Armband wearing detection method and system

Abstract

The invention discloses an armband wearing detection method and system, and relates to the field access qualification detection technology field, and the method comprises the steps: obtaining a to-be-analyzed human body image in real time; performing part segmentation on the human body image according to preset different human body part contours and contour position relations to form a plurality of human body part images; acquiring armband feature information; matching human body part images with features corresponding to the armband feature information in the plurality of human body part images; and if the human body part image with the feature corresponding to the armband feature information is not matched in the plurality of human body part images, generating alarm information, and sending the alarm information to an alarm device. The application has the effect of reducing subjectivity of armband identification and human resource investment at the entrance and exit of the place.

Description

Technical Field

[0001] This application relates to the technical field of site access qualification detection, and particularly to a method and system for detecting armband wearing. Background Art

[0002] In various industries and enterprises in our country, for special construction sites, such as underground construction sites (tunnels, deep foundation pits, etc.), dangerous goods warehouses (explosive tool warehouses, oil storage warehouses, etc.), various machinery workshops, and laboratories, etc., according to the relevant specific norms and regulations of the industry in our country, for the safety of personnel, equipment, and stable production, there are special management regulations for relevant staff to enter and leave the site.

[0003] For example, at the entrance of a construction site or the entrance of a special construction area, a sentry box is set up, and full-time inspectors are set up in the sentry box to check whether the staff entering and leaving the site have the qualification to enter and leave, so as to reduce the irrelevant personnel entering the site, and thus reduce the possibility of production accidents and safety accidents.

[0004] One existing method for checking access qualifications is that inspectors observe whether the entering and leaving personnel wear an armband through the naked eye on site or through a monitoring camera, and this armband is given the qualification to enter and leave by the construction management department.

[0005] However, the above-mentioned armband inspection method has the following disadvantages. On the one hand, the visual inspection by inspectors has certain subjectivity, and it may still cause irrelevant personnel to enter the construction area; on the other hand, since full-time inspectors need to be arranged at the entrance, it increases the investment in human resources, thus increasing the construction cost. Summary of the Invention

[0006] In order to reduce the subjectivity of armband recognition and the investment in human resources at the entrance of the site, this application provides a method and system for detecting armband wearing.

[0007] In the first aspect, a method for detecting armband wearing provided by this application adopts the following technical solution:

[0008] A method for detecting armband wearing, the method includes:

[0009] Obtain a human body image to be analyzed in real time;

[0010] Segment the human body image according to preset different human body part contours and contour position relationships to form multiple human body part images;

[0011] Obtain armband feature information;

[0012] In multiple human body part images, match the human body part images with the corresponding features of the armband feature information;

[0013] If among multiple human body part images, no human body part image with a feature corresponding to the armband feature information is matched, an alarm message is generated and sent to an alarm device.

[0014] Through the above technical solution, a camera device (such as a camera) is used to capture a human body image of a detection object (referring to a person about to enter a construction area), and then the human body image is segmented into multiple human body part images, including arm part images, head portraits, leg portraits, etc.

[0015] Then, armband feature information is obtained, such as special characters, special letters, special patterns, special colors set on the armband, and combinations of the above features, etc. Next, a human body part image with a feature corresponding to the armband feature information is matched in the human body part images.

[0016] If a human body part image with a feature corresponding to the armband feature information is matched, the detection object is released; if no human body part image with a feature corresponding to the armband feature information is matched, an alarm message is generated, and after receiving the alarm message, the alarm device issues an alarm to intercept the detection object.

[0017] Since by applying the system of this method and cooperating with a camera, the input of armband detection personnel in the construction area can be reduced, thus reducing the construction cost. In addition, since the time proportion of detection personnel for armband detection is reduced, the subjectivity of armband detection is reduced, and the possibility of unarmbanded detection personnel entering the construction area is also reduced.

[0018] Therefore, through this method, the subjectivity of armband recognition and the human resource input at the entrance and exit of this place are reduced.

[0019] In a preferred example of this application, after forming multiple human body part images, it further includes:

[0020] Based on the human body part image including the human head, identity matching is performed in a preset database;

[0021] If no corresponding identity information is matched, the armband detection ends.

[0022] Through the above technical solution, in the preset database, the identity information and related features of construction-related personnel are pre-stored. For example, in the human head feature, the face information. If the identity information of the detection object is not matched in the database, it means that the detection object is a person irrelevant to the construction area, and then there is no need to perform subsequent armband detection, and the detection object can be directly intercepted.

[0023] If the identity information of the detected object is matched in the database, which means the detected object is a person related to the construction area, then the subsequent armband detection steps are carried out.

[0024] By preferentially confirming the identity information, the steps of this armband detection method are simplified, the detection speed is accelerated, and the detection time is shortened.

[0025] In a preferred example of the present application, it can be further configured that, among the multiple human body part images, matching the human body part image with the corresponding feature of the armband feature information includes:

[0026] Group the multiple human body part images according to the corresponding body parts on the human body to form multiple different part groups;

[0027] Preferentially match the human body part image with the corresponding feature of the armband feature information in the part group including the human arm;

[0028] If no human body part image with the corresponding feature of the armband feature information is matched in the part group including the arm, then match the human body part image with the corresponding feature of the armband feature information in the remaining part groups.

[0029] Through the above technical solution, for the convenience of wearing and easy identification, generally the armband is worn on the human arm. In this method, in order to accelerate the armband detection speed, preferentially match the human body part image with the corresponding feature of the armband feature information in the part group including the human arm, which can effectively reduce the steps required for feature matching, thereby accelerating the armband detection speed.

[0030] In special cases such as when the arm part of the detected object is blocked by foreign objects, the detected object is a disabled person without an arm, and the detected object does not wear an armband on the arm, etc., the armband may be worn on other parts, such as fixed on the chest or leg, etc., and then match the feature information in the part group other than the arm.

[0031] By dividing the priority of matching the armband feature information for each part group of the human body image, the armband detection speed is effectively accelerated, and the armband detection time is shortened.

[0032] In a preferred example of the present application, it can be further configured that the real-time acquisition of the human body image to be analyzed includes:

[0033] Acquire the head feature and moving speed of the object to be detected, and match the identity information corresponding to the head feature of the object to be detected in the preset database;

[0034] Based on the identity information, obtain the stride information corresponding to the identity information from the database;

[0035] Based on the moving speed of the object to be detected and the corresponding stride information, calculate the stepping frequency of the object to be detected;

[0036] Based on the stepping frequency, calculate the arm-swing frequency of the object to be detected;

[0037] According to the arm-swing frequency, calculate the arm-swing period of the object to be detected;

[0038] Based on the arm-swing period, obtain consecutive body images of the object to be detected.

[0039] Through the above technical solution, after matching the identity information of the object to be detected according to the head characteristics of the object to be detected, then obtain the stride information corresponding to the identity information (here mainly referring to the distance that the human body advances each time it takes a step), and then, based on the moving speed and stride information of the object to be detected that have been obtained, the stepping frequency of the object to be detected can be calculated.

[0040] Since the human body usually has the same arm-swing frequency and stepping frequency in order to maintain balance during movement, therefore, the arm-swing frequency can be directly replaced by the stepping frequency.

[0041] At this time, knowing the arm-swing frequency of the object to be detected, and the arm-swing frequency and arm-swing period are reciprocal to each other, thus the arm-swing period of the object to be detected can be calculated, that is, the time required for one arm-swing. According to the arm-swing period, the camera can continuously capture two body images (including one with the left arm in front and one with the right arm in front), and at this time, the feature information of the armband can be captured in these two body images, thereby effectively reducing the error of armband occlusion and further improving the accuracy of armband detection.

[0042] In a preferred example of the present application, it can be further configured that after matching the identity information corresponding to the head characteristics of the object to be detected in the preset database, it further includes:

[0043] If no identity information corresponding to the head characteristics of the object to be detected is matched in the preset database, then end the armband detection.

[0044] Through the above technical solution, if the identity information of the detection object is not matched in the database, it means that the detection object is a person irrelevant to the construction area, then there is no need to perform subsequent armband detection, and the detection object can be directly intercepted.

[0045] In a preferred example of the present application, it can be further configured that the calculation formula for the stepping frequency of the object to be detected is as follows:

[0046]

[0047] Wherein, v is the moving speed of the object to be detected, and l is the stride of the object to be detected.

[0048] Through the above technical solution, the stepping frequency of the object to be detected can be directly calculated based on the moving speed and stride of the object to be detected.

[0049] In a preferred example of the present application, it can be further configured that before obtaining the stride information corresponding to the identity information from the database based on the identity information, it further includes:

[0050] In the database, multiple preset moving speed intervals and multiple preset stride intervals are entered for each person with access qualifications;

[0051] A mapping relationship is established between multiple said preset moving speed intervals and said preset stride intervals.

[0052] Through the above technical solution, for each person with access qualifications, moving speed data collection and stride data collection are performed in advance, multiple moving speed intervals and stride intervals are established, and a mapping relationship is established between multiple preset moving speed intervals and preset stride intervals, so as to match the corresponding stride information according to different moving speeds, thereby being able to calculate the stepping frequency more accurately, improving the accuracy of the swing period calculation, and thus improving the accuracy of the armband detection result.

[0053] In a preferred example of the present application, it can be further configured that if the magnitude of the moving speed of the object to be detected is equal to the preset value within the corresponding preset moving speed interval, the stepping frequency is calculated according to this moving speed;

[0054] If the magnitude of the moving speed of the object to be detected has no preset value within the corresponding preset moving speed interval, then within the preset moving speed interval, the average value of the two adjacent preset values at both ends of the moving speed is taken to calculate the stepping frequency.

[0055] Since everyone's moving speed is different in different specific situations, through the above technical solution, multiple groups of moving speeds and multiple groups of stride information are entered for each person with access qualifications, so that the corresponding stride information can be matched under different moving speeds of the detection object, thereby improving the accuracy of the armband detection.

[0056] In addition, regardless of the magnitude of the moving speed of the object to be detected and whether there is the same preset value in the database, as long as it is within the preset speed interval, a relatively accurate speed value can be calculated, thereby improving the accuracy of the stepping frequency calculation.

[0057] In the second aspect, based on the above armband wearing detection method, a kind of armband wearing detection system provided by the present application adopts the following technical solution:

[0058] A sleeve badge wearing detection system, comprising:

[0059] An information acquisition module: configured to acquire a human body image to be analyzed in real time;

[0060] An image segmentation module: configured to segment the human body image according to preset different human body part contours and contour position relationships to form multiple human body part images;

[0061] An information acquisition module: acquires sleeve badge feature information;

[0062] A feature matching module: matches the human body part image with the corresponding feature of the sleeve badge feature information in multiple human body part images;

[0063] An alarm information generation module: configured to generate alarm information for wearing a sleeve badge.

[0064] In summary, the present application includes the following beneficial technical effects:

[0065] 1. By applying the system of the present method and cooperating with a camera, the input of sleeve badge detection personnel in the construction area can be reduced, thereby reducing the construction cost. At the same time, the time ratio of the detection personnel for sleeve badge detection is also reduced, thereby reducing the subjectivity of sleeve badge detection, and also reducing the possibility of unbadged detection personnel entering the construction area;

[0066] 2. By dividing the priority levels of sleeve badge feature information matching for each part group of the human body image, the sleeve badge detection speed is effectively increased, and the sleeve badge detection time is shortened;

[0067] 3. By taking multiple human body images according to the human body arm swinging frequency and capturing the sleeve badge feature information in these multiple human body images, the error of arm swinging sleeve badge occlusion is effectively reduced, and thus the accuracy of sleeve badge detection is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 is a schematic flowchart of a sleeve badge wearing detection method provided by an embodiment of the present application.

[0069] Figure 2 is a schematic flowchart of identity information matching in a sleeve badge wearing detection method provided by an embodiment of the present application.

[0070] Figure 3 is a schematic flowchart of sleeve badge feature information matching in a sleeve badge wearing detection method provided by an embodiment of the present application.

[0071] Figure 4 is a schematic flowchart of acquiring multiple human body images in a sleeve badge wearing detection method provided by an embodiment of the present application.

[0072] Figure 5 It is a schematic flowchart of establishing a mapping relationship in a method for detecting armband wearing provided by an embodiment of the present application.

[0073] Figure 6 It is a system block diagram of a system for detecting armband wearing provided by an embodiment of the present application. Detailed implementation manners

[0074] The following further describes the present application in detail Figure 1 - Appendix Figure 6 with reference to the appendix.

[0075] An embodiment of the present application discloses a method for detecting armband wearing, and the method can be applied to a system for detecting armband wearing.

[0076] Referring to the appendix Figure 1 shown, a method for detecting armband wearing includes:

[0077] S101. Obtain a human body image to be analyzed in real time.

[0078] In implementation, a camera device, such as a camera, is installed at the entrance and exit of a construction area to take a picture of an object to be detected about to enter the construction area, forming a human body image including the whole body of the object to be detected.

[0079] S102. Segment the human body image according to preset contour and contour position relationships of different human body parts to form multiple human body part images.

[0080] In implementation, the outer contours of different human body parts are significantly different and easy to distinguish. And once the position of a certain human body part image in the human body image is determined, then according to the position relationship of the specific human body parts, the remaining parts can be quickly determined. For example, after determining the position of the arm, the human body part images including the head and the human body part images including the legs can be immediately determined according to the height where the arm is located, so as to facilitate the subsequent capture of armband features.

[0081] S103. Obtain armband feature information.

[0082] In implementation, the feature information of the armband can be special characters, special letters, special patterns, special colors and combinations of the above features set on the armband, etc., or the light reflection of special materials that are easy to distinguish, etc.

[0083] S104. In the multiple human body part images, match the human body part images with the corresponding features of the armband feature information.

[0084] In implementation, among multiple human body part images, match the human body part images with corresponding features of the armband feature information. For example, if there is a special word "access permit" on the armband, then match the human body part images with the word "access permit" among multiple human body part images. If there are other distinguishing features for the "access permit", then further feature matching can be performed.

[0085] S105. If among multiple human body part images, no human body part image with corresponding features of the armband feature information is matched, generate an alarm message and send the alarm message to the alarm device.

[0086] In implementation, if a human body part image with corresponding features of the armband feature information is matched, release the detection object.

[0087] If no human body part image with corresponding features of the armband feature information is matched, generate an alarm message. After the alarm device receives the alarm message, issue an alarm and intercept the detection object.

[0088] By applying the system of this method and cooperating with a camera, the input of armband detection personnel in the construction area can be reduced, thereby reducing the construction cost. At the same time, the time proportion of detection personnel for armband detection is also reduced, thereby reducing the subjective degree of armband detection, and also reducing the possibility of detection personnel without armbands entering the construction area.

[0089] Refer to Appendix Figure 1 and Appendix Figure 2 As shown, after step S102, the following steps are further included:

[0090] S201. Based on the human body part image including the human head, perform identity matching in a preset database.

[0091] In implementation, in the database, the identity information and related features of construction-related personnel are pre-stored. Based on the human body part image including the human head, such as face feature information, perform feature matching in a preset database.

[0092] S202. If no corresponding identity information is matched, end the armband detection.

[0093] In implementation, if the identity information of the detection object is not matched in the database, it means that the detection object is a person irrelevant to the construction area, and subsequent armband detection is not required. The detection object can be directly intercepted. If the identity information of the detection object is matched in the database, it means that the detection object is a person related to the construction area, and subsequent armband detection steps are performed.

[0094] By preferentially confirming the identity information, the steps of this armband detection method are simplified, the detection speed is accelerated, and the detection time is shortened.

[0095] Refer to the appendix Figure 1 and the appendix Figure 3 As shown, in step S104, the following processing can be performed:

[0096] S301. Group the multiple human body part images according to the corresponding body parts on the human body to form multiple different part groups.

[0097] In implementation, for a certain object to be detected, group the multiple human body part images according to the body parts to form multiple different part groups, including an arm group, a leg group, a head group, etc. For example, the left arm and the right arm of the human body are grouped into the arm group, and the left leg and the right leg are grouped into the leg group.

[0098] S302. Preferentially match the human body part images with the corresponding features of the armband feature information in the part group including the human arm.

[0099] In implementation, generally speaking, for the convenience of identification and wearing, the armband is worn on the human arm. Preferentially matching the human body part images with the corresponding features of the armband feature information in the part group including the human arm (i.e., the arm group) can handle most scenarios, thus accelerating the armband detection speed and shortening the detection time.

[0100] S303. If no human body part image with the corresponding features of the armband feature information is matched in the part group including the arm, then match the human body part images with the corresponding features of the armband feature information in the remaining part groups.

[0101] In implementation, in special cases such as when the arm part of the detection object is blocked, the detection object is a disabled person without an arm, and the detection object's arm is not wearing an armband, the armband may be worn on other parts, such as fixed on the chest or leg, etc. Then match the feature information in the part group other than the arm to make up for the logical gap in the special scenario where the detection object's arm is not wearing an armband, forming a closed loop for armband detection.

[0102] Refer to the appendix Figure 1 and the appendix Figure 4 As shown, in step S101, the following processing can be performed:

[0103] S401. Obtain the head features and moving speed of the object to be detected, and match the identity information corresponding to the head features of the object to be detected in the preset database.

[0104] In implementation, first obtain the head features (such as face features) of the object to be detected through the camera device, and match the identity information that coincides with the head features in the database.

[0105] The moving speed of the object to be detected can be achieved by devices such as a radar speed measurement device or a laser speed measurement device. The speed measurement device only needs to transmit the measured speed information to the system, and the system can obtain the measured speed information.

[0106] S402. Based on the identity information, obtain the stride information corresponding to the identity information from the database.

[0107] In implementation, the stride information of each person with access qualification is pre-entered in the database, including information such as the distance advanced by each person in each step and the distance between the left and right feet.

[0108] S403. Based on the moving speed of the object to be detected and the corresponding stride information, calculate the stride frequency of the object to be detected.

[0109] In implementation, the relevant calculation formulas and processes of the stride frequency will be described in detail in the following text and the relevant drawings, and will not be elaborated here.

[0110] S404. Based on the stride frequency, calculate the arm swing frequency of the object to be detected.

[0111] In implementation, in order to maintain balance during movement, the arm swing frequency and the stride frequency of the human body are usually the same. Therefore, the arm swing frequency can be directly replaced by the stride frequency.

[0112] S405. Based on the arm swing frequency, calculate the arm swing period of the object to be detected.

[0113] In implementation, for one arm swing action of the human body, the arm swing frequency and the arm swing period are reciprocal to each other. That is, the mathematical relationship between the arm swing frequency and the arm swing period is as follows:

[0114]

[0115] Among them, f is the arm swing frequency, and T is the arm swing period. Thus, the arm swing period can be calculated through the arm swing frequency, and the time required for one arm swing can also be calculated.

[0116] S406. Based on the arm swing period, obtain consecutive human body images of the object to be detected.

[0117] In implementation, using the arm swing period as the time interval for continuous photographing, continuously photograph the human body images of the object to be detected, including a human body image with the left arm in front and a human body image with the right arm in front. At this time, the feature information of the armband can be captured in these two human body images, effectively reducing the error of armband occlusion during arm swing, and thus improving the accuracy of armband detection.

[0118] Refer to the appendix Figure 4As shown, after step S401, the following processing can also be performed:

[0119] If the identity information corresponding to the head features of the object to be detected is not matched in the preset database, the armband detection is ended.

[0120] In implementation, if the identity information of the detected object is not matched in the database, it means that the detected object is a person irrelevant to the construction area. Then, there is no need to perform subsequent armband detection, and the detected object can be directly intercepted. If the identity information of the detected object is matched, subsequent armband detection is performed.

[0121] Refer to the appendix Figure 4 As shown, the calculation formula for the stepping frequency in step S403 is as follows:

[0122]

[0123] Among them, v is the moving speed of the object to be detected, and l is the stride of the object to be detected.

[0124] In implementation, based on the above formula, after obtaining the moving speed and stride of the object to be detected, the stepping frequency can be calculated.

[0125] Refer to the appendix Figure 4 and the appendix Figure 5 As shown, before step S402, the following processing can also be performed:

[0126] S501. In the database, enter multiple preset moving speed intervals and multiple preset stride intervals for each person with access qualifications.

[0127] In implementation, for each person with access qualifications, moving speed data collection and stride data collection are performed in advance, and multiple moving speed intervals and multiple stride intervals are entered into the database.

[0128] S502. Establish a mapping relationship between multiple said preset moving speed intervals and said preset stride intervals.

[0129] In implementation, a mapping relationship between multiple preset moving speed intervals and preset stride intervals is established to facilitate matching corresponding stride information according to different moving speeds, so as to calculate the stepping frequency more accurately, improve the accuracy of the swing period calculation, and thus improve the accuracy of the armband detection result.

[0130] Regarding the numerical processing of capturing the corresponding stride information according to the moving speed of the object to be detected, it can be as follows:

[0131] If the magnitude of the moving speed of the object to be detected is equal to the preset value within the corresponding preset moving speed interval, the stepping frequency is calculated according to the moving speed;

[0132] If the magnitude of the moving speed of the object to be detected has no preset value within the corresponding preset moving speed range, then within the preset moving speed range, take the average value of the two adjacent preset values at both ends of the moving speed to calculate the stepping frequency.

[0133] Since everyone's moving speed is different in different specific situations, therefore, in implementation, regardless of the moving speed of the object to be detected and whether there is the same preset value in the database for the moving speed, as long as it is within the preset speed range, a relatively accurate speed value can be calculated, thereby improving the accuracy of the stepping frequency calculation.

[0134] In addition, referring to Appendix Figure 1 and Appendix Figure 6 As shown, the present application also discloses an armband wearing detection system, and the above-mentioned armband wearing detection method is applied to this armband wearing detection system.

[0135] An armband wearing detection system includes:

[0136] An information acquisition module: used to acquire the human body image to be analyzed in real time.

[0137] An image segmentation module: used to segment the human body image according to the preset contour and contour position relationship of different human body parts to form multiple human body part images.

[0138] An information acquisition module: acquires the armband feature information.

[0139] A feature matching module: in multiple human body part images, matches the human body part image with the corresponding feature of the armband feature information.

[0140] An alarm information generation module: used to generate alarm information for not wearing an armband.

[0141] The embodiments of the present specific implementation manner are all preferred embodiments of the present application, and do not limit the protection scope of the present application in turn. Therefore, all equivalent changes made according to the structure, shape, and principle of the present application should be covered within the protection scope of the present application.

Claims

1. A method for detecting armband wearing, characterized in that: The method comprises: Acquire human body images to be analyzed in real time; Segmenting the human body image according to preset contours of different human body parts and contour position relationships to form a plurality of human body part images; Get the armband feature information; Matching a human body part image having features corresponding to the armband feature information among the plurality of human body part images; If no human body part image with features corresponding to the armband feature information is matched among the multiple human body part images, an alarm message is generated and sent to the alarm device.

2. A method for detecting armband wearing according to claim 1, characterized in that: After forming a plurality of human body part images, the method further comprises: Based on the images of human body parts including the human head, identity matching is performed in a preset database; If the corresponding identity information is not matched, the armband detection ends.

3. The armband wearing detection method according to claim 1, characterized in that: The step of matching a human body part image having a feature corresponding to the armband feature information among the plurality of human body part images comprises: Grouping the plurality of human body part images according to corresponding parts of the human body to form a plurality of different part groups; Preferentially matching the human body part image with the features corresponding to the armband feature information in the group of parts including the human body arm; If no human body part image with features corresponding to the armband feature information is matched in the part group including the arm, then human body part images with features corresponding to the armband feature information are matched in the remaining part groups.

4. The armband wearing detection method according to claim 1, characterized in that: The real-time acquisition of the human body image to be analyzed comprises: Acquire the head features and movement speed of the object to be detected, and match the identity information corresponding to the head features of the object to be detected in a preset database; Based on the identity information, obtaining stride information corresponding to the identity information from a database; Calculate the step frequency of the object to be detected based on the moving speed and corresponding stride information of the object to be detected; Based on the stepping frequency, calculating the arm swing frequency of the subject to be detected; Calculating the arm swing period of the object to be detected according to the arm swing frequency; Based on the arm swing cycle, continuous human body images of the object to be detected are acquired.

5. A method for detecting armband wearing according to claim 4, characterized in that: After matching the identity information corresponding to the head features of the object to be detected in the preset database, the method further includes: If the identity information corresponding to the head features of the object to be detected is not matched in the preset database, the armband detection is terminated.

6. A method for detecting armband wearing according to claim 4, characterized in that: The calculation formula of the step frequency of the object to be detected is as follows: Among them, v is the moving speed of the object to be detected, and l is the stride of the object to be detected.

7. A method for detecting armband wearing according to claim 4, characterized in that: Before acquiring the stride information corresponding to the identity information from the database based on the identity information, the method further includes: In the database, multiple preset moving speed intervals and multiple preset stride intervals are entered for each person with entry and exit qualifications; A mapping relationship between the plurality of preset moving speed intervals and the preset stride intervals is established.

8. A method for detecting armband wearing according to claim 7, characterized in that: If the moving speed of the object to be detected is equal to a preset value in the corresponding preset moving speed interval, the stepping frequency is calculated according to the moving speed; If the moving speed of the object to be detected has no preset value in the corresponding preset moving speed interval, then the average of the adjacent preset values ​​of the moving speed at both ends is taken in the preset moving speed interval to calculate the step frequency.

9. A sleeveband wearing detection system, characterized in that: include: Information acquisition module: used to acquire the human body image to be analyzed in real time; Image segmentation module: used for segmenting the human body image according to preset contours of different human body parts and contour position relationships to form multiple human body part images; Information acquisition module: obtains armband feature information; Feature matching module: matching the human body part image with the feature corresponding to the armband feature information among the plurality of human body part images; Alarm information generation module: used to generate alarm information for wearing armbands.

Images