A precise airport monitoring system based on face recognition

Abstract

The present invention provides a precise airport monitoring system based on facial recognition, comprising: a monitoring node module for deploying monitoring nodes within a preset control range of the airport; a monitoring image module for collecting information from the monitoring nodes based on a preset Internet of Things (IoT) and determining a monitoring image; a recognition result module for identifying facial features in the monitoring image and determining a recognition result, wherein the recognition result includes an abnormal recognition result, a questionable recognition result, and a normal recognition result; and a solution module for feeding the recognition result back to a preset control terminal and retrieving a corresponding solution. The present invention monitors complex airport environments in real time, deploys security protection based on primary and secondary priorities, improves the effectiveness of human and material resource investment, and better ensures passenger safety.

Description

Technical Field

[0001] The present invention relates to the fields of face recognition and artificial intelligence technology, and in particular to a lightweight-based Internet of Things terminal identity security authentication method. Background Art

[0002] Current airport security monitoring mainly relies on camera monitoring and security channel inspections, providing the same level of security protection for every passenger. This not only causes security checks to take too long and passengers to complain about the waiting time, but also lacks effective targeting of key monitoring targets, wasting a lot of manpower and material resources. Airports need a new security monitoring mechanism that pays special attention to key groups to improve efficiency. Summary of the Invention

[0003] The present invention provides an airport precision monitoring system based on face recognition, which is used to solve the problems mentioned above in the background technology.

[0004] The present invention provides an airport precision monitoring system based on face recognition, comprising:

[0005] Monitoring node module, used to deploy monitoring nodes within the preset control range of the airport;

[0006] A monitoring image module is used to collect information from the monitoring nodes based on a preset Internet of Things and determine a monitoring image;

[0007] A recognition result module is used to recognize facial features in the monitoring image and determine the recognition result;

[0008] The solution module is used to feed back the recognition result to a preset control terminal and retrieve the corresponding solution.

[0009] As an embodiment of the present technical solution, the monitoring node module includes:

[0010] A division area unit is used to divide the monitoring range of the airport and determine the division areas; wherein the division areas at least include a cross-monitoring area and a separate monitoring area;

[0011] A classification serial number unit is used to classify the divided areas into key monitoring areas and determine corresponding classification serial numbers;

[0012] A segmentation unit, configured to segment the divided areas based on the classification serial number to determine key monitoring areas and non-key monitoring areas;

[0013] The monitoring node unit is used to deploy monitoring nodes based on the key monitoring area and the non-key monitoring area.

[0014] As an embodiment of the present technical solution, the monitoring image module includes:

[0015] A frame image unit is used to identify surveillance video in real time, identify the surveillance video frame by frame, and determine the frame image;

[0016] The image-to-be-processed unit is configured to perform fuzziness screening on the frame images and delete the frame images that fail the screening, and determine the remaining images after deleting the images that fail the screening as images to be processed;

[0017] The monitoring image unit is used to collect image information of the image to be processed and determine that the image to be processed including the face image is a monitoring image.

[0018] As an embodiment of the present technical solution, the recognition result module includes:

[0019] a deduplication image unit, configured to deduplication images containing facial information in the surveillance image to determine a deduplication image;

[0020] An expression feature analysis data unit, configured to perform feature analysis on facial expressions in the screened image to determine expression feature analysis data;

[0021] A recognition result unit, configured to compare and recognize the expression feature analysis data with a preset expression feature analysis preset range to determine a recognition result;

[0022] a normal data unit, configured to determine that the expression feature analysis data is normal data when the recognition result indicates that the expression feature analysis data is within a preset expression feature analysis preset range, and temporarily store the normal data in a preset storage terminal;

[0023] The abnormal data unit is used to determine that the expression feature analysis data is abnormal data when the recognition result is that the expression feature analysis data is not within a preset expression feature analysis preset range, and to retrieve corresponding normal data from a preset storage terminal.

[0024] As an embodiment of the present technical solution, the image screening unit includes:

[0025] A facial feature element unit, configured to identify facial features in the surveillance image and determine facial feature elements;

[0026] a clothing characteristic element unit, configured to perform clothing analysis on images containing facial features in the surveillance image to determine clothing characteristic elements;

[0027] A feature object unit, configured to mark corresponding feature objects in the corresponding surveillance image using the facial feature elements and clothing feature elements;

[0028] The screening image unit is used to determine that the marked monitoring image is a screening image.

[0029] As an embodiment of the present technical solution, the expression feature analysis data unit further includes:

[0030] A feature learning data subunit is used to obtain facial expression samples, perform convolution training on the facial expression samples, and determine feature learning data;

[0031] A sample feature set subunit, configured to extract the facial expression sample through the feature learning data and determine a sample feature set;

[0032] A feature learning factor unit is used to transmit the sample feature set to a preset big data center for analysis and to extract feature learning factors;

[0033] The expression training model subunit is used to transmit the feature learning factors and facial expression samples to a preset neural convolutional network, perform secondary convolution training, and generate a corresponding expression training model;

[0034] The expression feature analysis data subunit is used to transmit the facial expressions in the screened image to the expression training model for feature analysis to determine expression feature analysis data.

[0035] As an embodiment of the present technical solution, the solution module includes:

[0036] an early warning unit, configured to feed back the abnormal recognition result to a preset control terminal and issue an early warning when the recognition result is an abnormal recognition result;

[0037] A solution unit is configured to, when the recognition result is questionable, feed back the abnormal recognition result to a preset control terminal and retrieve a corresponding solution; the solution at least includes image tracking, trajectory recording, and transmission to the control terminal for early warning;

[0038] The deleting unit is used to delete the monitoring image corresponding to the normal recognition result from a preset storage terminal when the recognition result is a normal recognition result.

[0039] As an embodiment of the present technical solution, the solution unit includes:

[0040] A keyword identification information subunit, configured to obtain the identification result and generate corresponding keyword identification information based on the identification result;

[0041] an association table subunit, configured to compare the keyword identification information with a preset standard comparison table and generate a corresponding association table;

[0042] A prediction scheme subunit, configured to transmit the association table and keyword recognition information to a preset training model to generate a corresponding prediction scheme;

[0043] The comparison subunit is used to compare the predicted solution with the historical solutions in the preset solution database, read at least one solution and feed it back to the control terminal.

[0044] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0045] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0047] Figure 1 This is a flow chart of an airport precision monitoring system based on face recognition in an embodiment of the present invention;

[0048] Figure 2 This is a flow chart of an airport precision monitoring system based on face recognition in an embodiment of the present invention;

[0049] Figure 3 The figure is a flow chart of an airport precision monitoring system based on face recognition in an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0051] Example 1:

[0052] According to the attached Figure 1 As shown, the embodiment of the present invention provides 1. A precise airport monitoring system based on face recognition, characterized by comprising:

[0053] Monitoring node module, used to deploy monitoring nodes within the preset control range of the airport;

[0054] A monitoring image module is used to collect information from the monitoring nodes based on a preset Internet of Things and determine a monitoring image;

[0055] A recognition result module is used to recognize facial features in the monitoring image and determine the recognition result;

[0056] The solution module is used to feed back the recognition result to a preset control terminal and retrieve the corresponding solution.

[0057] The working principle and beneficial effects of the above technical solution are:

[0058] In this technical solution, the monitoring node module is used to deploy monitoring nodes within the preset control range of the airport and dynamically monitor the control range at all times; the monitoring image module is used to collect information from the monitoring nodes based on the preset Internet of Things, determine the monitoring image, and effectively ensure the effective collection of monitoring graphics; the recognition result module is used to identify facial features in the monitoring image and determine the recognition results, where the recognition results include abnormal recognition results, questionable recognition results, and normal recognition results. Through hierarchical judgment, a priority sequence is formed for security protection, and the accuracy of human and material resource investment is improved; the solution module is used to feed back the recognition results to the preset control terminal and call the corresponding solution, which not only strengthens the precise control of abnormal recognition results, but also reduces unnecessary time consumption for monitoring normal results, and promotes airport safety construction.

[0059] Example 2:

[0060] According to the attached Figure 2 As shown, in one embodiment, the monitoring node module includes:

[0061] A division area unit is used to divide the monitoring range of the airport and determine the division areas; wherein the division areas at least include a cross-monitoring area and a separate monitoring area;

[0062] A classification serial number unit is used to classify the divided areas into key monitoring areas and determine corresponding classification serial numbers;

[0063] A segmentation unit, configured to segment the divided areas based on the classification serial number to determine key monitoring areas and non-key monitoring areas;

[0064] The monitoring node unit is used to deploy monitoring nodes based on the key monitoring area and the non-key monitoring area.

[0065] The working principle and beneficial effects of the above technical solution are:

[0066] In the present technical solution, the area division unit is used to divide the monitoring range of the airport and determine the divided areas, wherein the divided areas include at least cross-monitoring areas and separated monitoring areas, and the event coverage is expanded by the intersection of monitoring areas; the classification number unit is used to classify the divided areas and determine the corresponding classification numbers; the segmentation unit is used to segment the divided areas based on the classification numbers and determine the key monitoring areas and non-key monitoring areas; the monitoring node unit is used to deploy monitoring nodes based on the key monitoring areas and non-key monitoring areas, and based on the key division of the monitoring areas, form two key and non-key security detection mechanisms, thereby improving the deployment of manpower and material resources for key units.

[0067] Example 3:

[0068] According to the attached Figure 3 As shown, in one embodiment, the monitoring image module includes:

[0069] A frame image unit is used to identify surveillance video in real time, identify the surveillance video frame by frame, and determine the frame image;

[0070] The image-to-be-processed unit is configured to perform fuzziness screening on the frame images and delete the frame images that fail the screening, and determine the remaining images after deleting the images that fail the screening as images to be processed;

[0071] The monitoring image unit is used to collect image information of the image to be processed and determine that the image to be processed including the face image is a monitoring image.

[0072] The working principle and beneficial effects of the above technical solution are:

[0073] In this technical solution, the frame image unit is used to identify surveillance videos in real time, identify surveillance videos frame by frame, determine frame images, and effectively enhance details based on the frame image unit to achieve accurate monitoring of construction; the image unit to be processed is used to screen frame images and delete frame images that fail the screening, compare the retained space for excessive occupancy, and at the same time, determine the image to be processed; the monitoring image unit is used to collect information on the image to be processed, determine the monitoring image, grasp the image features through information collection, and expand the coverage of the monitoring content.

[0074] Example 4:

[0075] In one embodiment, the recognition result module includes:

[0076] a deduplication image unit, configured to deduplication images containing facial information in the surveillance image to determine a deduplication image;

[0077] An expression feature analysis data unit, configured to perform feature analysis on facial expressions in the screened image to determine expression feature analysis data;

[0078] A recognition result unit, configured to compare and recognize the expression feature analysis data with a preset expression feature analysis preset range to determine a recognition result;

[0079] a normal data unit, configured to determine that the expression feature analysis data is normal data when the recognition result indicates that the expression feature analysis data is within a preset expression feature analysis preset range, and temporarily store the normal data in a preset storage terminal;

[0080] The abnormal data unit is used to determine that the expression feature analysis data is abnormal data when the recognition result is that the expression feature analysis data is not within a preset expression feature analysis preset range, and to retrieve corresponding normal data from a preset storage terminal.

[0081] The working principle and beneficial effects of the above technical solution are:

[0082] In this technical solution, the screening image unit is used to screen images containing facial information in the monitoring image and determine the screening image; the expression feature analysis data unit is used to perform feature analysis on the facial expressions in the screening image and determine the expression feature analysis data; the comparison result unit is used to compare the expression feature analysis data with the preset expression feature analysis preset range and determine the comparison result, which saves manpower and provides comprehensive and complete monitoring; the normal data unit is used to determine normal data when the comparison result is that the expression feature analysis data is less than the preset expression feature analysis preset range, and temporarily store the normal data to a preset storage terminal as a regular sample to supplement the database. The abnormal data unit is used to determine abnormal data when the comparison result is that the expression feature analysis data is greater than the preset expression feature analysis preset range, and retrieve the corresponding normal data from the preset storage terminal. By intelligently analyzing and determining the abnormal data, the target image can be quickly grasped and accurate coverage can be achieved.

[0083] Example 5:

[0084] In one embodiment, the image screening unit includes:

[0085] A facial feature element unit, configured to identify facial features in the surveillance image and determine facial feature elements;

[0086] a clothing characteristic element unit, configured to perform clothing analysis on images containing facial features in the surveillance image to determine clothing characteristic elements;

[0087] A feature object unit, configured to mark corresponding feature objects in the corresponding surveillance image using the facial feature elements and clothing feature elements;

[0088] The screening image unit is used to determine that the marked monitoring image is a screening image.

[0089] The working principle and beneficial effects of the above technical solution are:

[0090] In this technical solution, the facial feature element unit is used to identify facial features in the surveillance image and determine the facial feature elements, wherein the facial feature elements include at least eyebrows, eyes, nose and mouth; the clothing feature element unit is used to perform corresponding clothing analysis on the image with facial features in the surveillance image and determine the clothing feature elements, wherein the clothing feature elements include at least clothing color and clothing structure; the feature object unit is used to mark the corresponding feature objects in the corresponding surveillance image through facial feature elements and clothing feature elements, and lock them from multiple dimensions such as facial features and clothing features, so as to effectively improve the timely screening of people with potential safety hazards; the screening image unit is used to determine the screening image through the marked surveillance image, and after screening, the workload of security personnel is greatly reduced.

[0091] Example 6:

[0092] In one embodiment, the facial expression feature analysis data unit further includes:

[0093] A feature learning data subunit is used to obtain facial expression samples, perform convolution training on the facial expression samples, and determine feature learning data;

[0094] A sample feature set subunit, configured to extract the facial expression sample through the feature learning data and determine a sample feature set;

[0095] A feature learning factor unit is used to transmit the sample feature set to a preset big data center for analysis and to extract feature learning factors;

[0096] The expression training model subunit is used to transmit the feature learning factors and facial expression samples to a preset neural convolutional network, perform secondary convolution training, and generate a corresponding expression training model;

[0097] The expression feature analysis data subunit is used to transmit the facial expressions in the screened image to the expression training model for feature analysis to determine expression feature analysis data.

[0098] The working principle and beneficial effects of the above technical solution are:

[0099] In this technical solution, the feature learning data subunit is used to obtain facial expression samples, perform convolution training on the facial expression samples, and determine feature learning data; the sample feature set subunit is used to extract facial expression samples through feature learning data, determine the sample feature set, and generate a sample database by the sample feature set of the passengers, so as to facilitate the induction and extraction of abnormal targets; the feature learning factor unit is used to transmit the sample feature set to the preset big data center for analysis and refine the feature learning factor; the expression training model subunit is used to transmit the feature learning factor and the facial expression sample to the preset neural convolution network for secondary convolution training to generate the corresponding expression training model, promote the construction of smart airport security based on the expression training model, and improve the scientific nature of the monitoring system; the expression feature analysis data subunit is used to transmit the facial expressions in the screened image to the expression training model for feature analysis, determine the expression feature analysis data, and accurately cover multiple element features through feature analysis, strengthen the prominent monitoring of key points, and strengthen security construction.

[0100] Example 7:

[0101] In one embodiment, the solution module includes:

[0102] an early warning unit, configured to feed back the abnormal recognition result to a preset control terminal and issue an early warning when the recognition result is an abnormal recognition result;

[0103] A solution unit is configured to, when the recognition result is questionable, feed back the abnormal recognition result to a preset control terminal and retrieve a corresponding solution; the solution at least includes image tracking, trajectory recording, and transmission to the control terminal for early warning;

[0104] The deleting unit is used to delete the monitoring image corresponding to the normal recognition result from a preset storage terminal when the recognition result is a normal recognition result.

[0105] The working principle and beneficial effects of the above technical solution are:

[0106] In this technical solution, the early warning unit is used to feed back the abnormal recognition result to the preset control terminal when the recognition result is an abnormal recognition result, and to issue an early warning, timely feedback danger information, and mobilize necessary personnel and material deployment; the solution unit is used to feed back the abnormal recognition result to the preset control terminal when the recognition result is a questionable recognition result, and to retrieve the corresponding solution; the deletion unit is used to delete the corresponding monitoring image from the preset storage terminal when the recognition result is a normal recognition result, thereby reducing unnecessary storage space occupancy and improving the timeliness and sensitivity of the monitoring module response.

[0107] Example 8:

[0108] In one embodiment, the solution unit includes:

[0109] A keyword identification information subunit, configured to obtain the identification result and generate corresponding keyword identification information based on the identification result;

[0110] an association table subunit, configured to compare the keyword identification information with a preset standard comparison table and generate a corresponding association table;

[0111] A prediction scheme subunit, configured to transmit the association table and keyword recognition information to a preset training model to generate a corresponding prediction scheme;

[0112] The comparison subunit is used to compare the predicted solution with the historical solutions in the preset solution database, read at least one solution and feed it back to the control terminal.

[0113] The working principle and beneficial effects of the above technical solution are:

[0114] In the present technical solution, the keyword recognition information subunit is used to obtain recognition results, and generate corresponding keyword recognition information based on the recognition results. By identifying key bytes, resource consumption is reduced and precise positioning is achieved; the association table subunit is used to compare the keyword recognition information with a preset standard comparison table and generate a corresponding association table; the prediction scheme subunit is used to transmit the association table and keyword recognition information to a preset training model to generate a corresponding prediction scheme; the comparison subunit is used to compare the prediction scheme with the historical scheme in the preset scheme database, read at least one solution and feed it back to the control terminal, and provide security personnel with a record by timely transmitting the solution, reducing the emergency buffer for emergencies and improving work efficiency.

[0115] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. An airport precision monitoring system based on face recognition, characterized by: include: Monitoring node module, used to deploy monitoring nodes within the airport's preset control range; A monitoring image module is used to collect information from the monitoring nodes based on a preset Internet of Things and determine a monitoring image; A recognition result module is used to recognize facial features in the monitoring image and determine the recognition result; A solution module is used to feed back the recognition result to a preset control terminal and retrieve a corresponding solution; The recognition result module includes: a deduplication image unit, configured to deduplication images containing facial information in the surveillance image to determine a deduplication image; The image screening unit is used to screen images containing facial information in the surveillance images and determine the screening images; An expression feature analysis data unit, configured to perform feature analysis on facial expressions in the screened image to determine expression feature analysis data; A recognition result unit, configured to compare and identify the expression feature analysis data with a preset expression feature analysis preset range to determine a recognition result; a normal data unit, configured to determine that the expression feature analysis data is normal data when the recognition result indicates that the expression feature analysis data is within a preset expression feature analysis preset range, and temporarily store the normal data in a preset storage terminal; an abnormal data unit, configured to, when the recognition result indicates that the expression feature analysis data is not within a preset expression feature analysis preset range, determine that the expression feature analysis data is abnormal data and retrieve corresponding normal data from a preset storage terminal; Wherein, the expression feature analysis data unit further includes: A feature learning data subunit is used to obtain facial expression samples, perform convolution training on the facial expression samples, and determine feature learning data; A sample feature set subunit, configured to extract the facial expression sample through the feature learning data and determine a sample feature set; A feature learning factor unit is used to transmit the sample feature set to a preset big data center for analysis and to extract feature learning factors; The expression training model subunit is used to transmit the feature learning factors and facial expression samples to a preset neural convolutional network, perform secondary convolution training, and generate a corresponding expression training model; The expression feature analysis data subunit is used to transmit the facial expressions in the screened image to the expression training model for feature analysis to determine expression feature analysis data.

2. The airport precision monitoring system based on face recognition according to claim 1, characterized in that: The monitoring node module includes: A division area unit is used to divide the monitoring range of the airport and determine the division areas; wherein the division areas at least include a cross-monitoring area and a separate monitoring area; A classification serial number unit is used to classify the divided areas into key monitoring areas and determine corresponding classification serial numbers; A segmentation unit, configured to segment the divided areas based on the classification serial number to determine key monitoring areas and non-key monitoring areas; The monitoring node unit is used to deploy monitoring nodes based on the key monitoring area and the non-key monitoring area.

3. The airport precision monitoring system based on face recognition according to claim 1, characterized in that: The monitoring image module includes: A frame image unit is used to identify surveillance video in real time, identify the surveillance video frame by frame, and determine the frame image; The image-to-be-processed unit is configured to perform fuzziness screening on the frame images and delete the frame images that fail the screening, and determine the remaining images after deleting the images that fail the screening as images to be processed; The monitoring image unit is used to collect image information of the image to be processed and determine that the image to be processed including the face image is a monitoring image.

4. The airport precision monitoring system based on face recognition according to claim 1, characterized in that: The solution module includes: an early warning unit, configured to feed back the abnormal recognition result to a preset control terminal and issue an early warning when the recognition result is an abnormal recognition result; A solution unit is configured to, when the recognition result is an abnormal recognition result, feed back the abnormal recognition result to a preset control terminal and retrieve a corresponding solution; the solution at least includes image key tracking, trajectory recording, and transmission to the control terminal for early warning; The deleting unit is used to delete the monitoring image corresponding to the normal recognition result from a preset storage terminal when the recognition result is a normal recognition result.

5. The airport precision monitoring system based on face recognition according to claim 4, characterized in that: The solution unit includes: A keyword identification information subunit, configured to obtain the identification result and generate corresponding keyword identification information based on the identification result; an association table subunit, configured to compare the keyword identification information with a preset standard comparison table and generate a corresponding association table; A prediction scheme subunit, configured to transmit the association table and keyword recognition information to a preset training model to generate a corresponding prediction scheme; The comparison subunit is used to compare the predicted solution with the historical solutions in the preset solution database, read at least one solution and feed it back to the control terminal.

Images