Method for matching passenger information with luggage security check image based on artificial intelligence

Abstract

The invention discloses a passenger information and luggage security check image matching method based on artificial intelligence, and relates to the technical field of security check, and the method comprises the steps: arranging a human body camera and an article camera at a security check entrance, capturing front and side images of a passenger, extracting face feature vectors, and generating a unique identity tag; meanwhile, recording external characteristics of the luggage, and binding the characteristics with the passenger identity tag to generate matching information; after a passenger enters a security check channel, a human body camera extracts a dynamic behavior track of the passenger by using a gait recognition technology, monitors the relative position of the passenger and luggage in real time by using a multi-target tracking algorithm, and continuously updates binding information to cope with a cross scene; and when the luggage passes through the security inspection equipment, the article camera captures an X-ray image and extracts the internal characteristics of the luggage through a YOLOv8 model, the internal characteristics, the external characteristics and the dynamic behavior track are combined, joint analysis is performed through a multi-modal fusion algorithm, and when an abnormal condition is detected, backtracking analysis is automatically triggered.

Description

Technical Field

[0001] The present invention relates to the field of security inspection technology, and in particular to a method for matching passenger information with baggage security inspection images based on artificial intelligence. Background Art

[0002] At public transportation hubs such as airports or high-speed rail stations, security checks on passenger luggage are an important part of ensuring transportation safety. The method of matching passenger information with luggage mainly relies on binding passenger identity information (such as boarding passes, ID cards) with luggage tags (such as barcodes, RFID tags). When the luggage passes through the security inspection equipment, the equipment will record its X-ray image and associate the image with the luggage tag for subsequent tracking and identification.

[0003] However, in the case of crossed luggage or wrong luggage, the recognition effect of traditional solutions cannot meet the usage requirements. First, traditional solutions have poor adaptability to dynamic scenes. When luggage and passengers are separated or cross-moved after security checks, it is difficult to accurately track the ownership of the luggage, which can easily lead to matching errors. Secondly, traditional methods are highly dependent on the integrity of luggage tags. Once the tags fall off or are damaged, the matching process will be interrupted. In addition, when multiple passengers enter the security check area at the same time and there are many pieces of luggage, the manual matching of luggage tags and X-ray images may be delayed or erroneous, further affecting the efficiency and accuracy of security checks.

[0004] Some solutions address the above problems by strengthening manual intervention, but such measures not only increase labor costs, but also easily cause new problems due to communication problems. Therefore, how to improve matching accuracy and efficiency in dynamic and complex scenarios has become an urgent problem that needs to be solved in the current security inspection field. Summary of the invention

[0005] In view of the above existing problems, the present invention is proposed.

[0006] The present invention provides a method for matching passenger information with baggage security inspection images based on artificial intelligence to solve the problem that traditional solutions have poor adaptability to dynamic scenes. When the luggage and passengers are separated or cross-moved after security inspection, it is difficult to accurately track the ownership of the luggage, which easily leads to matching errors.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] The embodiment of the present invention provides a method for matching passenger information with baggage security inspection images based on artificial intelligence, which includes:

[0009] Step S1, a body camera and an item camera are arranged at an additional location at the security check entrance. The body camera captures the front and side images of the passenger and generates a unique passenger identity tag in combination with face recognition technology; the item camera records the external features of the luggage from a top perspective and binds them to the passenger identity tag to generate binding information;

[0010] External features of luggage include geometric shape, color, and specific markings, including iconic stickers and unusual appearance;

[0011] Step S2: After the luggage enters the security inspection channel, the human body camera uses gait recognition technology to capture the passenger's body features and dynamic behavior trajectory, and monitors the relative relationship between the passenger and the luggage in real time; the multi-target tracking DeepSORT algorithm is used to track the trajectory of the passenger and the luggage, and the binding information is continuously updated;

[0012] Step S3: When the luggage passes through the security inspection equipment, the item camera captures the X-ray image and extracts the internal features of the luggage through the YOLOv8 model. The internal features are combined with the external features of step S1 and the behavior trajectory of step S2 for joint analysis to determine the completeness of the matching relationship between the luggage and the passenger.

[0013] As a preferred solution of the method for matching passenger information with baggage security inspection images based on artificial intelligence described in the present invention, the step of generating a unique passenger identity tag by combining face recognition technology is as follows:

[0014] Input passenger image set I = {I f ,I s}, where I f Represents the front image of the passenger, I s Represents the profile image of the passenger,

[0015] Extract the facial feature vector, the extraction formula is:

[0016] F=Softmax(W3·tanh(W2·tanh(W1·[I f ,I s ]+b1)+b2)+b3),

[0017] in, is the passenger feature vector, f j is the jth eigenvalue, j is the feature index, tanh(x) is the hyperbolic tangent activation function, W i is the weight matrix of the i-th layer network, b i is the bias vector of the i-th layer network, d f is the dimension of the feature vector,

[0018] Generate a unique ID tag based on the feature vector k, the generating formula is:

[0019]

[0020] Among them, ID k is the identity label of the kth passenger, w j is the weight factor of the eigenvalue, φ(f j )=ln(1+|f j |) is the feature normalization function, j is the feature index, and M is the modular operation base.

[0021] As a preferred solution of the method for matching passenger information with baggage security inspection images based on artificial intelligence described in the present invention, the step of binding with the passenger identity tag to generate binding information is:

[0022] Input luggage image I L , extract the external feature vector, the extraction formula is:

[0023] L ext =[f shape (I L ),f color (I L ),f mark (I L )],

[0024] in,

[0025]

[0026] N is the number of points of the luggage contour, (x i ,y i ) is the coordinate of the ith point, is the center point of the point set, α i is the weight factor of the point,

[0027]

[0028] c m is the color component mean, is the overall color mean, R, G, B are the red, green, and blue channels,

[0029]

[0030] m i is the grayscale value of the iconic sticker,

[0031] Generate binding matrix B k , B k =ID k L ext .

[0032] As a preferred solution of the method for matching passenger information with baggage security inspection images based on artificial intelligence described in the present invention, the step of capturing the passenger's body features and dynamic behavior trajectory through gait recognition technology and monitoring the relative relationship between the passenger and the baggage in real time is as follows:

[0033] Input gait image sequence Among them G t Represents the gait image at time t, extracting the gait feature vector G k , the extraction formula is:

[0034]

[0035] in, is the gait feature, T is the time series length, G t is the gait image, W g ,b g are the network weights and biases,

[0036] Generate dynamic behavior trajectory, the generation formula is:

[0037]

[0038] in,

[0039] are the lateral and longitudinal velocities of the trajectory, t ′ is the time variable.

[0040] As a preferred solution of the method for matching passenger information with baggage security inspection images based on artificial intelligence described in the present invention, the steps of tracking the trajectories of passengers and baggage using the multi-target tracking DeepSORT algorithm and continuously updating the binding information are as follows:

[0041] Define the trajectory update model, the model formula is:

[0042]

[0043] in, is the updated trajectory,

[0044] Δ t is the time interval, and γ is the correction factor.

[0045] As a preferred solution of the method for matching passenger information and baggage security inspection images based on artificial intelligence described in the present invention, wherein: in the process of the joint analysis, a global attention mechanism is simultaneously introduced to optimize feature association;

[0046] During the joint analysis, when an abnormal situation is detected, a backtracking analysis is automatically performed.

[0047] As a preferred solution of the method for matching passenger information with baggage security inspection images based on artificial intelligence described in the present invention, the retrospective analysis method is:

[0048] Reconfirm the ownership relationship between luggage and passengers by combining historical internal features, external features and behavioral trajectories;

[0049] If the anomaly cannot be resolved during retrospective analysis, an alarm is immediately generated, prompting security personnel to intervene and perform manual confirmation.

[0050] As a preferred solution of the method for matching passenger information with baggage security inspection images based on artificial intelligence described in the present invention, wherein: the behavior action includes the baggage picking and placing action and path trajectory;

[0051] The abnormal situations include: the luggage and the passenger's tracks are separated and the passenger mistakenly picks up other people's luggage.

[0052] As a preferred solution of the method for matching passenger information with baggage security inspection images based on artificial intelligence described in the present invention, the steps of extracting internal features of the baggage through the YOLOv8 model, combining the internal features with the external features of step S1 and the behavior trajectory of step S2, and performing joint analysis are as follows:

[0053] Extract the internal features of luggage based on X-ray images, input X-ray images Extract the internal features of the luggage, the extraction formula is:

[0054]

[0055] in, is the input X-ray image, representing the internal structure of the luggage, is the extracted luggage internal feature vector, including d x eigenvalues, d x is the dimension of the internal features, indicating the number of features extracted from the image, (x i ,y i ) is the coordinate of the i-th feature point on the image, is the center coordinate of all feature points, defined as:

[0056]

[0057] exp(x) = e x is the natural exponential function, σ is the scale parameter of the Gaussian distribution, cos(2πx i / λ) is the frequency domain transformation of the feature point, and λ is the wavelength parameter.

[0058] As a preferred solution of the method for matching passenger information with baggage security inspection images based on artificial intelligence described in the present invention, the step of automatically performing retrospective analysis when an abnormality is detected is:

[0059] The joint eigenvector is defined as U, which is expressed as:

[0060]

[0061] Among them, U is the joint feature vector, which integrates passenger features and luggage features for matching analysis. Attention(·) is the global attention mechanism, which is defined as:

[0062]

[0063] Among them, z represents the input feature, z i is the i-th eigencomponent, exp(z i ) is used to calculate the weight distribution, K is the total number of passengers, α k is the feature weight factor, which indicates the importance of the k-th passenger feature and is defined as:

[0064]

[0065] F k is the feature vector of the kth passenger, L int ,L ext are the internal and external feature vectors representing the luggage respectively;

[0066] Perform matching relationship detection, the detection formula is:

[0067]

[0068] Among them, S match is the matching score, which is used to measure the joint feature vector U and the passenger trajectory T k The similarity of k =[t x,k ,t y,k ] is the trajectory vector of the kth passenger, where

[0069]

[0070] v x (t) is the speed of the passenger in the x direction, v y (t) is the velocity in the y direction,

[0071] |U| is the Euclidean norm of the joint eigenvector, defined as:

[0072]

[0073] Among them, U i is the i-th component of the joint eigenvector.

[0074] The beneficial effects of the present invention are as follows: the present invention arranges a human camera and an object camera at the security check entrance to capture the front and side images of the passenger and extract the facial feature vector to generate a unique identity tag, while recording the external features of the luggage, and binding these features with the passenger identity tag to generate matching information; after the passenger enters the security check channel, the human camera uses gait recognition technology to extract the passenger's dynamic behavior trajectory, adopts a multi-target tracking algorithm to monitor the relative position of the passenger and the luggage in real time, and continuously updates the binding information to cope with cross-scenes; when the luggage passes through the security check equipment, the object camera captures the X-ray image and extracts the internal features of the luggage through the YOLOv8 model, combines the internal features with the external features and the dynamic behavior trajectory, performs joint analysis through a multimodal fusion algorithm, and introduces a global attention mechanism to optimize feature association; when an abnormal situation is detected, a retrospective analysis is automatically triggered, and the ownership relationship is reconfirmed in combination with the historical internal features, external features and behavior trajectories.

[0075] The present invention introduces joint analysis and global attention mechanism, which not only improves the matching efficiency, but also can accurately determine the source of anomalies and respond quickly, significantly enhancing the accuracy and reliability of passenger and luggage matching in dynamic scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0077] Figure 1 The present invention is a flowchart of a method for matching passenger information with baggage security inspection images based on artificial intelligence. DETAILED DESCRIPTION

[0078] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0079] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0080] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0081] Example 1, reference Figure 1 , this embodiment provides a method for matching passenger information with baggage security inspection images based on artificial intelligence, comprising the following steps:

[0082] Step S1, a body camera and an item camera are arranged at an additional location at the security check entrance. The body camera captures the front and side images of the passenger and generates a unique passenger identity tag in combination with face recognition technology; the item camera records the external features of the luggage from a top perspective and binds them to the passenger identity tag to generate binding information;

[0083] External features of luggage include geometric shape, color, and specific markings, including iconic stickers and unusual appearance;

[0084] The steps to generate a unique passenger identity tag using facial recognition technology are:

[0085] Input passenger image set I = {I f ,I s}, where I f Represents the front image of the passenger, I s Represents the profile image of the passenger,

[0086] Extract the facial feature vector, the extraction formula is:

[0087] F=Softmax(W3·tanh(W2·tanh(W1·[I f ,I s ]+b1)+b2)+b3),

[0088] in, is the passenger feature vector, f j is the jth eigenvalue, j is the feature index, tanh(x) is the hyperbolic tangent activation function, W i is the weight matrix of the i-th layer network, b i is the bias vector of the i-th layer network, d f is the dimension of the feature vector,

[0089] Generate a unique ID tag based on the feature vector k , the generating formula is:

[0090]

[0091] Among them, IDk is the identity label of the kth passenger, w j is the weight factor of the eigenvalue, φ(f j )=ln(1+|f j |) is the feature normalization function, j is the feature index, and M is the modular operation base;

[0092] Specifically, passenger characteristics are extracted through a three-layer neural network, and identity tags are generated by combining normalization functions and modular operations to ensure uniqueness and distinctiveness.

[0093] The steps for binding with the passenger identity tag to generate binding information are:

[0094] Input luggage image I L , extract the external feature vector, the extraction formula is:

[0095] L ext =[f shape (I L ),f color (I L ),f mark (I L )],

[0096] in,

[0097]

[0098] N is the number of points of the luggage contour, (x i ,y i ) is the coordinate of the ith point, is the center point of the point set, α i is the weight factor of the point,

[0099]

[0100] c m is the color component mean, is the overall color mean, R, G, B are the red, green, and blue channels,

[0101]

[0102] m i is the grayscale value of the iconic sticker,

[0103] Generate binding matrix B k , B k =ID k L ext ;

[0104] Specifically, through multi-scale feature extraction, the external features of the luggage are obtained and bound with the passenger identity tags to generate a matrix, which establishes a one-to-one correspondence between passengers and luggage.

[0105] Step S2: After the luggage enters the security inspection channel, the human body camera uses gait recognition technology to capture the passenger's body features and dynamic behavior trajectory, and monitors the relative relationship between the passenger and the luggage in real time; the multi-target tracking DeepSORT algorithm is used to track the trajectory of the passenger and the luggage, and the binding information is continuously updated;

[0106] The steps of capturing the passenger's body features and dynamic behavior trajectory through gait recognition technology and monitoring the relative relationship between the passenger and the luggage in real time are as follows:

[0107] Input gait image sequence Among them G t Represents the gait image at time t, extracting the gait feature vector G k , the extraction formula is:

[0108]

[0109] in, is the gait feature, T is the time series length, G t is the gait image, W g ,b g are the network weights and biases,

[0110] Generate dynamic behavior trajectory, the generation formula is:

[0111]

[0112] in,

[0113] are the lateral and longitudinal velocities of the trajectory, t ′ is the time variable;

[0114] Specifically, the dynamic behavior information of passengers is captured through time series feature extraction and trajectory derivation.

[0115] The steps of using the multi-target tracking DeepSORT algorithm to track the trajectory of passengers and luggage and continuously update the binding information are as follows:

[0116] Define the trajectory update model, the model formula is:

[0117]

[0118] in, is the updated trajectory,

[0119] Δ t is the time interval, γ is the correction factor;

[0120] Specifically, the model is updated through trajectory correction to achieve dynamic binding of passengers and luggage.

[0121] Step S3: When the luggage passes through the security inspection equipment, the item camera captures the X-ray image and extracts the internal features of the luggage through the YOLOv8 model. The internal features are combined with the external features of step S1 and the behavior trajectory of step S2 for joint analysis to determine the integrity of the matching relationship between the luggage and the passenger.

[0122] In the process of joint analysis, a global attention mechanism is introduced to optimize feature association;

[0123] During the joint analysis process, when an abnormal situation is detected, a retrospective analysis is automatically performed;

[0124] The backtracking analysis method is:

[0125] Reconfirm the ownership relationship between luggage and passengers by combining historical internal features, external features and behavioral trajectories;

[0126] If the anomaly cannot be resolved during retrospective analysis, an alarm is immediately generated to prompt security personnel to intervene and conduct manual confirmation;

[0127] Behavior actions include luggage pick-up and drop-down actions and path trajectories;

[0128] Abnormal situations include: luggage being separated from passengers and passengers mistakenly picking up other people’s luggage;

[0129] The internal features of the luggage are extracted through the YOLOv8 model. The internal features are combined with the external features of step S1 and the behavior trajectory of step S2. The steps for joint analysis are as follows:

[0130] Extract the internal features of luggage based on X-ray images, input X-ray images Extract the internal features of the luggage, the extraction formula is:

[0131]

[0132] in, is the input X-ray image, representing the internal structure of the luggage, is the extracted luggage internal feature vector, including d x eigenvalues, d x is the dimension of the internal features, indicating the number of features extracted from the image, (x i ,y i ) is the coordinate of the i-th feature point on the image, is the center coordinate of all feature points, defined as:

[0133]

[0134] exp(x) = e x is the natural exponential function, σ is the scale parameter of the Gaussian distribution, cos(2πx i / λ) is the frequency domain transformation of the feature point, and λ is the wavelength parameter;

[0135] Specifically, by combining Gaussian weighting in the spatial domain and cosine transform in the frequency domain, multi-scale features of the internal structure of luggage are extracted, enhancing the ability to detect abnormal items.

[0136] When an abnormality is detected, the steps for automatic backtracking analysis are:

[0137] The joint eigenvector is defined as U, which is expressed as:

[0138]

[0139] Among them, U is the joint feature vector, which integrates passenger features and luggage features for matching analysis. Attention(·) is the global attention mechanism, which is defined as:

[0140]

[0141] Among them, z represents the input feature, z i is the i-th eigencomponent, exp(z i ) is used to calculate the weight distribution, K is the total number of passengers, α k is the feature weight factor, which indicates the importance of the k-th passenger feature and is defined as:

[0142]

[0143] F k is the feature vector of the kth passenger, L int ,L ext are the internal and external feature vectors representing the luggage respectively;

[0144] Perform matching relationship detection, the detection formula is:

[0145]

[0146] Among them, S match is the matching score, which is used to measure the joint feature vector U and the passenger trajectory T k The similarity of k =[t x,k ,t y,k ] is the trajectory vector of the kth passenger, where

[0147]

[0148] v x(t) is the speed of the passenger in the x direction, v y (t) is the velocity in the y direction,

[0149] |U| is the Euclidean norm of the joint eigenvector, defined as:

[0150]

[0151] Among them, U i is the i-th component of the joint eigenvector;

[0152] Specifically, the joint features are calculated through the global attention mechanism, combined with trajectory similarity detection, to accurately match passengers and luggage, and judge abnormal situations based on the matching scores.

[0153] It should be noted that the above embodiments are only used to illustrate the technical solutions 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 solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for matching passenger information with baggage security inspection images based on artificial intelligence, characterized in that: include, Step S1, placing a body camera and an object camera at an additional location at the security check entrance, the body camera captures the front and side images of the passenger, and generates a unique passenger identity tag in combination with face recognition technology; The item camera records the external features of the luggage from a top perspective and binds them to the passenger's identity tag to generate binding information; External features of luggage include geometric shape, color, and specific markings, including iconic stickers and unusual appearance; Step S2: After the luggage enters the security inspection channel, the human body camera uses gait recognition technology to capture the passenger's body features and dynamic behavior trajectory, and monitors the relative relationship between the passenger and the luggage in real time; the multi-target tracking DeepSORT algorithm is used to track the trajectory of the passenger and the luggage, and the binding information is continuously updated; Step S3: When the luggage passes through the security inspection equipment, the item camera captures the X-ray image and extracts the internal features of the luggage through the YOLOv8 model. The internal features are combined with the external features of step S1 and the behavior trajectory of step S2 for joint analysis to determine the completeness of the matching relationship between the luggage and the passenger.

2. The method for matching passenger information with baggage security inspection images based on artificial intelligence as claimed in claim 1, characterized in that: The steps of generating a unique passenger identity tag by combining face recognition technology are: Input passenger image set I = {I f ,I s }, where I f Represents the front image of the passenger, I s Represents the profile image of the passenger, Extract the facial feature vector, the extraction formula is: F=Softmax(W3·tanh(W2·tanh(W1·[I f ,I s ]+b1)+b2)+b3), in, is the passenger feature vector, f j is the jth eigenvalue, j is the feature index, tanh(x) is the hyperbolic tangent activation function, W i is the weight matrix of the i-th layer network, b i is the bias vector of the i-th layer network, d f is the dimension of the feature vector, Generate a unique ID tag based on the feature vector k , the generating formula is: Among them, ID k is the identity label of the kth passenger, w j is the weight factor of the eigenvalue, φ(f j )=ln(1+|f j |) is the feature normalization function, j is the feature index, and M is the modular operation base.

3. The method for matching passenger information with baggage security inspection images based on artificial intelligence as claimed in claim 2, characterized in that: The step of binding with the passenger identity tag to generate binding information is: Input luggage image I L , extract the external feature vector, the extraction formula is: L ext =[f shape (I L ),f color (I L ),f mark (I L )], in, N is the number of points of the luggage contour, (x i ,y i ) is the coordinate of the ith point, is the center point of the point set, α i is the weight factor of the point, c m is the color component mean, is the overall color mean, R, G, B are the red, green, and blue channels, m i is the grayscale value of the iconic sticker, Generate binding matrix B k , B k =ID k L ext .

4. The method for matching passenger information with baggage security inspection images based on artificial intelligence as claimed in claim 3, characterized in that: The steps of capturing the passenger's body features and dynamic behavior trajectory by gait recognition technology and monitoring the relative relationship between the passenger and the luggage in real time are: Input gait image sequence Among them G t Represents the gait image at time t, extracting the gait feature vector G k , the extraction formula is: in, is the gait feature, T is the time series length, G t is the gait image, W g ,b g are the network weights and biases, Generate dynamic behavior trajectory, the generation formula is: in, are the lateral and longitudinal velocities of the trajectory, and t′ is the time variable.

5. The method for matching passenger information with baggage security inspection images based on artificial intelligence as claimed in claim 4, characterized in that: The steps of using the multi-target tracking DeepSORT algorithm to track the trajectories of passengers and luggage and continuously updating the binding information are as follows: Define the trajectory update model, the model formula is: in, is the updated trajectory, Δ t is the time interval, and γ is the correction factor.

6. The method for matching passenger information with baggage security inspection images based on artificial intelligence as claimed in claim 5, characterized in that: During the joint analysis, a global attention mechanism is introduced to optimize feature association; During the joint analysis, when an abnormal situation is detected, a backtracking analysis is automatically performed.

7. The method for matching passenger information with baggage security inspection images based on artificial intelligence as claimed in claim 6, characterized in that: The backtracking analysis method is: Reconfirm the ownership relationship between luggage and passengers by combining historical internal features, external features and behavioral trajectories; If the anomaly cannot be resolved during retrospective analysis, an alarm is immediately generated, prompting security personnel to intervene and perform manual confirmation.

8. The method for matching passenger information with baggage security inspection images based on artificial intelligence as claimed in claim 7, characterized in that: The behavior actions include luggage pick-up and drop-down actions and path trajectories; The abnormal situations include: the luggage and the passenger's tracks are separated and the passenger mistakenly picks up other people's luggage.

9. The method for matching passenger information with baggage security inspection images based on artificial intelligence as claimed in claim 8, characterized in that: The steps of extracting the internal features of the luggage through the YOLOv8 model, combining the internal features with the external features of step S1 and the behavior trajectory of step S2, and performing joint analysis are as follows: Extract the internal features of luggage based on X-ray images, input X-ray images Extract the internal features of the luggage, the extraction formula is: in, is the input X-ray image, representing the internal structure of the luggage, is the extracted luggage internal feature vector, including d x eigenvalues, d x is the dimension of the internal features, indicating the number of features extracted from the image, (x i ,y i ) is the coordinate of the i-th feature point on the image, is the center coordinate of all feature points, defined as: exp(x) = e x is the natural exponential function, σ is the scale parameter of the Gaussian distribution, cos(2πx i / λ) is the frequency domain transformation of the feature point, and λ is the wavelength parameter.

10. The method for matching passenger information with baggage security inspection images based on artificial intelligence as claimed in claim 9, characterized in that: When an abnormal situation is detected, the steps of automatically performing a backtracking analysis are: The joint eigenvector is defined as U, which is expressed as: Among them, U is the joint feature vector, which integrates passenger features and luggage features for matching analysis, and Attention(·) is the global attention mechanism, which is defined as: Among them, z represents the input feature, z i is the i-th eigencomponent, exp(z i ) is used to calculate the weight distribution, K is the total number of passengers, α k is the feature weight factor, which indicates the importance of the k-th passenger feature and is defined as: F k is the feature vector of the kth passenger, L int ,L ext are the internal and external feature vectors representing the luggage respectively; Perform matching relationship detection, the detection formula is: Among them, S match is the matching score, which is used to measure the joint feature vector U and the passenger trajectory T k The similarity of k =[t x,k ,t y,k ] is the trajectory vector of the kth passenger, where v x (t) is the speed of the passenger in the x direction, v y (t) is the velocity in the y direction, |U| is the Euclidean norm of the joint eigenvector, defined as: Among them, U i is the i-th component of the joint eigenvector.

Images