Method, device and system for identifying that a desktop is not cleaned in time

Abstract

The present application relates to the field of computer vision technology, and in particular to a method, device and system for identifying a desktop that has not been cleaned in time, the method comprising: acquiring an image of a target environment area taken at intervals; wherein the image includes at least one detection area, the detection area includes at least one desktop, and the desktop is used to place dining items; identifying whether there are dining item targets and customer targets in the detection area; identifying the position of the dining item target in the detection area; if within a preset duration, the customer target is not identified in the detection area, the dining item target is identified, and the position movement of the dining item target is less than a preset threshold, it is determined that the desktop has not been cleaned in time. This method can accurately and efficiently identify that the desktop has not been cleaned in time, so that the desktop can be cleaned in time after the meal.

Description

Technical Field

[0001] The present application relates to the field of computer vision technology, and in particular to a method, device and system for identifying a desktop that has not been cleaned in time. Background Art

[0002] In the on-site management needs of catering stores, there are requirements for the dining environment of customers. One of them is to clean the tableware and dining garbage in time after the user finishes the meal. At present, the on-site inspection of unclean tables can only be achieved by people watching the video, which has the problems of increased costs, increased management difficulty and low efficiency. Summary of the invention

[0003] Based on this, it is necessary to provide a method, device, computer equipment and storage medium that can accurately and efficiently identify that the desktop is not cleaned in time in response to the above technical problems.

[0004] In a first aspect, the present application provides a method for identifying a tabletop that has not been cleaned in a timely manner, the method comprising: acquiring an image of a target environment area taken at intervals; wherein the image includes at least one detection area, the detection area includes at least one tabletop, and the tabletop is used to place dining items; identifying whether there are dining item targets and customer targets in the detection area; identifying the position of the dining item target in the detection area; if within a preset duration, no customer target is identified in the detection area, and a dining item target is identified, and the position movement of the dining item target is less than a preset threshold, it is determined that there is a tabletop that has not been cleaned in a timely manner.

[0005] In one embodiment, acquiring images of the target environment area taken at intervals includes: acquiring N1 images of the target environment area taken at intervals; if within a preset duration, no customer target is identified in the detection area, and a dining item target is identified, and the position movement of the dining item target is less than a preset threshold, determining that a desktop has not been cleaned in time includes: within a preset duration, if for N2 consecutive images within the preset duration, no customer target is identified in the detection area, and a dining item target is identified, and the position movement of the dining item target is less than a preset threshold, determining that a desktop has not been cleaned in time; wherein the N2 images are partial images of the N1 images.

[0006] In one of the embodiments, identifying that the position movement of a dining item target in a detection area in N2 consecutive images is less than a preset threshold includes: when no customer target is identified in the detection area and the presence of a dining item target is identified in N3 consecutive images, obtaining the coordinates of the dining item target in the detection area of ​​the first image in N3 images as the original coordinates; wherein the N3 images are partial images in the N1 images; obtaining the coordinates of the dining item target in the detection area in multiple consecutive images, taking the coordinates with the largest change value relative to the original coordinates as the end point coordinates, and determining the movement value of the dining item target based on the original coordinates and the end point coordinates; until the movement value in N2 consecutive images is less than the preset threshold, N3 is greater than or equal to N2, and the N2 images are partial images in the N3 images.

[0007] In one embodiment, the coordinate with the largest change value relative to the original coordinate is taken as the end point coordinate, and the movement value of the dining item target is determined based on the original coordinate and the end point coordinate, including: the coordinate of the dining item target includes each vertex of the identification frame of the dining item target, in the identification frame of the dining item target, the original coordinate of each vertex corresponds to the position of the end point coordinate one by one, and the vertex coordinate with the largest difference between the original coordinate and the end point coordinate among each vertex is taken as the movement value of the dining item target.

[0008] In one embodiment, the method for identifying that a desktop has not been cleaned in time also includes: if it is identified within the duration that there is a customer target in the detection area who is not facing the table in the detection area, the duration or the number of consecutive images is not reset.

[0009] In one embodiment, a method for identifying that a customer target is not facing toward a table within the detection area includes: determining whether the customer target is facing toward a camera; determining the relative positions of the camera, the customer target, and the table; and determining the orientation of the customer target relative to the table within the detection area based on the orientation of the customer target relative to the camera and the relative positions of the customer target and the table.

[0010] In one embodiment, the orientation of the customer target relative to the table in the detection area is determined based on the orientation of the customer target relative to the camera and the relative positions of the customer target and the table, including: if the customer target is toward the camera and the customer target is closer to the camera relative to the table, determining that the customer target is not toward the table in the detection area; and / or, if the customer target is toward the camera and the customer target is farther away from the camera relative to the table, determining that the customer target is toward the table in the detection area; and / or, if the customer target is away from the camera and the customer target is farther away from the camera relative to the table, determining that the customer target is not toward the table in the detection area; and / or, if the customer target is away from the camera and the customer target is closer to the camera relative to the table, determining that the customer target is toward the table in the detection area.

[0011] In one embodiment, the method for identifying that a desktop is not cleaned in time also includes: acquiring multiple images of the target environment area taken at intervals within a first preset time period, wherein the first preset time period is set long enough to ensure that the employee will definitely clean the desktop within the first preset time period; identifying a first object target on the desktop that has not moved within the first preset time period; and excluding the first object target when identifying dining items in the detection area.

[0012] In one of the embodiments, the method for identifying that a desktop is not cleaned in time also includes: acquiring multiple images of the target environment area taken at intervals within a second preset time period, wherein the second preset time period is set long enough to ensure that the employee will definitely clean the desktop within the second preset time period; among the multiple images, if a second object target is identified on the desktop in the detection area of ​​images greater than a preset number, the second object target is excluded when identifying dining items in the detection area; wherein the second object target can be moved or separated on the desktop.

[0013] In a second aspect, the present application also provides a device for identifying that a desktop has not been cleaned in time, the device comprising: an acquisition module, for acquiring images of a target environment area taken at intervals; wherein the image includes at least one detection area, the detection area includes at least one desktop, and the desktop is used to place dining items; an identification module, for identifying whether there are dining item targets and customer targets in the detection area; and for identifying the position of the dining item target in the detection area; a control module, if within a preset duration, no customer target is identified in the detection area, and the presence of a dining item target is identified, and the position movement of the dining item target is less than a preset threshold, it is determined that there is a desktop that has not been cleaned in time.

[0014] On the third aspect, the present application also provides a system for identifying that a desktop is not cleaned in time, the system comprising: an image acquisition device, used to take images of a target environment area at intervals; wherein the image includes at least one detection area, the detection area includes at least one desktop, and the desktop is used to place dining items; a server, used to identify whether there are dining item targets and customer targets in the detection area; and also used to identify the position of the dining item target in the detection area; if within a preset duration, no customer target is identified in the detection area, and no dining item target is identified, and the position movement of the dining item target is less than a preset threshold, it is determined that there is a desktop that is not cleaned in time.

[0015] In a fourth aspect, the present application further provides a computer device, wherein the computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of any one of the methods described in the first aspect are implemented.

[0016] In a fifth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any one of the methods described in the first aspect are implemented.

[0017] The above-mentioned method, device, system, computer equipment and storage medium for identifying a tabletop that has not been cleaned in time, obtains an image of the target environment area taken at intervals; wherein the image includes at least one detection area, the detection area includes at least one tabletop, and the tabletop is used to place dining items; further identifies whether there are dining item targets and customer targets in the detection area and the positions of the dining item targets; if within a preset duration, no customer target is identified in the detection area, and a dining item target is identified, and the position movement of the dining item target is less than a preset threshold, that is, there is a tabletop where the dining customer has left but there are still dining items on the tabletop, so it is determined that the tabletop has not been cleaned in time. Compared with manual inspections or manual viewing of video recordings in related technologies to determine whether there is a tabletop to be cleaned, the present application uses images of the target environment area taken at intervals to further determine whether the identified images meet certain conditions within a preset duration, so as to accurately and efficiently determine that the tabletop has not been cleaned in time, thereby ensuring that the customer's tabletop can be cleaned in time after the meal, thereby improving the management efficiency of restaurant tabletop cleaning and reducing management difficulty and management costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A diagram of an application environment of a method for identifying a desktop that is not cleaned in time in one embodiment;

[0019] Figure 2 A schematic diagram of a first process of a method for identifying a desktop that has not been cleaned in time in one embodiment;

[0020] Figure 3a is a schematic diagram of a detection area in one embodiment;

[0021] Figure 3b is a schematic diagram of a detection area in another embodiment;

[0022] Figure 4 A second flow chart of a method for identifying a desktop that has not been cleaned in time in one embodiment;

[0023] Figure 5 is a schematic diagram of N1 images of a target environment area in one embodiment;

[0024] Figure 6 A schematic diagram of obtaining a movement value of a dining item target in one embodiment;

[0025] Figure 7 A third flow chart of a method for identifying a desktop that has not been cleaned in time in one embodiment;

[0026] Figure 8 is a schematic diagram of multiple images of a target environment area in one embodiment;

[0027] Fig. 9 A structural block diagram of a device for identifying a desktop that has not been cleaned in time in one embodiment;

[0028] Fig.10 A system block diagram of a system for identifying a desktop that is not cleaned in time in one embodiment;

[0029] Fig.11 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0031] In the on-site management needs of catering stores, there are requirements for the dining environment of customers. One of them is to clean the tableware and dining garbage in time after the user finishes the meal. At present, the on-site inspection of unclean tables can only be achieved by people watching the video, which has the problems of increased costs, increased management difficulty and low efficiency.

[0032] Based on this, the present application provides a method for identifying that a desktop is not cleaned in time, which can be applied to Figure 1 In the application environment shown in Figure 1, the application environment contains multiple desktops. Figure 1 It can be seen that the application environment includes a table with remaining food residues or tableware and a clean and tidy tabletop. In order to maintain the hygiene level of the restaurant, the tabletop needs to be cleaned in time when there are food residues or tableware on the table and there are no customers.

[0033] Combination Figure 1 The application environment shown is described in detail below. The implementation process of the method for identifying that the desktop is not cleaned in time of the present application is described in detail.

[0034] In one embodiment, Figure 2 As shown, a method for identifying whether a tabletop is not cleaned in time is provided, which is used to identify whether a tabletop is cleaned in time in a restaurant or other dining environment. The method specifically includes the following steps:

[0035] Step S201, acquiring an image of a target environment area shot at intervals; wherein the image includes at least one detection area, and the detection area includes at least a desktop, and the desktop is used to place dining items.

[0036] The target environment area includes areas in restaurants, cafeterias, beverage shops or other dining environments. The images of the target environment area refer to images of the target environment area taken at intervals (e.g., every few minutes) by a camera or other imaging device. For example, according to the customer flow and cleaning requirements of the restaurant, the images of the target environment area are taken every 3 minutes, 10 minutes or 20 minutes. One shot can be a single picture, several consecutive images, a video image lasting several seconds, or a moving image lasting several seconds.

[0037] The image includes at least one detection area, which includes at least one tabletop. The tabletop is used to place dining items. Whether the dining items on the tabletop in the detection area need to be cleaned in time can be identified and analyzed as a unit of a detection area. For example, the camera is installed in a position facing the tabletop and avoiding obstructions to ensure that dining items on the tabletop in the detection area and customers around the tabletop can be captured. It is understandable that the camera used to shoot the target environment area can be fixed in position and at a fixed angle to facilitate users to plan the detection area.

[0038] Step S202: Identify whether there are dining item targets and customer targets in the detection area.

[0039] Among them, dining item targets refer to dining items on the table in the detection area, such as plates, bowls and chopsticks, cups, food, etc. Customer targets refer to customers around the table in the detection area. The dining item targets and customer targets in the detection area can be effectively identified through the deep learning target detection algorithm.

[0040] Step S203: Identify the location of the dining item target within the detection area.

[0041] The position of the dining item target refers to the position of the dining item target in the detection area. For example, a rectangular coordinate system can be established based on the detection area, and the position of the dining item target is marked in the form of coordinates in the rectangular coordinate system.

[0042] Step S204: If within the preset duration, no customer target is identified in the detection area, dining item targets are identified, and the position movement of the dining item targets is less than a preset threshold, it is determined that the tabletop has not been cleaned in time.

[0043] Among them, the preset duration can be the time required to clean the table within a certain period of time after the customer finishes the meal, based on the customer flow and cleaning needs of the restaurant, for example, 8 minutes or 10 minutes. The preset duration can also be an estimated time, such as the time from the customer starting to eat, to leaving the table after the meal, until the table needs to be cleaned, for example, it can be 40 minutes or 50 minutes. Further, the preset threshold is a distance threshold used as a judgment of the position movement of the dining item target. Considering that after the customer finishes the meal and leaves, the dining items on the uncleaned table are slightly moved due to the waiter or other personnel passing by, or due to the wind in the store. Therefore, if the position movement of the dining item on the table is less than the preset threshold, it can be considered to be caused by the above factors. If the position movement is greater than the preset threshold, it is considered that the customer has moved the dining item. For example, the preset threshold can be set to 2 cm. In the rectangular coordinate system, the position of the dining item target in multiple images is tracked to determine whether the position movement distance of the dining item target is less than 2 cm.

[0044] In summary, the method of identifying a tabletop that has not been cleaned in time in this embodiment is by acquiring an image of a target environment area shot at intervals; wherein the image includes at least one detection area, and the detection area includes at least one tabletop, and the tabletop is used to place dining items; further identifying whether there are dining item targets and customer targets in the detection area and the positions of the dining item targets; if within a preset duration, no customer target is identified in the detection area, and a dining item target is identified, and the position movement of the dining item target is less than a preset threshold, that is, there is a tabletop where the dining customer has left but there are still dining items on the tabletop, and therefore it is determined that the tabletop has not been cleaned in time. Compared with manual inspections or manual review of video recordings in related technologies to determine whether there is a tabletop to be cleaned, this embodiment can accurately and efficiently identify that the tabletop has not been cleaned in time. It ensures that the customer's tabletop can be cleaned in time after the meal, thereby improving the efficiency of identification and management of restaurant tabletop cleaning and reducing management difficulty and cost.

[0045] In order to more accurately and efficiently identify the cleanliness of the desktop, in one embodiment, the method further includes: obtaining the coordinates of each pre-generated detection area, and determining the detection area in the image according to the coordinates of each detection area. The detection area including the desktop under the camera's viewing angle can be determined by image segmentation, edge detection, etc., and the detection area including the desktop can also be defined in the camera by manual annotation to generate the coordinates of each detection area. For example, Figure 3a and Figure 3b As shown, Figure 1The application environment of the example is used as the target environment area. For the image of the target environment area, it can be seen that the image includes 4 detection areas. Each detection area includes at least a tabletop, seats around the tabletop, and dining items on the tabletop. The detection area includes rectangles and polygons. The coordinates of the detection area can be represented by the coordinates of the boundaries and vertices of the rectangle or polygon, so as to determine the detection area in the image based on the coordinates of each detection area. It should be noted that if the layout of the camera or the target environment area changes, the coordinates of each detection area can be updated and adjusted accordingly.

[0046] By using the pre-generated coordinates of each detection area, the detection areas in several images can be simply and efficiently located, which is conducive to effectively and accurately monitoring the cleanliness of the desktops in multiple detection areas.

[0047] In one embodiment, the coordinates of each pre-generated detection area are obtained, including: obtaining a baseline image of the target environment area, as well as the angle of a camera used for interval shooting and the position of each desktop in the target environment area; generating each detection area in the baseline image based on the angle of the camera and the position of each desktop; and determining the coordinates of each detection area according to the position of each detection area in the baseline image.

[0048] The reference image is an image containing multiple desktops acquired with the camera as the reference point. The reference image will be used as a reference or benchmark for subsequent analysis. The angles of the camera that captures the image include pitch angle, yaw angle, roll angle, etc. The position of each desktop in the target environment area can be inferred through spatial measurement or reference objects in the image.

[0049] For example, in a three-dimensional coordinate system established based on a reference image with a camera as a fixed reference object, the X-axis and Y-axis represent the horizontal position with respect to the desktop, and the Z-axis represents the height of the desktop, so the position of the desktop can be represented by three-dimensional coordinates (x, y, z). Then, the position coordinates of the desktop can be expanded as a whole so that the detection area completely includes the desktop, thereby determining the coordinates of the detection area.

[0050] By adopting the above-mentioned method, the accuracy and efficiency of the acquired detection area and the coordinates of the detection area can be improved, so that the subsequent recognition and analysis of the customer targets in the detection area and the dining item targets on the table can be more accurate.

[0051] In order to improve the accuracy of identifying whether the desktop is cleaned in time, in one embodiment, acquiring images of the target environment area taken at intervals includes: acquiring N1 images of the target environment area taken at intervals.

[0052] Wherein, N1 is a preset integer. For example, an image of the target environment area is captured every 3 minutes, and a total of 10 (ie, N1 is 10) images of the target environment area are acquired in 30 minutes.

[0053] If within the preset duration, no customer target is identified in the detection area, and the presence of dining item target is identified, and the position movement of the dining item target is less than the preset threshold, it is determined that the desktop has not been cleaned in time. The method includes: within the preset duration, if for N2 consecutive images, no customer target is identified in the detection area, and the presence of dining item target is identified, and the position movement of the dining item target is less than the preset threshold, it is determined that the desktop has not been cleaned in time; wherein the N2 images are partial images of the N1 images.

[0054] Among them, N2 is a preset integer, and N2 represents the number of consecutive images used to determine whether the desktop is not cleaned in time within the preset duration. Exemplarily, one image of the target environment area is acquired every 1 minute, and a total of 10 images are acquired in 10 minutes, and N1 is 10. Within the preset duration, for example, 5 minutes are taken out of 10 minutes, and 5 images of the target environment area are acquired in 5 minutes, respectively, of which 4 consecutive images (N2 is 4) meet: no customer target is identified in the detection area, the presence of dining item target is identified, and the position movement of the dining item target is less than the preset threshold, that is, it is determined that the desktop in the detection area is not cleaned in time.

[0055] It can be understood that if three consecutive images satisfy the following conditions: no customer target is identified in the detection area, dining item target is identified, and the position movement of the dining item target is less than the preset threshold, but the fourth image does not meet the above conditions, the counting will be restarted until four consecutive images meet the above conditions, that is, until N2 consecutive images meet the following conditions: no customer target is identified in the detection area, dining item target is identified, and the position movement of the dining item target is less than the preset threshold, it is determined that the table top in the detection area is not cleaned in time.

[0056] In this embodiment, by acquiring N1 images of the target environment area taken at intervals, if, in the N1 images, within the preset duration, if N2 consecutive images do not identify the presence of a customer target in the detection area, identify the presence of a dining item target, and the position movement of the dining item target is less than a preset threshold, it is determined that the tabletop has not been cleaned in time. This method avoids inaccurate judgments caused by possible contingencies or errors in a single image, thereby improving the accuracy of identifying that the tabletop has not been cleaned in time.

[0057] Further, in order to improve the accuracy of determining that the position movement of the dining item target is less than a preset threshold, in one embodiment, Figure 4As shown, identifying that the position movement of the dining object target in the detection area in the continuous N2 images is less than a preset threshold includes:

[0058] Step S401, when no customer target is identified in the detection area and dining item target is identified in the continuous N3 images, the coordinates of the dining item target in the detection area of ​​the first image in the N3 images are obtained as the original coordinates; wherein the N3 images are partial images of the N1 images.

[0059] Wherein, N3 is a preset integer. Specifically, there are N3 consecutive images among the N1 images in which the customer target is not recognized in the detection area and the dining item target is recognized in the detection area. The first image among the N3 consecutive images is obtained, the coordinates of the dining item target in the detection area of ​​the image are determined, and the coordinates are used as the original coordinates.

[0060] Step S402, obtaining the coordinates of the dining item target in the detection area in multiple consecutive images, taking the coordinate with the largest change value relative to the original coordinate as the end point coordinate, and determining the movement value of the dining item target based on the original coordinate and the end point coordinate.

[0061] For example, assuming that the coordinates of the dining object target in the detection area in the continuous multiple images are (x1, y1), (x2, y2), (x3, y3), and the original coordinates are (x0, y0), the change values ​​of the coordinates of the dining object target relative to the original coordinates are sqrt((x1-x0) 2 +(y1-y0) 2 )、sqrt((x2-x0) 2 +(y2-y0) 2 )、sqrt((x3-x0) 2 +(y3-y0) 2 ), assuming that the coordinate with the largest change value is (x3, y3), then the coordinate (x3, y3) is marked as the end coordinate, and the moving value of the dining item target is sqrt((x3-x0) 2 +(y3-y0) 2 ). Among them, sqrt() is a function used to calculate the square root of a number.

[0062] Step S403, until the movement value in the consecutive N2 images is less than the preset threshold, N3 is greater than or equal to N2, and the N2 images are partial images in the N3 images.

[0063] Specifically, first, among the N1 images, there are N3 consecutive images that satisfy: no customer target is identified in the detection area, and dining item target is identified in the detection area. Secondly, among the N3 consecutive images, there are N2 consecutive images that satisfy that the position movement of the dining item target is less than a preset threshold, thereby determining that the tabletop in the detection area is not cleaned in time.

[0064] Exemplarily, 6 (i.e., N1) images of the target environment area taken at intervals are acquired in sequence as follows: Figure 5 Shown are (i), (ii), (iii), (iv), (v), and (vi). Observation Figure 5 It can be seen that images (1) and (2) are both in which the customer target 110 is dining, image (3) is in which the customer target 110 stands up and leaves the table, and images (4), (5), and (6) are all in which the dining item target 120 exists on the table but the customer target 110 is not identified. It can be seen that in the three consecutive images (4), (5), and (6), the customer target 110 is not identified in the detection area, and the dining item target 120 is identified. The coordinates of the dining item target 120 in the detection area of ​​the first image of these three (i.e., N3) images, i.e., image (4), are taken as the original coordinates. The movement values ​​of the coordinates of the dining item target 120 in the detection area of ​​images (5) and (6) relative to the original coordinates are obtained in sequence. Assuming that the movement values ​​of the coordinates of the dining item target 120 in the detection area of ​​images (5) and (6) relative to the original coordinates are less than the preset threshold, i.e., the movement values ​​in two consecutive (i.e., N2) images are less than the preset threshold, it is determined that the table in the detection area has not been cleaned in time.

[0065] Through the above method, the position changes of dining item targets and customer targets over a period of time can be effectively identified, and the accuracy of judging that the position movement of dining item targets is less than a preset threshold can be improved, thereby improving the accuracy of determining that the table is not cleaned in time when the customer target is not identified in the detection area and the dining item target is identified.

[0066] In one embodiment, the coordinates with the largest change value relative to the original coordinates are taken as the end point coordinates, and the movement value of the dining item target is determined based on the original coordinates and the end point coordinates, including: the coordinates of the dining item target include each vertex of the identification frame of the dining item target, in the identification frame of the dining item target, the original coordinates of each vertex correspond to the position of the end point coordinates one by one, and the vertex coordinates with the largest difference between the original coordinates and the end point coordinates among each vertex are taken as the movement value of the dining item target.

[0067] For example, Figure 6As shown in the figure, assume that the recognition box of the dining object target is a rectangular box, and each recognition box has four vertices. Assume that the original coordinates of the upper left corner vertex of one of the recognition boxes are (x1, y1), and the end point coordinates of the vertex in the subsequent image become (x2, y2), then the moving distance of the vertex is sqrt((x2-x1) 2 +(y2-y1) 2 ), perform this calculation for all vertices, and select the largest movement distance as the movement value of the dining item target.

[0068] In this way, the movement value of the position change of the recognition frame of the meal item target in the image can be accurately determined.

[0069] In one embodiment, the method for identifying that a desktop has not been cleaned in time also includes: if it is identified within the duration that there is a customer target in the detection area who is not facing the table in the detection area, the duration or the number of consecutive images is not reset.

[0070] Specifically, the timing starts when the camera recognizes that there is no customer target in the detection area. If during the timing process the camera recognizes that there is a customer target in the detection area, but the customer target is not facing the table in the detection area, it may be that the customer target is sideways or has its back to the table, which means that the customer target is not a dining customer at the table in the detection area, but a dining customer or waiter at another table, etc., so the duration will not be reset, which means that the duration will continue to accumulate, and whether the table in the detection area needs to be cleaned will not be re-identified because the customer target is not facing the table. If the customer target is facing the table, the duration will be reset, which means that the customer may be dining or preparing to dine, and the cleanliness of the table has not yet reached a level that requires attention.

[0071] In one embodiment, if the camera only has the function of capturing images but does not have the function of identifying customers, the camera can capture images and upload them to the server for analysis. When there is a customer target in the captured image but the customer is not facing the table in the detection area, the number of consecutive images will not be reset. Specifically, if N2 is set to 4, when two consecutive images meet the following conditions: the customer target is not identified in the detection area, the dining item target is identified, and the position movement of the dining item target is less than the preset threshold; the third image meets the following conditions: the customer target is in the detection area, but the customer target is not facing the table in the detection area, then the count will not be re-counted; if the fourth image meets the following conditions: the customer target is not identified in the detection area, the dining item target is identified, and the position movement of the dining item target is less than the preset threshold, then it is determined that the tabletop in the detection area has not been cleaned in time; if the fourth image does not meet the above conditions, it means that the customer may be dining or preparing to dine, and the cleanliness of the tabletop has not yet reached the level that requires attention.

[0072] It can be understood that since the time interval between the images taken by the camera is known, it can also be understood that if the following conditions are met during the duration: no customer target is identified in the detection area, dining item targets are identified, and the position movement of the dining item targets is less than a preset threshold, then it is determined that the table top in the detection area is not cleaned in time.

[0073] By not resetting the duration or the number of consecutive images when it is identified within the duration that there is a customer target in the detection area who is not facing the table in the detection area, and continuing to identify whether the table in the detection area needs to be cleaned, the accuracy of judging the cleaning status of the table is improved, avoiding the problem of misjudgment due to the customer temporarily leaving or engaging in other non-dining activities.

[0074] In order to improve the accuracy of determining whether the customer target is facing the table, in one embodiment, a method for identifying that the customer target is not facing the table within the detection area includes: determining whether the customer target is facing the camera; determining the relative positions of the camera, the customer target and the table; and determining the orientation of the customer target relative to the table within the detection area based on the orientation of the customer target relative to the camera and the relative positions of the customer target and the table.

[0075] Whether the customer target is facing the camera can be determined by the customer target's body posture, facial features or other biometric features in the image. For example, if the customer target's eyes and nose are identified in the image and the body is not obviously turned sideways, it indicates that the customer target is facing the camera.

[0076] The relative positions among the camera, the customer target and the table can be determined by spatial geometric analysis. For example, a rectangular coordinate system is established based on the image, the camera position is used as the origin of the rectangular coordinate system, the coordinates of the customer target and the table in the rectangular coordinate system are obtained, the vector between the camera and the table is calculated, and the vector between the customer target and the table is calculated, thereby determining the relative positions among the camera, the customer target and the table.

[0077] Exemplary, reference Figure 3a and Figure 3b , customer targets 111 and 112 face the camera 310, while customer targets 113 and 114 do not face the camera 310. The red dotted line indicates the shortest distance between the camera 310 and each customer target, and the green dotted line indicates the shortest distance between the camera 310 and each table. By comparing these shortest distances, the relative positions of the camera, customer targets, and tables are determined. Furthermore, by comparing the relative positions of the camera, customer targets, and tables in combination with the orientations of each customer target, the orientation of the customer target relative to the table in the detection area is determined.

[0078] In this embodiment, by first determining whether the customer target is facing the camera, the relative positions of the camera, the customer target and the table are determined, and then based on the orientation of the customer target relative to the camera and the relative positions of the customer target and the table, the orientation of the customer target relative to the table is determined, thereby improving the accuracy of judging whether the customer target is facing the table, so as to make more accurate decisions in desktop cleaning management.

[0079] In one embodiment, based on the orientation of the customer target relative to the camera and the relative positions of the customer target and the table, the orientation of the customer target relative to the table in the detection area is determined, including: if the customer target is facing the camera and the customer target is closer to the camera relative to the table, determining that the customer target is not facing the table in the detection area.

[0080] Exemplary, reference Figure 3a , the customer target 111 is facing the camera 310, and the customer target 111 is closer to the camera 310 relative to the table in the detection area 211, that is, the back of the customer target 111 is closer to the table in the detection area 211, so it can be determined that the customer target 111 is not facing the table in the detection area 211.

[0081] In another embodiment, the orientation of the customer target is determined based on the orientation of the customer target and the relative position of the customer target and the table, including: if the customer target is toward the camera and the customer target is farther away from the camera relative to the table, determining that the customer target is toward the table of the detection area.

[0082] Exemplary, reference Figure 3a , the customer target 112 is facing the camera 310, and the customer target 112 is farther away from the camera 310 than the table in the detection area 212, that is, the front of the customer target 112 is facing the camera 310, and the customer target 112 is farther away from the camera 310 than the table in the detection area 212, so it can be determined that the customer target 112 is facing the table in the detection area 212.

[0083] In another embodiment, the orientation of the customer target is determined based on the orientation of the customer target and the relative position of the customer target and the table, including: if the customer target is facing away from the camera, the customer target is farther away from the camera relative to the table, and it is determined that the customer target is not facing the table in the detection area.

[0084] Exemplary, reference Figure 3b , the customer target 114 turns away from the camera 310, and the customer target 114 is farther away from the camera 310 relative to the table in the detection area 212, that is, when the customer target 114 turns away from the camera 310 and is farther away from the table, the back of the customer target 114 faces the table in the detection area 212, so it can be determined that the customer target 114 is not facing the table in the detection area 212.

[0085] In another embodiment, the orientation of the customer target is determined based on the orientation of the customer target and the relative position of the customer target and the table, including: if the customer target is facing away from the camera, the customer target is closer to the camera relative to the table, and determining that the customer target is facing the table in the detection area.

[0086] Exemplary, reference Figure 3b , the customer target 113 is facing away from the camera 310, and the customer target 113 is closer to the camera 310 relative to the table in the detection area 211, that is, although the customer target 113 is facing away from the camera 310, it is closer to the table in the detection area 211, so it is determined that the customer target 113 is facing the table in the detection area 211.

[0087] By using the above four methods of determining the orientation of the customer's target based on the orientation of the customer's target and the relative position of the customer's target and the table, it is possible to more accurately determine whether the customer is facing the table. It should be noted that the above four methods of determining the orientation of the customer's target based on the orientation of the customer's target and the relative position of the customer's target and the table can also be combined with two or more of them to more accurately determine the orientation of the customer's target.

[0088] In the process of monitoring the cleaning status of the desktop, ensuring accurate monitoring of the desktop cleaning requires the ability to identify and exclude items that have not been moved or cleaned, based on this, such as Figure 7 As shown, in one embodiment, the method for identifying that the desktop is not cleaned in time also includes:

[0089] Step S701: Acquire multiple images of a target environment area taken at intervals within a first preset time period, wherein the first preset time period is set long enough to ensure that the employee will definitely clean the desk within the first preset time period.

[0090] The first preset time period may be set according to the restaurant's operating model, average dining time, and cleaning habits, and the first preset time period may be long enough, such as 12 hours.

[0091] Step S702: Identify a first object target on the desktop that has not moved within a first preset time period.

[0092] The first object target refers to an object that has existed for a long time or has not been moved, such as a QR code for ordering food, an advertising sticker, a decorative sticker, etc. on the desktop.

[0093] Specifically, within a first preset time period, multiple images within the first preset time period are analyzed, positions of objects in the multiple images are tracked and compared, and objects on the desktop whose positions have not moved are found as first object targets.

[0094] Step S703: when identifying dining items in the detection area, exclude the first object target.

[0095] For example, Figure 8 As shown, four images of the target environment area taken at intervals within a first preset time period are obtained. It can be seen that the blue framed parts in the figure are fixed ordering QR codes and advertising stickers. The ordering QR code and advertising stickers are used as the first object target 121. When the dining object target 120 in the subsequent identification detection area is used, these items marked as the first object target 121 will no longer be considered.

[0096] In this way, items that do not need to be moved or cleaned can be accurately identified and excluded, thereby reducing false alarms and improving the accuracy of desktop cleaning monitoring.

[0097] In the process of monitoring the cleaning status of the desktop, accurately identifying and excluding items that are often present but may move or separate is crucial to improving the accuracy of identifying whether the desktop is cleaned in a timely manner. Based on this, in one embodiment, the method for identifying that the desktop is not cleaned in a timely manner also includes: acquiring multiple images of the target environment area taken at intervals within a second preset time period, wherein the second preset time period is set long enough to ensure that the employee will definitely clean the desktop within the second preset time period; among the multiple images, if the detection area of ​​images greater than a preset number has identified the presence of a second object target on the desktop, then when identifying dining items in the detection area, the second object target is excluded; wherein the second object target can move or separate on the desktop.

[0098] The second preset time period can be set according to the restaurant's operating mode, average dining time and cleaning habits, and the second preset time period is long enough, such as 12 hours. The second object target represents an object that can be moved or separated on the desktop. For example, a movable billboard, table number plate, flower bouquet, napkin box, etc. on the desktop. The preset number represents the number of images in which the second object target exists in the detection area.

[0099] For example, assuming that the first preset time and the second preset time are the same, the preset number is 4. The multiple images of the target environment area taken at intervals within the second preset time period are as follows: Figure 8 Taking the four images shown as an example, in the four images, the red framed part is a movable and detachable bouquet, that is, the second object target 122 is identified on the table in the detection area of ​​the four images. When the dining object target 120 in the detection area is subsequently identified, the second object target 122 is excluded.

[0100] In this way, items that are often present but may move or become detached can be accurately identified and excluded, thereby reducing false alarms and improving the accuracy of desktop cleaning monitoring.

[0101] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0102] Based on the same inventive concept, the embodiment of the present application also provides a device for implementing the above-mentioned method for identifying that the desktop has not been cleaned in time. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in the embodiment of the device for identifying that the desktop has not been cleaned in time provided below can be referred to the limitations of the method for identifying that the desktop has not been cleaned in time above, and will not be repeated here.

[0103] In one embodiment, Fig. 9 As shown, a device for identifying that a desktop is not cleaned in time is provided, the device comprising:

[0104] The acquisition module 901 is used to acquire the images of the target environment area taken at intervals; wherein the images include at least one detection area, and the detection area includes at least a desktop, and the desktop is used to place dining items.

[0105] The identification module 902 is used to identify whether there are dining item targets and customer targets in the detection area; and is also used to identify the location of the dining item targets in the detection area.

[0106] Control module 903 determines that the tabletop has not been cleaned in time if no customer target is identified in the detection area within a preset duration, the presence of dining item targets is identified, and the position movement of the dining item targets is less than a preset threshold.

[0107] In one embodiment, the acquisition module 901 acquires the images of the target environment area taken at intervals, including: the acquisition module 901 acquires N1 images of the target environment area taken at intervals.

[0108] The control module 903 determines that the tabletop is not cleaned in time if no customer target is identified in the detection area within the preset duration, the dining item target is identified, and the position movement of the dining item target is less than the preset threshold value:

[0109] The acquisition module 901 is used to determine that the table top has not been cleaned in time if, within a preset duration, N2 consecutive images fail to identify the existence of a customer target in the detection area, the existence of a dining item target is identified, and the position movement of the dining item target is less than a preset threshold; wherein the N2 images are partial images of the N1 images.

[0110] In one embodiment, the recognition module 902 is used to recognize that the position movement of the dining object target in the detection area in the consecutive N2 images is less than a preset threshold, including:

[0111] The recognition module 902 is used to obtain the coordinates of the dining item target in the detection area of ​​the first image in N3 images as the original coordinates when no customer target is recognized in the detection area and the dining item target is recognized in N3 consecutive images; wherein the N3 images are partial images of the N1 images.

[0112] The acquisition module 901 is used to obtain the coordinates of the dining item target in the detection area in multiple consecutive images, take the coordinate with the largest change value relative to the original coordinate as the end coordinate, and determine the movement value of the dining item target based on the original coordinate and the end coordinate.

[0113] The control module 903 continues until the movement value in N2 consecutive images is less than the preset threshold, N3 is greater than or equal to N2, and the N2 images are partial images in the N3 images.

[0114] In one embodiment, the coordinate with the largest change value relative to the original coordinate is used as the end point coordinate, and the movement value of the dining item target is determined based on the original coordinate and the end point coordinate, including:

[0115] Acquisition module 901, the coordinates of the dining item target include the vertices of the identification frame of the dining item target. In the identification frame of the dining item target, the original coordinates of each vertex correspond to the position of the end point coordinates one by one, and the vertex coordinates with the largest difference between the original coordinates and the end point coordinates are used as the movement value of the dining item target.

[0116] In one embodiment, the control module 903 is further configured not to reset the duration or the number of consecutive images if it is identified within the duration that there is a customer target in the detection area but the customer target is not facing the table in the detection area.

[0117] In one embodiment, the method in which the recognition module 902 is used to recognize that the customer target is not facing the table in the detection area includes: the recognition module 902 is used to determine whether the customer target is facing the camera; and is also used to determine the relative positions of the camera, the customer target and the table; based on the orientation of the customer target relative to the camera and the relative positions of the customer target and the table, determine the orientation of the customer target relative to the table in the detection area.

[0118] In one embodiment, the recognition module 902 determines the orientation of the customer target relative to the table in the detection area based on the orientation of the customer target relative to the camera and the relative positions of the customer target and the table, including:

[0119] A recognition module 902 for determining that the customer target is not facing the table within the detection area if the customer target is facing the camera and the customer target is closer to the camera than the table; and / or,

[0120] A recognition module 902 for determining that the customer target is facing the table within the detection area if the customer target is facing the camera and the customer target is farther from the camera than the table; and / or,

[0121] Identification module 902, for determining that the customer target is not facing the table within the detection area if the customer target is facing away from the camera and the customer target is further away from the camera relative to the table; and / or,

[0122] The identification module 902 is used to determine that the customer target is facing the table in the detection area if the customer target is facing away from the camera and the customer target is closer to the camera relative to the table.

[0123] In one embodiment, the acquisition module 901 is also used to acquire multiple images of the target environment area taken at intervals within a first preset time period, wherein the first preset time period is set long enough to ensure that the employee will definitely clean the desktop within the first preset time period; the identification module 902 is also used to identify a first object target that has not moved on the desktop within the first preset time period; the control module 903 is also used to exclude the first object target when identifying dining items in the detection area.

[0124] In one embodiment, the acquisition module 901 is also used to acquire multiple images of the target environment area taken at intervals within a second preset time period, wherein the second preset time period is set long enough to ensure that the employee will definitely clean the desktop within the second preset time period; the control module 903 is also used to exclude the second object target from being identified when dining items are identified in the detection area if a second object target is identified on the desktop in the detection area of ​​images greater than a preset number among the multiple images; wherein the second object target can be moved or separated on the desktop.

[0125] Each module in the above-mentioned device for identifying a desktop that has not been cleaned in time can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each of the above modules.

[0126] In one embodiment, Fig.10 As shown, a system for identifying that a desktop is not cleaned in time is also provided, the system comprising:

[0127] The image acquisition device 1001 is used to take images of the target environment area at intervals; wherein the image includes at least one detection area, and the detection area includes at least one desktop, and the desktop is used to place dining items;

[0128] Server 1002 is used to identify whether there are dining item targets and customer targets in the detection area; it is also used to identify the position of the dining item targets in the detection area; if within a preset duration, no customer target is identified in the detection area, the presence of dining item targets is identified, and the position movement of the dining item targets is less than a preset threshold, it is determined that the table has not been cleaned in time.

[0129] It should be noted that the image acquisition device 1001 and the server 1002 in the system for identifying illegally picked food cooperate with each other to execute the implementation scheme recorded in the above-mentioned method for identifying that the desktop is not cleaned in time. Therefore, the specific implementation scheme in the image acquisition device 1001 and the server 1002 refers to the above-mentioned definition of the method for identifying that the desktop is not cleaned in time, and will not be repeated here.

[0130] In one embodiment, a computer device is also provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program. The computer device may be a server, and its internal structure diagram may be as follows: Fig.11 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The network interface of the computer device is used to communicate with an external terminal through a network connection.

[0131] Those skilled in the art will understand that Fig.11The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0132] In one embodiment, a computer-readable storage medium is further provided, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0133] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive memory, magnetic memory, ferroelectric memory, phase change memory, graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in the embodiments provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, etc., but is not limited thereto.

[0134] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0135] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for identifying that a desktop is not cleaned in time, characterized in that: The method comprises: Acquire an image of the target environment area taken at intervals; wherein the image includes at least one detection area, and the detection area includes at least a desktop, and the desktop is used to place dining items; Identify whether there are dining item targets and customer targets in the detection area; Identifying the location of the dining object target within the detection area; If within a preset duration, the customer target is not identified in the detection area, the dining item target is identified, and the position movement of the dining item target is less than a preset threshold, it is determined that the tabletop is not cleaned in time; Wherein, acquiring the images of the target environment area photographed at intervals includes: acquiring N1 images of the target environment area photographed at intervals; If within a preset duration, the customer target is not identified as existing in the detection area, the dining item target is identified as existing, and the position movement of the dining item target is less than a preset threshold, it is determined that the desktop has not been cleaned in time, including: within a preset duration, if there are N2 consecutive images, the customer target is not identified as existing in the detection area, the dining item target is identified as existing, and the position movement of the dining item target is less than a preset threshold, it is determined that the desktop has not been cleaned in time; wherein the N2 images are partial images of the N1 images.

2. The method for identifying that a desktop is not cleaned in time according to claim 1, characterized in that: Identifying that the position movement of the dining object target in the detection area in the consecutive N2 images is less than a preset threshold, comprising: When the customer target is not identified in the detection area and the dining item target is identified in the continuous N3 images, the coordinates of the dining item target in the detection area of ​​the first image in the N3 images are obtained as the original coordinates; wherein the N3 images are partial images of the N1 image; Acquire the coordinates of the dining object target in the detection area in the continuous multiple images, take the coordinate with the largest change value relative to the original coordinate as the end point coordinate, and determine the movement value of the dining object target based on the original coordinate and the end point coordinate; Until the movement value in the continuous N2 images is less than the preset threshold, the N3 is greater than or equal to N2, and the N2 images are partial images of the N3 images.

3. The method for identifying that a desktop is not cleaned in time according to claim 2, characterized in that: The step of taking the coordinate with the largest change value relative to the original coordinate as the end point coordinate, and determining the movement value of the dining item target based on the original coordinate and the end point coordinate includes: The coordinates of the dining item target include the vertices of the identification frame of the dining item target. In the identification frame of the dining item target, the original coordinates of each vertex correspond to the position of the end point coordinates one by one, and the vertex coordinates with the largest difference between the original coordinates and the end point coordinates are used as the movement value of the dining item target.

4. The method for identifying that a desktop is not cleaned in time according to any one of claims 1 to 3, characterized in that: The method further includes: if it is identified within the duration that the customer target exists in the detection area but is not facing a table within the detection area, the duration or the number of consecutive images is not reset.

5. The method for identifying that a desktop is not cleaned in time according to claim 4, characterized in that: The method for identifying that the customer target is not facing the table within the detection area includes: determining whether the customer target is facing toward a camera; Determine the relative positions of the camera, the customer target and the table; and determine the orientation of the customer target relative to the table within the detection area based on the orientation of the customer target relative to the camera and the relative positions of the customer target and the table.

6. The method for identifying that a desktop is not cleaned in time according to claim 5, characterized in that: Determining the orientation of the customer target relative to the table within the detection area based on the orientation of the customer target relative to the camera and the relative positions of the customer target and the table includes: If the customer target is facing the camera, and the customer target is closer to the camera than the table, determining that the customer target is not facing the table within the detection area; and / or, If the customer target is facing the camera, and the customer target is farther from the camera than the table, determining that the customer target is facing the table within the detection area; and / or, If the customer target faces away from the camera, the customer target is further away from the camera relative to the table, and it is determined that the customer target is not facing the table within the detection area; and / or, If the customer target faces away from the camera, the customer target is closer to the camera relative to the table, and it is determined that the customer target is facing the table within the detection area.

7. The method for identifying that a desktop is not cleaned in time according to claim 1, characterized in that: The method further comprises: Acquire multiple images of the target environment area taken at intervals within a first preset time period, wherein the first preset time period is set long enough to ensure that the employee will definitely clean the desk within the first preset time period; Identifying a first object target that has not moved on the desktop within a first preset time period; When identifying the dining items within the detection area, the first object target is excluded.

8. The method for identifying that a desktop is not cleaned in time according to claim 1, characterized in that: The method further comprises: Acquire multiple images of the target environment area taken at intervals within a second preset time period, wherein the second preset time period is set long enough to ensure that the employee will definitely clean the desk within the second preset time period; If, in the detection areas of the images greater than a preset number in the multiple images, a second object target is identified on the table, then when identifying the dining items in the detection area, the second object target is excluded; wherein, the second object target can be moved or separated on the table.

9. A device for identifying that a desktop is not cleaned in time, characterized in that: The device comprises: An acquisition module, used for acquiring an image of a target environment area photographed at intervals; wherein the image includes at least one detection area, and the detection area includes at least a desktop, and the desktop is used to place dining items; An identification module, used to identify whether there are dining object targets and customer targets in the detection area; and also used to identify the location of the dining object targets in the detection area; The control module determines that the tabletop is not cleaned in time if the customer target is not identified in the detection area within a preset duration, the dining item target is identified, and the position movement of the dining item target is less than a preset threshold; Wherein, acquiring the images of the target environment area photographed at intervals includes: acquiring N1 images of the target environment area photographed at intervals; If within a preset duration, the customer target is not identified as existing in the detection area, the dining item target is identified as existing, and the position movement of the dining item target is less than a preset threshold, it is determined that the desktop has not been cleaned in time, including: within a preset duration, if there are N2 consecutive images, the customer target is not identified as existing in the detection area, the dining item target is identified as existing, and the position movement of the dining item target is less than a preset threshold, it is determined that the desktop has not been cleaned in time; wherein the N2 images are partial images of the N1 images.

10. A system for identifying a desktop that has not been cleaned in time, characterized in that: The system comprises: An image acquisition device, used for taking images of a target environment area at intervals; wherein the image includes at least one detection area, and the detection area includes at least a desktop, and the desktop is used to place dining items; The server is used to identify whether there are dining item targets and customer targets in the detection area; and is also used to identify the position of the dining item targets in the detection area; if within a preset duration, the presence of the customer target in the detection area is not identified, the presence of the dining item target is identified, and the position movement of the dining item target is less than a preset threshold, it is determined that the tabletop is not cleaned in time; Wherein, acquiring the images of the target environment area photographed at intervals includes: acquiring N1 images of the target environment area photographed at intervals; If within a preset duration, the customer target is not identified as existing in the detection area, the dining item target is identified as existing, and the position movement of the dining item target is less than a preset threshold, it is determined that the desktop has not been cleaned in time, including: within a preset duration, if there are N2 consecutive images, the customer target is not identified as existing in the detection area, the dining item target is identified as existing, and the position movement of the dining item target is less than a preset threshold, it is determined that the desktop has not been cleaned in time; wherein the N2 images are partial images of the N1 images.

11. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for identifying that a desktop is not cleaned in time according to any one of claims 1 to 8 are implemented.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for identifying that a desktop is not cleaned in time according to any one of claims 1 to 8 are implemented.

Images