ represents the reference gray value; hci represents the gray difference value of the corresponding unit area;

[0050] When the image deviation value is greater than the threshold value X3, the camera parameters are judged to be unqualified;

[0051] When the image deviation value is not greater than the threshold value X3, the camera parameters are determined to be qualified.

[0052] Step 3: Obtain the collected image and the standard luggage information corresponding to the luggage to be inspected, set the corresponding damage standard data for each part in the collected image according to the standard luggage information, perform damage detection on the luggage to be inspected in the collected image according to the damage standard data, and determine the corresponding appearance damage detection result;

[0053] Step 4: Establish a quality library, which is used to store quality verification data. The quality verification data includes abnormal deformation data and standard deformation data. Collect deformation analysis data of the bags to be inspected, match the corresponding quality verification data according to the bags to be inspected, perform quality damage detection according to the quality verification data and deformation analysis data, and obtain corresponding quality damage detection results.

[0054] A multifunctional intelligent bag damage detection system, comprising a bag information library, an information module and a detection module;

[0055] The luggage information database is used to store information of each standard luggage;

[0056] The information module is used for the user to input the luggage information to be inspected, identify the luggage information to be inspected, match the identified luggage information to be inspected with the standard luggage information in the luggage information library, and determine whether it falls within the detection range;

[0057] When it is judged that it does not fall within the detection range, the analysis is stopped and no subsequent operations are performed;

[0058] When it is determined that it belongs to the detection range, the corresponding standard bag information is output; when the user collects image data, the image of the bag to be inspected is collected according to the preset image collection method;

[0059] The detection module is used to perform damage detection on the luggage to be inspected, obtain the collected image and the standard luggage information of the luggage to be inspected, set corresponding damage standard data for each part in the collected image according to the standard luggage information, perform damage detection on the luggage to be inspected in the collected image according to the damage standard data, and determine the corresponding appearance damage detection result;

[0060] A quality library is established, deformation analysis data of bags to be inspected are collected, corresponding quality verification data is matched according to bags to be inspected, quality damage detection is performed according to the quality verification data and deformation analysis data, and corresponding quality damage detection results are obtained.

[0061] Compared with the prior art, the present invention has the following beneficial effects:

[0062] The present invention uses advanced image processing and artificial intelligence technologies to achieve rapid and accurate identification of luggage damage. Compared with traditional manual visual inspection, the present invention not only greatly improves the detection speed, but also significantly improves the accuracy of the detection results, and reduces errors caused by human factors; at the same time, it solves the problem that users find it difficult to rely on their own abilities to perform accurate luggage damage detection, and at the same time, it intelligently adjusts the acquisition parameters to solve the problem of damage detection errors caused by differences in acquired images. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] 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.

[0064] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0065] 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, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0066] like Figure 1 As shown, a multifunctional intelligent bag damage detection method comprises:

[0067] Step 1: Obtain various luggage information, such as luggage name, type, model, brand, price, shape, color, second-hand recycling rate, image and other related information. The second-hand recycling rate refers to the probability that the luggage can be sold if the consumer wants to resell it. It is collected according to the corresponding data statistics, or set according to the evaluation data of the merchant who purchased the luggage;

[0068] The platform sets the detection scope, that is, the platform sets under what circumstances the bags and luggage can be used as the detection target of the system according to business needs. In actual applications, many users of low-end brands and low-priced bags and luggage do not have the need for damage detection. Therefore, the platform needs to set the detection scope according to its business needs, form a bag and luggage evaluation model, and use the bag and luggage evaluation model to screen the information of each bag and luggage to determine the bags and luggage within the detection range; integrate and establish a bag and luggage database, that is, the bag and luggage database is used to store various bag and luggage data that can be detected for damage.

[0069] The detailed process is as follows:

[0070] Collect various luggage information currently on the market, and the platform management personnel manually select K or more luggage information that meet the requirements from the luggage information, and mark them as standard luggage information, where K is a positive integer, for example, K=100; then select U or more luggage information that does not meet the requirements, and mark them as reference luggage information, where U is a positive integer;

[0071] Identify the brand, price and recycling rate corresponding to each standard bag information and reference bag information;

[0072] Formula for setting identification value: DW = β1×PB+β2×BG+β3×e λ ;

[0073] In the formula: DW is the identification value; PB is the brand value; BG is the price; λ is the recovery rate; e is a constant; β1, β2, and β3 are the corresponding adjustment coefficients, and the value range is (0, 1]; each parameter is calculated by removing the dimension and taking its numerical value; the brand value is set according to the various luggage brands, and each luggage brand will be divided into different levels first. For example, the current mainstream recognized luggage brand ranking can be applied, such as the first-level LV and Hermes, etc., mainly to achieve the level recognition of large and medium-sized brands and the detailed setting of the brand value of the corresponding level, and the subsequent many niche brands can be set uniformly; that is, the large and medium-sized brands with higher popularity can set the corresponding brand values ​​one by one, and the number of large and medium-sized brands is small; many small brands can directly set a unified brand value, or make corresponding corrections based on the corresponding popularity to achieve the brand value of each luggage brand, which is specifically set by professionals; and many small brands are generally not the target audience of the platform, so they can have a larger allowable error;

[0074] The threshold value X1 is determined according to the information of each standard bag, the reference bag information and the identification value formula, that is, a value greater than the threshold value X1 can be considered to meet the detection range, and is set according to the calculated minimum value of each standard identification value and the maximum value of each reference identification value;

[0075] Analyze each piece of luggage information through the identification value formula to obtain the identification value corresponding to each piece of luggage information, and mark the luggage information with an identification value greater than the threshold value X1 as standard luggage information;

[0076] Summarize the information of each standard luggage and establish a luggage database.

[0077] A threshold value X2 smaller than the threshold value X1 is set, and the luggage information with identification values ​​between the threshold values ​​X1 and X2 is marked as retained information, which is used by management personnel to query whether each retained information can be used as standard luggage information; at the same time, the identification formula can be optimized and adjusted according to the selection of management personnel.

[0078] The adjustment coefficient may be set according to the prior art in one embodiment, or may be directly set by professionals.

[0079] In another embodiment, the adjustment coefficient may be set as follows; the method includes:

[0080] Continuous simulation verification is carried out according to the information of each standard bag and the reference bag, and the adjustment coefficient is optimized through the genetic algorithm. The genetic algorithm toolbox provided in the MATLAB software can be used for iterative calculation.

[0081] Generate random numbers for the adjustment coefficients β1, β2, and β3 in (0, 1] and establish an optimization function. The expression of the optimization function is:

[0082] δ is a constant, and different values ​​correspond to different optimization functions, which can be 1, 2, 3, etc.; when μ=min{DW}, YP(DW)=0.5, σ is the distance from DW to min{DW}, and the initial values ​​of μ and σ can be set manually. Usually, the values ​​of σ and μ are continuously corrected by the next generation of crossover and mutation evolution of the genetic algorithm to obtain the value of DW, which is substituted into the evaluation value formula to optimize the initial adjustment coefficients β1, β2, β3, and the optimal solution of the adjustment coefficient can be obtained;

[0083] In the future, recommendations will be made based on the information of each bag, and the adjustment coefficient will be optimized again according to the subsequent user's choices.

[0084] Step 2: An information module is set according to the luggage information library, and the user collects image data of the luggage to be inspected through the information module to obtain the collected image corresponding to the luggage to be inspected;

[0085] The information module is used for users to input information of bags to be inspected, identify the bags to be inspected information, match the identified bags to be inspected information with each standard bag information in the bag information library, and determine whether it falls within the detection range; when it is determined that it does not fall within the detection range, stop the analysis and do not perform subsequent operations; when it is determined that it falls within the detection range, output the corresponding standard bag information; when the user performs image data acquisition, the image of the bags to be inspected is acquired according to the preset image acquisition method.

[0086] In one embodiment, the preset image acquisition method is an existing acquisition method, such as retrieving the camera permission of the user's mobile phone, and the user uses the mobile phone to acquire images of the bag to be inspected.

[0087] In another embodiment, although the data collection of the bags to be inspected can be realized according to the above implementation, in the actual application process, due to the influence of various factors such as mobile phone model, camera parameters, lighting, etc., the collected images uploaded by the user have certain differences from the appearance of the actual bags to be inspected, which makes the subsequent damage detection have large errors, especially most users do not have professional camera knowledge; based on this, this embodiment provides the following collection method:

[0088] Identify the user's mobile phone model, and obtain corresponding camera data based on the identified mobile phone model, such as display mode, version, various camera parameters, and other camera data;

[0089] Guide users to collect comparison image combinations. For example, display a color card with various colors on it. Users select objects with the same color as those on the color card and mark them as comparison objects. Use a mobile phone to collect images of the comparison objects to obtain comparison images. Generate a reference image based on the color card color corresponding to the comparison image. Arrange the reference image and the comparison image into a comparison image combination.

[0090] The contrast image and the reference image in the contrast image combination are gray-scale processed to obtain a corresponding contrast gray-scale image and a reference gray-scale image, because the reference gray-scale image identifies a standard pure color image, that is, the gray-scale value is the same, and there is only one gray-scale value; the gray-scale value of the reference gray-scale image is identified and marked as the reference gray-scale value;

[0091] Identify the grayscale value of each pixel in the contrast grayscale image and mark it as the contrast grayscale value; merge adjacent pixels with the same grayscale value to obtain a number of unit areas, and the grayscale values ​​in the unit areas are the same; mark the contrast grayscale value corresponding to the unit area as the unit value;

[0092] Calculate the grayscale difference between each unit area and the reference grayscale value, and determine the grayscale difference that may appear in the actual shooting process according to the color of the color card. Because even if there is a deviation, the grayscale difference of the same color is also small, so set the standard deviation value corresponding to each color card color according to each grayscale difference; that is, if it exceeds the standard deviation value, it can be identified as an image of a non-contrast object, because there must be other background images during the acquisition process; establish a segmentation model based on the preset standard deviation value, and the expression of the segmentation model is Where: s is the grayscale difference, s′ is the standard deviation; the output data is the segmentation value;

[0093] The grayscale difference of each unit area is analyzed through the segmentation model to obtain the segmentation value corresponding to each unit area, and each unit area with a segmentation value of 1 is eliminated to achieve the segmentation of the contrast grayscale image and obtain the grayscale segmentation image;

[0094] Identify the area of ​​each unit region, marked as unit area Ai, i represents the corresponding unit region, i = 1, 2, ..., n, n is a positive integer;

[0095] According to the formula Calculate the image deviation value; where: PC is the image deviation value; h′ represents the reference gray value; hci represents the gray difference value of the corresponding unit area;

[0096] When the image deviation value is greater than the threshold value X3, the camera parameters are judged to be unqualified; otherwise, the camera parameters are judged to be qualified;

[0097] When the camera acquisition fails, the weight value of each unit area is calculated, that is, the ratio of its area to the sum of the areas of all unit areas, and the weight value and grayscale difference of the unit area are marked as adjustment points, and each adjustment point is integrated into an adjustment set;

[0098] Based on the camera data, various camera parameter failures that may be encountered are obtained, and the corresponding adjustment set is identified. Professional camera personnel determine how to adjust the camera parameters to overcome the problems corresponding to the adjustment set, and form corresponding camera adjustment parameters, or download the adjustment parameters of the specified camera; the camera is adjusted according to the obtained camera adjustment parameters. If automatic adjustment can be achieved, intelligent adjustment is performed, otherwise the user is guided to make adjustments.

[0099] Step 3: Obtain the collected image and the standard luggage information corresponding to the luggage to be inspected, and set the corresponding damage standard data for each part in the collected image according to the standard luggage information, such as the standard image, color and other related data of the position; perform damage detection on each part of the luggage to be inspected in the collected image according to the obtained damage standard data, use the existing image recognition technology to compare, judge whether it meets the damage standard data, and determine the corresponding appearance damage detection result according to the judgment result;

[0100] Step 4: According to the standard luggage data of the luggage to be inspected, the normal deformation data of the luggage to be inspected when it is subjected to external forces such as pressure is obtained, that is, the posture and shape of each part under normal conditions under force. Because when the luggage material is damaged, the surface may be normal, but the deformation process after being subjected to force is quite different; therefore, the corresponding normal deformation data can be determined according to its standard luggage data, and the corresponding identification features can be determined and integrated into standard deformation data. Similarly, a large amount of abnormal deformation data is obtained, and the corresponding damage type, grade, degree and other labels are marked; the obtained abnormal deformation data and standard deformation data are integrated into the quality verification data of the luggage to be inspected, and the quality verification data of each luggage are integrated to establish the corresponding quality library; that is, the quality library is established by the platform.

[0101] Match the corresponding quality verification data from the quality library according to the standard luggage information, and identify the standard deformation data in the quality verification data;

[0102] The force-collected image data of the bags to be inspected are marked as deformation analysis data, and the deformation analysis data are compared with the standard deformation data to determine whether they meet the standard deformation data requirements. If they meet the standard deformation requirements, the quality damage test is judged to be qualified; if they do not meet the standard deformation requirements, the quality loss test is judged to be unqualified, and the deformation analysis data is compared with each abnormal deformation data to determine the damage type and degree; and the corresponding quality damage test results are obtained through integration.

[0103] A multifunctional intelligent bag damage detection system, comprising a bag information library, an information module and a detection module;

[0104] The luggage information database is used to store information of each standard luggage;

[0105] The information module is used for the user to input the luggage information to be inspected, identify the luggage information to be inspected, match the identified luggage information to be inspected with the standard luggage information in the luggage information library, and determine whether it falls within the detection range;

[0106] When it is judged that it does not fall within the detection range, the analysis is stopped and no subsequent operations are performed;

[0107] When it is determined that it belongs to the detection range, the corresponding standard luggage information is output; when the user collects image data, the image of the luggage to be inspected is collected according to the preset image collection method.

[0108] The detection module is used to perform damage detection on the luggage to be inspected, obtain the collected image and the standard luggage information of the luggage to be inspected, set corresponding damage standard data for each part in the collected image according to the standard luggage information, perform damage detection on the luggage to be inspected in the collected image according to the damage standard data, and determine the corresponding appearance damage detection result;

[0109] A quality library is established, deformation analysis data of bags to be inspected are collected, corresponding quality verification data is matched according to bags to be inspected, quality damage detection is performed according to the quality verification data and deformation analysis data, and corresponding quality damage detection results are obtained.

[0110] The above formulas are all calculated by removing dimensions and taking numerical values. The formula is a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained by simulating a large amount of data.

[0111] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A multifunctional intelligent bag damage detection method, characterized in that: Methods include: Step 1: Acquire the information of each bag in real time, screen the information of each bag, obtain standard bag information, and establish a bag information database according to the information of each standard bag; Step 2: An information module is set according to the luggage information library, and the user collects image data of the luggage to be inspected through the information module to obtain the collected image corresponding to the luggage to be inspected; The information module is used for the user to input the luggage information to be inspected, identify the luggage information to be inspected, match the identified luggage information to be inspected with the standard luggage information in the luggage information library, and determine whether it falls within the detection range; When it is judged that it does not fall within the detection range, the analysis is stopped and no subsequent operations are performed; When it is judged to be within the detection range, the corresponding standard bag information is output; When the user collects image data, the image of the bag to be inspected is collected according to a preset image collection method; Step 3: Obtain the collected image and the standard luggage information corresponding to the luggage to be inspected, set the corresponding damage standard data for each part in the collected image according to the standard luggage information, perform damage detection on the luggage to be inspected in the collected image according to the damage standard data, and determine the corresponding appearance damage detection result; Step 4: Establish a quality library, which is used to store quality verification data. The quality verification data includes abnormal deformation data and standard deformation data. Collect deformation analysis data of the bags to be inspected, match the corresponding quality verification data according to the bags to be inspected, perform quality damage detection according to the quality verification data and deformation analysis data, and obtain corresponding quality damage detection results.

2. A multifunctional intelligent bag damage detection method according to claim 1, characterized in that: The default image acquisition method is: Identify the user's mobile phone model, and obtain corresponding camera data according to the mobile phone model; Instructing the user to collect a comparison image combination, wherein the comparison image combination includes a comparison image and a reference image; Performing grayscale processing on the contrast image and the reference image in the contrast image combination to obtain a corresponding contrast grayscale image and a reference grayscale image; identifying the grayscale value of the reference grayscale image and marking it as a reference grayscale value; Identify the grayscale value of each pixel in the contrast grayscale image and mark it as the contrast grayscale value; merge the pixels according to the contrast grayscale values ​​to obtain each unit area; mark the contrast grayscale value corresponding to each unit area as a unit value; segment the contrast grayscale image according to each unit value to obtain a grayscale segmentation image; Determine whether the camera parameters are qualified based on the grayscale segmentation image and the reference grayscale value; When it is determined that the camera acquisition is unqualified, the specific gravity value and grayscale difference of each unit area are calculated, the specific gravity value and grayscale difference of the unit area are integrated into adjustment points, and each adjustment point is integrated into an adjustment set; The corresponding camera adjustment parameters are determined according to the adjustment set, and the camera is adjusted according to the obtained camera adjustment parameters; and the image is acquired through the adjusted camera.

3. A multifunctional intelligent bag damage detection method according to claim 2, characterized in that: The method of combining pixels according to the contrasting gray values ​​includes: Step SA1: randomly select a number of non-adjacent pixels in non-unit areas as initial pixels, and identify the contrast grayscale value of each initial pixel; mark the pixels adjacent to the initial pixels as candidate pixels, and identify the contrast grayscale value of the candidate pixels; When the contrast grayscale value of the initial pixel is the same as the contrast grayscale value of the corresponding pixel to be selected, the initial pixel and the pixel to be selected are merged to obtain the initial area; Otherwise, no merger will be performed; Step SA2: Mark the pixels adjacent to the initial area as candidate pixels, and identify the contrast grayscale values ​​of the candidate pixels; When the contrast grayscale value of the initial region is the same as the contrast grayscale value of the corresponding pixel point to be selected, the initial region and the pixel point to be selected are merged to obtain a new initial region; Otherwise, no merger will be performed; Step SA3: looping step SA2 until there are no pixels to be merged in the initial area, marking the current initial area as a unit area; Step SA4: loop step SA1 to step SA3 until there is no initial pixel point.

4. A multifunctional intelligent bag damage detection method according to claim 2, characterized in that: Methods for segmenting contrasting grayscale images based on cell values ​​include: Calculate the grayscale difference between each unit area and the reference grayscale value; establish a segmentation model, the expression of the segmentation model is ; Where: s is the grayscale difference, s' is the standard deviation; the output data is the segmentation value; The grayscale difference of each unit area is analyzed through the segmentation model to obtain the segmentation value corresponding to each unit area, and each unit area with a segmentation value of 1 is eliminated to obtain a grayscale segmentation image.

5. The multifunctional intelligent bag damage detection method according to claim 2 is characterized in that: Methods for determining whether camera parameters are qualified based on grayscale segmentation images and reference grayscale values ​​include: Identify the area of ​​each unit region, marked as unit area Ai, i represents the corresponding unit region, i=1, 2, ..., n, n is a positive integer; According to the formula Calculate image deviation value; Where: PC is the image deviation value; h´ represents the reference grayscale value; hci represents the grayscale difference of the corresponding unit area; When the image deviation value is greater than the threshold value X3, the camera parameters are judged to be unqualified; When the image deviation value is not greater than the threshold value X3, the camera parameters are determined to be qualified.

6. A multifunctional intelligent bag damage detection method according to claim 1, characterized in that: Methods for filtering information of each bag include: The management personnel select less than K pieces of standard bag information and no less than U pieces of reference bag information; Identify the brand, price and recycling rate corresponding to each standard bag information and reference bag information; Set the identification value formula: ; Where: DW is the identification value; PB is the brand value; BG is the price; λ is the recovery rate; e is a constant; β1, β2, β3 are the corresponding adjustment coefficients, and the value range is (0, 1]; Optimize each adjustment coefficient according to the information of each bag; The screening model is established by the identification value formula, and the expression of the screening model is: ; X1 and X2 are both thresholds; and X1>X2; the output data is the screening value; Analyze the information of each bag through the screening model to obtain the screening value corresponding to the information of each bag; Mark the luggage information with a filter value of 1 as standard luggage information; mark the luggage information with a filter value of 0 as retained information; Integrate all retained information to establish a process library.

7. A multifunctional intelligent bag damage detection system, characterized in that: A multifunctional intelligent bag damage detection method according to any one of claims 1 to 6 is implemented, comprising a bag information library, an information module and a detection module; The luggage information database is used to store information of each standard luggage; The information module is used for the user to input the luggage information to be inspected, identify the luggage information to be inspected, match the identified luggage information to be inspected with the standard luggage information in the luggage information library, and determine whether it falls within the detection range; When it is judged that it does not fall within the detection range, the analysis is stopped and no subsequent operations are performed; When it is determined that it belongs to the detection range, the corresponding standard bag information is output; when the user collects image data, the image of the bag to be inspected is collected according to the preset image collection method; The detection module is used to perform damage detection on the luggage to be inspected, obtain the collected image and the standard luggage information of the luggage to be inspected, set corresponding damage standard data for each part in the collected image according to the standard luggage information, perform damage detection on the luggage to be inspected in the collected image according to the damage standard data, and determine the corresponding appearance damage detection result; A quality library is established, deformation analysis data of bags to be inspected are collected, corresponding quality verification data is matched according to bags to be inspected, quality damage detection is performed according to the quality verification data and deformation analysis data, and corresponding quality damage detection results are obtained.

Images