Queue monitoring and guiding system and method based on machine vision
By using machine vision monitoring cameras and specialized algorithms, queue anomalies can be monitored and corrected in real time, solving the problem of queue order relying on manual supervision and improving the intelligence and automation of queue monitoring and guidance.
Patent Information
- Application Number
- CN202411003818.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-25
- Publication Date
- 2024-11-15
AI Technical Summary
In existing technologies, queue order relies on manual supervision, which leads to high pressure on supervisors and is prone to false detections and missed detections, especially in densely populated places and industrial assembly lines where abnormalities such as objects running off course or flying away cannot be effectively monitored.
A machine vision-based queue monitoring and guidance system is adopted, which uses surveillance cameras and dedicated visual inspection algorithms to monitor the queue status in real time, and guides the queue through digital display screens, projection, and speaker devices, and automatically corrects abnormalities.
It enables real-time monitoring and automated correction of queue status, reduces manual intervention, improves the intelligence level of queue monitoring and guidance, and avoids safety accidents.
Smart Images

Figure CN118968405A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of safe operation and maintenance technology, and in particular to a queue monitoring and guiding system and method based on machine vision. Background Art
[0002] Queuing is a common phenomenon in daily life, especially in crowded places such as supermarkets and stations. Currently, queue order relies on manual work, and correcting queue distortion and stopping queue jumping bring huge pressure to supervisors.
[0003] In the industrial field, objects such as finished industrial products and ores that are transported in queues on the assembly line may sometimes deviate or fly away, and currently they are mainly supervised by artificial vision. After working for a long time, human eyes will inevitably become fatigued, resulting in false detection and missed detection. It can be seen that queue status monitoring is of great significance to the field of safe operation and maintenance. However, a search of existing technical information (literature, patents, etc.) shows that there is no relevant technology for queue monitoring and guidance. Summary of the invention
[0004] The purpose of the present invention is to provide a queue monitoring and guidance system and method based on machine vision, which uses a monitoring camera as a video acquisition device and designs a special visual detection algorithm to replace artificial vision to achieve real-time monitoring of the queue status. By analyzing the image characteristics of queue abnormalities and formulating scientific and reasonable guidance strategies, corrective measures can be taken in time when abnormalities occur to solve the problems raised in the above background technology.
[0005] To achieve the above object, the present invention provides the following technical solutions:
[0006] A queue monitoring and guiding system based on machine vision comprises a monitoring camera, a queue monitoring module, a queue guiding module, a digital display screen, a projection device and a speaker device; the monitoring camera is installed above and in front of the queue; the queue monitoring module is connected to the monitoring camera; the queue guiding module is connected to the queue monitoring module; the digital display screen is connected to both the queue monitoring module and the queue guiding module; the projection device and the speaker device are connected to the queue guiding module; the speaker device is used to play reminder information.
[0007] Furthermore, the models, quantities and installation methods of the monitoring cameras, digital display screens, projection devices and speaker devices are adjusted according to the actual needs of the queue monitoring site.
[0008] Furthermore, the surveillance camera is used to shoot the marching queue; the queue monitoring module has a built-in queue monitoring algorithm for monitoring the queue status, and displaying the queue status through a digital display screen; the queue guiding module has a built-in queue guiding algorithm for generating a queue guiding method according to the current queue status, and guiding the queue through a projection device and a speaker device.
[0009] The present invention provides another technical solution: a queue monitoring and guiding method based on machine vision, comprising the following steps:
[0010] S1: The monitoring camera shoots the queue video from the front and top of the queue, and the queue monitoring module calls the queue monitoring algorithm to detect the number and position of the targets in the queue, and displays them in real time on the digital display screen;
[0011] S2: The queue monitoring module calls the queue monitoring algorithm to generate theoretical order boundaries l1 and l2 on the left and right sides of the queue according to the image position and size of the first target in the queue, monitors whether the queue is distorted, and monitors whether there is queue jumping or queue dropping. If there is an abnormality in the queue, go to S3;
[0012] S3: The queue guidance module calls the queue guidance algorithm to generate a queue guidance method according to the current queue status; and projects guidance information in the area where the queue is abnormal through the projection device, and broadcasts the guidance information through the loudspeaker device, and updates the digital display screen information according to the situation.
[0013] Furthermore, the queue position in S1 includes the image coordinates of the target and the order number of the target in the queue, and the number of targets and position information are displayed in a scrolling manner on a digital display screen.
[0014] Furthermore, the specific monitoring process in S2 is as follows:
[0015] S201: The queue monitoring module calculates the image X coordinate F of the center of the first target F in the queue Q. x Generate the theoretical center axis M of the queue, that is, let M x =F x , so that the image X coordinate M of the theoretical central axis M x Equal to F x ;
[0016] S202: Taking the target detection box width W1 and M of the first target x As the reference, along the camera field of view, (W1 / 2)·τ n-1 After the central axis M is widened to the left and right in sections n times with equal intervals, n endpoints are generated on the left and right sides of M respectively, and the endpoints are connected to generate the theoretical order boundaries l1 and l2 corresponding to the left and right sides of the queue; where τ is the common ratio;
[0017] S203: Traverse the X coordinates of all targets in the queue Draw k straight lines passing through the X coordinates of each target and parallel to the X axis of the image, calculate the intersection points of each straight line with l1 and l2, and generate a set of X coordinates of the intersection points According to the following formula:
[0018]
[0019] In the formula, Represents the center X coordinate of the mth target in queue Q; and Respectively represent the And the X coordinate of the intersection of the straight line parallel to the X axis of the image and l1 and l2, if the flag T1 is 1, it means that the queue has a distortion exception, and go to S3; otherwise, there is no distortion exception.
[0020] Furthermore, in S203, two monitoring areas R1 and R2 are set on the left and right sides of l1 and l2 respectively. The queue monitoring algorithm saved by the queue monitoring module continuously monitors the target and generates its motion trajectory. If the target motion trajectory is: leaving the queue Q after passing through R1 or R2, it means that there is a queue-out abnormality; if the target motion trajectory is: entering the queue Q after passing through R1 or R2, and the y coordinate of the target center is greater than the y coordinate of the last target in the original queue, it means that there is a queue-jumping abnormality, and go to S3; otherwise, it is a normal queuing behavior, and the digital display screen information is updated.
[0021] Furthermore, in S3, if T1=1, the queue guidance module calls the queue guidance algorithm according to the X-axis coordinate of the target causing the distortion Making a decision: If The projection device is triggered to project a guide arrow near the target, the guide arrow points to the target horizontally from the left side of the target, and the speaker device is triggered to play a reminder to merge right; if The guide arrow points to the target from the right side and points horizontally to the target, and at the same time triggers the speaker device to play a reminder to merge to the left; if there is a phenomenon of falling out of line, the queue guidance module triggers the speaker device to play a reminder for the entire queue to move forward and updates the digital display screen information; if there is a phenomenon of queue jumping, the queue guidance module triggers the projection device to project an exclamation mark sign near the target, and at the same time triggers the speaker device to play a reminder for queue jumping.
[0022] Compared with the prior art, the present invention has the following beneficial effects:
[0023] The queue monitoring and guidance system and method based on machine vision of the present invention uses a monitoring camera as a video acquisition device. By designing a special visual detection algorithm, it can replace artificial vision, realize real-time monitoring of the queue status, and formulate a scientific and reasonable guidance strategy by analyzing the image characteristics of queue abnormalities. Corrective measures can be taken in time when abnormalities occur, thereby improving the automation and intelligence of queue monitoring and guidance, reducing labor costs on the basis of replacing artificial vision, effectively filling the gap in the field of queue safety operation and maintenance, and avoiding safety accidents caused by abnormal queue status. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] Figure 1 It is a system framework diagram of the present invention;
[0025] Figure 2 It is a logic diagram of the implementation method of the present invention;
[0026] Figure 3 It is a schematic diagram of monitoring area line calculation of the present invention.
[0027] In the figure: 1. Monitoring camera; 2. Queue monitoring module; 3. Queue guiding module; 4. Digital display screen; 5. Projection device; 6. Speaker device. DETAILED DESCRIPTION
[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0029] See also Figure 1In an embodiment of the present invention, a queue monitoring and guiding system based on machine vision is provided, comprising a monitoring camera 1, a queue monitoring module 2, a queue guiding module 3, a digital display screen 4, a projection device 5 and a speaker device 6; the monitoring camera 1 is installed above and in front of the queue; the queue monitoring module 2 is connected to the monitoring camera 1; the queue guiding module 3 is connected to the queue monitoring module 2; the digital display screen 4 is simultaneously connected to the queue monitoring module 2 and the queue guiding module 3; the projection device 5 and the speaker device 6 are connected to the queue guiding module 3; the speaker device 6 is used to play reminder information; wherein, the model, quantity, installation method and other parameters of the monitoring camera 1, the digital display screen 4, the projection device 5 and the speaker device 6 can be adjusted according to the actual needs of the queue monitoring site; specifically, the monitoring camera 1 is used to shoot the moving queue; the queue monitoring module 2 has a built-in queue monitoring algorithm for monitoring the queue status, and displays the queue status through the digital display screen 4; the queue guiding module 3 has a built-in queue guiding algorithm for generating a queue guiding method according to the current queue status, and guiding the queue through the projection device 5 and the speaker device 6.
[0030] like Figure 2 As shown, based on the above system, in order to further better explain the embodiment of the present invention, a queue monitoring and guiding method based on machine vision is also provided, comprising the following steps:
[0031] S1: The monitoring camera 1 shoots a queue video from the front and top of the queue, and sends the video to the queue monitoring module 2. The queue monitoring module 2 calls the queue monitoring algorithm to detect the number and position of targets contained in the queue, and displays them in real time through the digital display screen 4; wherein the queue position information includes the image coordinates of each target and its sequence number in the queue, and the number and position information of the targets are displayed in a scrolling manner through the digital display screen 4.
[0032] S2: Queue monitoring module 2 calls the queue monitoring algorithm to generate theoretical order boundaries l1 and l2 on the left and right sides of the queue according to the image position and size of the first target in the queue, monitors whether the queue is distorted, and monitors whether there is queue cutting or queue dropping. If there is an abnormality in the queue, it goes to S3. In specific implementation, the monitoring process is as follows:
[0033] S201: The queue monitoring module 2 detects the image X coordinate F of the center of the first target F in the queue Q. x Generate the theoretical center axis M of the queue, that is, let M x =F x , so that the image X coordinate M of the theoretical central axis M x Equal to F x ;
[0034] S202: Taking the target detection box width W1 and M of the first target xAs the reference, along the camera field of view, (W1 / 2)·τ n-1 After the central axis M is widened to the left and right in sections n times with equal intervals, n endpoints are generated on the left and right sides of M respectively, and the endpoints are connected to generate the theoretical order boundaries l1 and l2 corresponding to the left and right sides of the queue; where τ is the common ratio;
[0035] S203: Traverse the X coordinates of all targets in the queue Draw k straight lines passing through the X coordinates of each target and parallel to the X axis of the image, calculate the intersection points of each straight line with l1 and l2, and generate a set of X coordinates of the intersection points According to the following formula:
[0036]
[0037] In the formula, Represents the center X coordinate of the mth target in queue Q; and Respectively represent the And the X coordinate of the intersection of the straight line parallel to the X axis of the image and l1 and l2, if the flag T1 is 1, it means that the queue has a distortion exception, and go to S3; otherwise, there is no distortion exception.
[0038] like Figure 3 As shown, in S203, two monitoring areas R1 and R2 are respectively set on the left and right sides of l1 and l2. The queue monitoring algorithm saved by the queue monitoring module 2 continuously monitors the target and generates its motion trajectory. If the target motion trajectory is: leaving the queue Q after passing through R1 or R2, it means that there is a queue-out abnormality; if the target motion trajectory is: entering the queue Q after passing through R1 or R2, and the y coordinate of the target center is greater than the y coordinate of the last target in the original queue, it means that there is a queue-jumping abnormality, and go to S3; otherwise, it is a normal queuing behavior, and the information on the digital display screen 4 is updated.
[0039] S3: The queue guidance module 3 calls the queue guidance algorithm to generate a queue guidance method according to the current queue state; and projects guidance information in the abnormal area of the queue through the projection device 5, and broadcasts the guidance information through the speaker device 6, and updates the information on the digital display screen 4 according to the situation; in specific implementation, if T1=1, the queue guidance module 3 calls the queue guidance algorithm, according to the X-axis coordinate of the target causing the distortion Making a decision: If The projection device 5 is triggered to project a guide arrow near the target, the guide arrow points to the target horizontally from the left side of the target, and the speaker device 6 is triggered to play a reminder to merge right; if The guiding arrow points to the target from the right side and points horizontally to the target, and at the same time triggers the speaker device 6 to play a reminder to merge to the left; if there is a phenomenon of falling out of line, the queue guidance module 3 triggers the speaker device 6 to play a reminder for the entire queue to move forward, and updates the information on the digital display screen 4; if there is a phenomenon of queue jumping, the queue guidance module 3 triggers the projection device 5 to project an exclamation mark near the target, and at the same time triggers the speaker device 6 to play a reminder for queue jumping.
[0040] In summary: the queue monitoring and guidance system and method based on machine vision provided by the present invention uses a monitoring camera 1 as a video acquisition device. By designing a special visual detection algorithm, it can replace artificial vision and realize real-time monitoring of the queue status. By analyzing the image characteristics of queue abnormalities, a scientific and reasonable guidance strategy can be formulated. Corrective measures can be taken in time when an abnormality occurs, thereby improving the automation and intelligence of queue monitoring and guidance, and reducing labor costs on the basis of replacing artificial vision.
[0041] It should be noted that the above embodiments of the present invention are merely examples for the purpose of clearly illustrating the contents of the present invention, and are not limitations on the implementation methods of the present invention. For example: the queue covers all objects that can form a queue; the surveillance camera is not limited to a visible light camera, but also covers all instruments that can form images. The projection device covers all devices that can form a warning sign in three-dimensional space, and the speaker device covers all reminder devices that can transmit in three-dimensional space that can be perceived and understood by humans or devices. Obvious changes in systems, methods, functions and applications made by various technical personnel in this field based on the above embodiments are still within the scope of protection of the present invention.
Claims
1. A queue monitoring and guiding system based on machine vision, characterized in that: The invention comprises a monitoring camera (1), a queue monitoring module (2), a queue guiding module (3), a digital display screen (4), a projection device (5) and a speaker device (6); the monitoring camera (1) is installed above and in front of the queue; the queue monitoring module (2) is connected to the monitoring camera (1); the queue guiding module (3) is connected to the queue monitoring module (2); the digital display screen (4) is simultaneously connected to the queue monitoring module (2) and the queue guiding module (3); the projection device (5) and the speaker device (6) are connected to the queue guiding module (3); and the speaker device (6) is used for playing reminder information.
2. A queue monitoring and guiding system based on machine vision as claimed in claim 1, characterized in that: The model, quantity and installation method of the monitoring camera (1), the digital display screen (4), the projection device (5) and the speaker device (6) are adjusted according to the actual needs of the queue monitoring site.
3. A queue monitoring and guiding system based on machine vision as claimed in claim 2, characterized in that: The monitoring camera (1) is used to photograph a moving queue; the queue monitoring module (2) has a built-in queue monitoring algorithm for monitoring the queue state, and displays the queue state through a digital display screen (4); the queue guiding module (3) has a built-in queue guiding algorithm for generating a queue guiding method according to the current queue state, and guiding the queue through a projection device (5) and a speaker device (6).
4. A queue monitoring and guiding method based on machine vision, implemented based on a queue monitoring and guiding system based on machine vision as claimed in claim 3, characterized in that: The following steps are involved: S1: The monitoring camera (1) takes a video of the queue from the front and top of the queue, and the queue monitoring module (2) calls the queue monitoring algorithm to detect the number and position of the targets in the queue, and displays them in real time on the digital display screen (4); S2: The queue monitoring module (2) calls the queue monitoring algorithm to generate theoretical order boundaries l1 and l2 on the left and right sides of the queue according to the image position and size of the first target in the queue, monitors whether the queue is distorted, and monitors whether there is queue jumping or queue dropping. If there is an abnormality in the queue, go to S3; S3: The queue guidance module (3) calls the queue guidance algorithm to generate a queue guidance method according to the current queue state; and projects guidance information in the area where the queue is abnormal through the projection device (5), and broadcasts the guidance information through the loudspeaker device (6), and updates the information on the digital display screen (4) according to the situation.
5. A queue monitoring and guiding method based on machine vision as claimed in claim 4, characterized in that: The queue position in S1 includes the image coordinates of the target and the sequence number of the target in the queue, and the number of targets and position information are displayed in a scrolling manner on the digital display screen (4).
6. A queue monitoring and guiding method based on machine vision as claimed in claim 4, characterized in that: The specific monitoring process in S2 is as follows: S201: Queue monitoring module (2) calculates the image X coordinate F of the center of the first target F in queue Q. x Generate the theoretical center axis M of the queue, that is, let M x =F x , so that the image X coordinate M of the theoretical central axis M x Equal to F x ; S202: Taking the target detection box width W1 and M of the first target x As the reference, along the camera field of view, (W1 / 2)·τ n-1 After the central axis M is widened to the left and right in sections n times with equal intervals, n endpoints are generated on the left and right sides of M respectively, and the endpoints are connected to generate the theoretical order boundaries l1 and l2 corresponding to the left and right sides of the queue; where τ is the common ratio; S203: Traverse the X coordinates of all targets in the queue Draw k straight lines passing through the X coordinates of each target and parallel to the X axis of the image, calculate the intersection points of each straight line with l1 and l2, and generate a set of X coordinates of the intersection points According to the following formula: In the formula, Represents the center X coordinate of the mth target in queue Q; and Respectively represent the And the X coordinate of the intersection of the straight line parallel to the X axis of the image and l1 and l2, if the flag T1 is 1, it means that the queue has a distortion exception, and go to S3; otherwise, there is no distortion exception.
7. A queue monitoring and guiding method based on machine vision as claimed in claim 6, characterized in that: In S203, two monitoring areas R1 and R2 are respectively set on the left and right sides of l1 and l2. The queue monitoring algorithm saved by the queue monitoring module (2) continuously monitors the target and generates its motion trajectory. If the target motion trajectory is: leaving the queue Q after passing R1 or R2, it indicates that there is a queue-out anomaly; if the target motion trajectory is: entering the queue Q after passing R1 or R2, and the y coordinate of the target center is greater than the y coordinate of the last target in the original queue, it indicates that there is a queue-jumping anomaly, and go to S3. Otherwise, it is normal queuing behavior and the information on the digital display screen (4) is updated.
8. A queue monitoring and guiding method based on machine vision as claimed in claim 7, characterized in that: In S3, if T1=1, the queue guidance module (3) calls the queue guidance algorithm according to the X-axis coordinate of the target causing the distortion Making a decision: If The projection device (5) is triggered to project a guide arrow near the target, the guide arrow points to the target horizontally from the left side of the target, and the speaker device (6) is triggered to play a reminder to merge rightwards; if The guiding arrow points to the target from the right side and horizontally to the target, and at the same time triggers the speaker device (6) to play a reminder to merge to the left; if there is a queue dropout, the queue guidance module (3) triggers the speaker device (6) to play a reminder for the entire queue to move forward, and updates the information on the digital display screen (4); if there is a queue jump, the queue guidance module (3) triggers the projection device (5) to project an exclamation mark near the target, and at the same time triggers the speaker device (6) to play a queue jump reminder.
