Method, apparatus, and system for identifying abnormal operations in the back kitchen

Abstract

The present application discloses a method, apparatus, and system for identifying abnormal operations in a kitchen. The method includes: obtaining a target image of a target environmental area in the kitchen, where the target environmental area includes at least one detection area; identifying a first item and a second item in the detection area, and determining whether the first item and the second item intersect according to the position information of the first item and the second item; where the first item is a tool for processing food ingredients, the second item is a movable item, and the second item is prohibited from being placed on the first item; if the first item intersects with the second item, an alarm instruction is output. The present application can improve the efficiency of identifying the illegal operation behaviors of kitchen staff.

Description

Technical Field

[0001] This application relates to the field of computer vision technology, and in particular, to a method, device, and system for identifying abnormal operations in the back kitchen. Background Art

[0002] In the back kitchen scenario of a catering store, there are strict hygiene requirements for back kitchen staff. One of them is that it is not allowed to wash cleaning tools such as brooms, mops, and dustpans in the food sink. Currently, such illegal operation behaviors can only be avoided through the management means of the enterprise, resulting in problems such as increased costs, increased management difficulties, and low efficiency. Summary of the Invention

[0003] Based on this, it is necessary to provide a method, device, and system for identifying abnormal operations in the back kitchen to solve the problem of difficult management of illegal operations in the back kitchen, improve the efficiency of identifying illegal operation behaviors in the back kitchen, and reduce management costs and management difficulties.

[0004] In a first aspect, this application provides a method for identifying abnormal operations in the back kitchen. The method includes:

[0005] Obtain a target image of a target environmental area in the back kitchen, where the target environmental area includes at least one detection area;

[0006] Identify a first item and a second item in the detection area, and determine whether the first item and the second item intersect according to the position information of the first item and the second item; where the first item is a tool for processing food ingredients, the second item is a movable item, and the second item is prohibited from being placed on the first item;

[0007] If the first item intersects with the second item, output an alarm instruction.

[0008] In one embodiment, a first image acquisition device is configured to be able to identify the first item and the second item, a second image acquisition device is configured to be able to identify the first item and the second item, and the installation positions and shooting angles of the first image acquisition device and the second image acquisition device are different. The method further includes:

[0009] The first image acquisition device acquires and obtains a first target image of the target environmental area, identifies the first item and the second item in the detection area of the first target image, and determines whether the first item and the second item intersect;

[0010] The second image acquisition device acquires and obtains a second target image of the target environmental area, identifies the first item and the second item in the detection area of the second target image, and determines whether the first item and the second item intersect;

[0011] If the first image acquisition device determines that the first item and the second item intersect, and the second image acquisition device determines that the first item and the second item intersect, an alarm instruction is output.

[0012] In one embodiment, the first image acquisition device is a first camera, and the second image acquisition device is a second camera. The method further includes:

[0013] When the first camera does not determine that the first item and the second item intersect, the second camera captures second target images at a first time interval;

[0014] When the first camera determines that the first item and the second item intersect, the second camera captures second target images at a second time interval, where the second time interval is less than the first time interval.

[0015] In one embodiment, if the first image acquisition device determines that the first item and the second item intersect, and the second image acquisition device determines that the first item and the second item intersect, outputting an alarm instruction includes:

[0016] The first image acquisition device determines that the first item and the second item intersect, and outputs a first recognition result to the server;

[0017] The second image acquisition device determines that the first item and the second item intersect, and outputs a second recognition result to the server;

[0018] The server outputs an alarm instruction in response to the first recognition result and the second recognition result.

[0019] In one embodiment, the method further includes:

[0020] Obtain the false alarm information feedback by the user within a preset time period, determine the movable items with a false alarm rate higher than the preset false alarm rate threshold, and divide the movable items into at least a first part and a second part. Among the false alarm information feedback by the user: the first part is determined to intersect the first item more than a first threshold, or the proportion of the number of images in which the first part is determined to intersect the first item in the number of images of the false alarm in which the user feedbacks that the first item intersects the second item is greater than a second threshold;

[0021] If it is determined that the first item intersects the first part of the movable item, no alarm instruction is output.

[0022] In one embodiment, the method further includes:

[0023] Obtain a plurality of specific false alarm images, where the specific false alarm images are false alarm images in which the first part is determined to intersect the first item;

[0024] Mark the intersection positions of the first part and the first item in the plurality of specific false alarm images respectively;

[0025] According to the intersection position of the first part and the first item after annotation, the first part is at least divided into a third sub - part and a fourth sub - part, where the third sub - part is determined to intersect the first item more times than a third threshold, or the proportion of the number of images in which the third sub - part intersects the first item among a specific number of false - alarm images is greater than a fourth threshold;

[0026] If it is determined that the third sub - part intersects the first item in the obtained target image, no alarm instruction is output;

[0027] If it is determined that the fourth sub - part intersects the first item in the obtained target image, an alarm instruction is output.

[0028] In one embodiment, the first item is a first cleaning container for cleaning food ingredients, and the second item is a floor cleaning tool. The method further includes:

[0029] Determine a first detection area in the target image, and determine whether the first cleaning container and the floor cleaning tool exist simultaneously in the first detection area, where the first detection area includes at least one first cleaning container;

[0030] If the first cleaning container and the floor cleaning tool exist simultaneously in the first detection area, determine whether the first cleaning container and the floor cleaning tool intersect, or whether the first cleaning container contains the floor cleaning tool. If they intersect or the first cleaning container contains the floor cleaning tool, output a first alarm instruction;

[0031] and / or,

[0032] The first item includes a second cleaning container and a third cleaning container, the second item includes tableware and food ingredients, the second cleaning container is used for cleaning tableware, and the third cleaning container is used for cleaning food ingredients. The method further includes:

[0033] Determine a second detection area in the target image, where the second detection area includes the second cleaning container and the third cleaning container;

[0034] If it is recognized that the second cleaning container intersects the food ingredients in the second detection area, or the third cleaning container intersects the tableware, output a second alarm instruction;

[0035] and / or,

[0036] The first item includes a first placement board, the second item includes a first type of food ingredients and a second type of food ingredients, and the first placement board is used for processing the first type of food ingredients. The method further includes:

[0037] Determine a third detection area in the target image, where the third detection area includes at least the first placement board;

[0038] If the first placement board and the second type of food ingredient are simultaneously recognized in the third detection area and the first placement board and the second type of food ingredient intersect, output a third warning instruction;

[0039] And / or,

[0040] The first item includes a second placement board, the second item includes the first type of food ingredient and the second type of food ingredient, and the second placement board is used for processing the second type of food ingredient. The method further includes:

[0041] Determine a fourth detection area in the target image, where the fourth detection area at least includes the second placement board;

[0042] If the second placement board and the first type of food ingredient are simultaneously recognized in the fourth detection area and the second placement board and the first type of food ingredient intersect, output a fourth warning instruction;

[0043] And / or,

[0044] The first item is a food processing board, the second item is a cleaning cloth, and the method further includes:

[0045] Determine a fifth detection area in the target image, where the fifth detection area includes the food processing board;

[0046] If the food processing board and the cleaning cloth are simultaneously recognized in the fifth detection area and the food processing board and the cleaning cloth intersect, output a fifth warning instruction.

[0047] In one embodiment, determining whether the first cleaning container and the floor cleaning tool intersect or the first cleaning container contains the floor cleaning tool includes:

[0048] Obtain the recognition frame of the first cleaning container in the first detection area;

[0049] Obtain the recognition frame of the floor cleaning tool in the first detection area;

[0050] According to the position coordinates of the recognition frame of the first cleaning container in the target image and the position coordinates of the recognition frame of the floor cleaning tool in the target image, determine whether there is an overlapping area between the recognition frame of the first cleaning container and the recognition frame of the floor cleaning tool;

[0051] If the area ratio of the overlapping area to the recognition frame of the first cleaning container is greater than or equal to a preset first ratio threshold, determine that the first cleaning container contains the floor cleaning tool, otherwise determine that the first cleaning container and the floor cleaning tool intersect;

[0052] And / or,

[0053] Simultaneously recognizing the first placement board and the second type of food ingredient in the third detection area and the first placement board and the second type of food ingredient intersecting includes:

[0054] Obtain the recognition frame of the first placement board in the third detection area;

[0055] Obtain the recognition frame of the second type of food ingredients in the third detection area;

[0056] Determine whether there is an overlapping area between the recognition frame of the first placement board and the recognition frame of the second type of food ingredients according to the position coordinates of the recognition frame of the first placement board in the target image and the position coordinates of the recognition frame of the second type of food ingredients in the target image;

[0057] If the area ratio of the overlapping area to the recognition frame of the first placement board is greater than or equal to a preset second ratio threshold, determine that the first placement board and the second type of food ingredients intersect;

[0058] And / or

[0059] If the second placement board and the first type of food ingredients are simultaneously recognized in the fourth detection area, and the second placement board and the first type of food ingredients intersect, including:

[0060] Obtain the recognition frame of the second placement board in the fourth detection area;

[0061] Obtain the recognition frame of the first type of food ingredients in the fourth detection area;

[0062] Determine whether there is an overlapping area between the recognition frame of the second placement board and the recognition frame of the first type of food ingredients according to the position coordinates of the recognition frame of the second placement board in the target image and the position coordinates of the recognition frame of the first type of food ingredients in the target image;

[0063] If the area ratio of the overlapping area to the recognition frame of the second placement board is greater than or equal to a preset third ratio threshold, determine that the second placement board and the first type of food ingredients intersect;

[0064] And / or

[0065] The food processing board is configured with a first color, and the cleaning cloth is configured with a second color, and the first color is different from the second color. Among them, if the food processing board and the cleaning cloth are simultaneously recognized in the fifth detection area, and the food processing board and the cleaning cloth intersect, output a fifth warning instruction, including:

[0066] Perform color recognition on the target object in the fifth detection area, recognize the target object of the first color as the food processing board, and recognize the target object of the second color as the cleaning cloth;

[0067] If the food processing board and the cleaning cloth intersect, output a fifth warning instruction.

[0068] In a second aspect, the present application provides a device for identifying abnormal operations in the back kitchen. The device includes:

[0069] An acquisition module, configured to acquire a target image of a target environment area in the back kitchen, where the target environment area includes at least one detection area;

[0070] An identification module, configured to identify a first item and a second item in the detection area, and determine whether the first item and the second item intersect according to the position information of the first item and the second item; wherein, the first item is a tool for processing food materials, the second item is a movable item, and the second item is prohibited from being placed on the first item;

[0071] An alarm module, configured to output an alarm instruction if the first item intersects the second item.

[0072] In a third aspect, the present application provides a system for identifying abnormal operations in the back kitchen, the system including:

[0073] An image acquisition device, configured to acquire a target image of a target environment area in the back kitchen, where the target environment area includes at least one detection area, identify a first item and a second item in the detection area, determine whether the first item and the second item intersect according to the position information of the first item and the second item, and if they intersect, output an identification result; wherein, the first item is a tool for processing food materials, the second item is a movable item, and the second item is prohibited from being placed on the first item;

[0074] A server, configured to output an alarm instruction in response to the identification result.

[0075] In a fourth aspect, the present application provides a computer device, including a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, the steps of the method for identifying abnormal operations in the back kitchen as described above are implemented.

[0076] In a fifth aspect, the present application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method for identifying abnormal operations in the back kitchen as described above are implemented.

[0077] By acquiring a target image of a target environment area in the back kitchen, identifying the detection area in the target image, identifying the first item and the second item in the detection area, where the first item is a tool for processing food materials, the second item is a movable item, and the second item is prohibited from being placed on the first item, and determining whether the first item and the second item intersect, if they intersect, it is determined that there is a violation operation behavior of the back kitchen staff, such as washing a mop in the sink for washing food materials, thereby solving the problem of difficult management of the violation operation behavior of the back kitchen staff, being able to automatically and effectively identify the violation operation behavior, timely discover the violation operation behavior of the staff, improving the efficiency of store management, and reducing the management difficulty and management cost of the staff. Description of the Drawings

[0078] Figure 1 The first process schematic diagram of the method for identifying abnormal operations in the back kitchen provided by the embodiments of the present application;

[0079] Figure 2 The second process schematic diagram of the method for identifying abnormal operations in the back kitchen provided by the embodiments of the present application;

[0080] Figure 3 The third process schematic diagram of the method for identifying abnormal operations in the back kitchen provided by the embodiments of the present application;

[0081] Figure 4 The fourth process schematic diagram of the method for identifying abnormal operations in the back kitchen provided by the embodiments of the present application;

[0082] Figure 5 The fifth process schematic diagram of the method for identifying abnormal operations in the back kitchen provided by the embodiments of the present application;

[0083] Figure 6 The sixth process schematic diagram of the method for identifying abnormal operations in the back kitchen provided by the embodiments of the present application;

[0084] Figure 7 The seventh process schematic diagram of the method for identifying abnormal operations in the back kitchen provided by the embodiments of the present application;

[0085] Figure 8 The eighth process schematic diagram of the method for identifying abnormal operations in the back kitchen provided by the embodiments of the present application;

[0086] Figure 9 The ninth process schematic diagram of the method for identifying abnormal operations in the back kitchen provided by the embodiments of the present application;

[0087] Figure 10 The tenth process schematic diagram of the method for identifying abnormal operations in the back kitchen provided by the embodiments of the present application;

[0088] Figure 11 The eleventh process schematic diagram of the method for identifying abnormal operations in the back kitchen provided by the embodiments of the present application;

[0089] Figure 12 The twelfth process schematic diagram of the method for identifying abnormal operations in the back kitchen provided by the embodiments of the present application;

[0090] Figure 13 The system block diagram of the device for identifying abnormal operations in the back kitchen provided by the embodiments of the present application;

[0091] Figure 14 The system block diagram of the system for identifying abnormal operations in the back kitchen provided by the embodiments of the present application;

[0092] Figure 15It is a system block diagram of the computer device provided by the embodiment of the present application. Detailed implementation manners

[0093] The present application will be described in detail below in conjunction with the specific implementation manners shown in the accompanying drawings. However, these implementation manners do not limit the present application, and any structural, method, or functional transformation made by those of ordinary skill in the art based on these implementation manners is included in the protection scope of the present application.

[0094] The embodiment of the present application provides a method for identifying abnormal operations in the back kitchen. The execution subject of this method can be a camera or a camera, and the camera or camera can be connected to a server. A processor, a transceiver, and a camera device can be set in the camera or camera. Among them, the camera device can be used to capture images, that is, target images can be obtained. The processor identifies the obtained target images to identify whether a first item and a second item intersect. The transceiver can be used to perform data transmission with the server. That is, when it is determined that the first item and the second item intersect, the recognition result is output to the server so that the server outputs an alarm instruction.

[0095] In the embodiment of the present application, the execution subject of the method for identifying abnormal operations in the back kitchen can be an electronic device. The camera is connected to the electronic device. The camera is used to capture a target image. The electronic device obtains the target image captured by the camera for image processing to identify whether a first item and a second item intersect, and when it is determined that the first item and the second item intersect, an alarm instruction is output.

[0096] Exemplarily, the electronic device can be a device such as a terminal, including but not limited to a mobile terminal and a fixed terminal. For example, the mobile terminal includes but not limited to a smart phone, a smart watch, a tablet computer, a laptop computer, a smart vehicle, a smart in-vehicle device, etc. Among them, the fixed terminal includes but not limited to a desktop computer, a smart TV, etc.

[0097] Exemplarily, the electronic device can also be a device such as a server. The server can be an independent physical server, or a server cluster or a distributed system composed of multiple physical servers. It can also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms, but is not limited thereto.

[0098] The implementation manner of the method for identifying abnormal operations in the back kitchen according to the embodiment of the present application is described in detail below.

[0099] Please refer to Figure 1 , the embodiment of the present application provides a method for identifying abnormal operations in the back kitchen, and this method includes steps S101-S103.

[0100] Step S101, obtain a target image of a target environment area in the back kitchen, where the target environment area includes at least one detection area.

[0101] Exemplarily, the target environment area may be an environment area inside the back kitchen of a restaurant.

[0102] Exemplarily, the detection area may be an area where a violation operation behavior occurs. For example, the detection area includes the area of the sink for washing ingredients. For another example, the detection area includes the area of the ingredient processing board.

[0103] Exemplarily, collect and obtain the target image corresponding to the target environment area through a camera or a camera installed in the back kitchen. The target image can be a real-time captured image, or an image frame extracted from video recording, or an image captured at periodic intervals. The installation position of the camera can be determined according to the site conditions of the actual use place so as to be able to completely capture the behaviors of the staff in the back kitchen. This application does not make any limitations in this regard.

[0104] Based on the obtained target image, determine the detection area in the target image. Exemplarily, based on the environmental layout in the back kitchen, the target environment area can be divided into areas to obtain the required detection areas. The general installation position of the camera is fixed. Based on the pre-divided areas, the detection area can be directly determined in the obtained target image. The detection area in the camera's view can also be determined through image segmentation, edge detection, etc. The detection area can also be defined manually in the camera to generate the coordinates of the detection area. Through the coordinates of each pre-generated detection area, the detection area in the target image can be located simply and efficiently, which is beneficial to accurately and effectively identify the target object in the detection area.

[0105] Step S102, identify a first item and a second item in the detection area, and determine whether the first item and the second item intersect according to the position information of the first item and the second item; where the first item is a tool for processing ingredients, the second item is a movable item, and the second item is prohibited from being placed on the first item.

[0106] Perform target object recognition on the detection area to identify the first item and the second item in the detection area. The first item is a tool for processing ingredients, and the second item is a movable item. Exemplarily, computer vision algorithms or deep learning algorithms can be used to perform image recognition on the detection area to identify the first item and the second item in the image. According to the position coordinates of the first item in the target image and the position coordinates of the second item in the target image, determine whether the first item and the second item intersect.

[0107] The first item is a tool for processing food ingredients, and the second item is a movable item. The second item is prohibited from being placed on the first item. Exemplarily, the first item is a sink for washing food ingredients, and the second item is a floor cleaning tool, such as a broom, a mop, a dustpan, etc. In the hygiene operation specifications of the back kitchen, the floor cleaning tool is prohibited from being placed in the sink. Exemplarily, the first item is a meat ingredient processing board, and the second item is a fruit ingredient or a vegetable ingredient. The fruit or vegetable ingredient is prohibited from using the meat ingredient processing board.

[0108] Step S103, if the first item and the second item intersect, output an alarm instruction.

[0109] If it is determined that the first item and the second item intersect, it indicates that the staff has performed a violation operation. Output an alarm instruction, which is used to indicate that there is a violation operation by the staff in the back kitchen.

[0110] In this embodiment, by acquiring the target image of the target environmental area in the back kitchen, identifying the detection area in the target image, identifying the first item and the second item in the detection area, and determining whether the first item and the second item intersect, it is determined that the staff has a violation operation behavior. For example, washing the mop in the sink for washing food ingredients, or placing fruits or vegetables on the meat ingredient processing board, etc. Thus, the problem of difficult management of the violation operation behavior of the staff in the back kitchen is solved, the management cost is reduced, the violation operation behavior can be automatically and effectively identified, the violation operation behavior of the staff can be discovered in time, the management efficiency of the store is improved, the management difficulty and management cost for the staff are reduced, and the supervision effect is improved.

[0111] In some embodiments, as Figure 2 shown, the first image acquisition device is configured to be able to identify the first item and the second item, the second image acquisition device is configured to be able to identify the first item and the second item, the installation positions and shooting angles of the first image acquisition device and the second image acquisition device are different, and the method further includes steps S201 - S203.

[0112] Step S201, the first image acquisition device acquires and obtains the first target image of the target environmental area, identifies the first item and the second item in the detection area of the first target image, and determines whether the first item and the second item intersect.

[0113] In order to improve the accuracy of identifying illegal operation behaviors in the back kitchen, a first image acquisition device and a second image acquisition device are installed in the back kitchen. The installation positions and shooting angles of the first image acquisition device and the second image acquisition device are different, so as to shoot the target environmental area from different angles and obtain target images with different shooting angles. The first image acquisition device can be a camera or a camcorder, and the second image acquisition device can be a camera or a camcorder. The camera can be an edge camera with image recognition function, and send the recognition output result to the server, which can effectively reduce the data traffic transmitted from the camera to the server.

[0114] The first image acquisition device acquires and obtains the first target image of the target environmental area, determines the detection area in the first target image, performs image recognition on the detection area, recognizes the first item and the second item in the detection area, and determines whether the first item and the second item intersect based on the position coordinates of the first item in the first target image and based on the position coordinates of the second item in the first target image.

[0115] Step S202: The second image acquisition device acquires and obtains the second target image of the target environmental area, recognizes the first item and the second item in the detection area of the second target image, and determines whether the first item and the second item intersect.

[0116] The second image acquisition device acquires and obtains the second target image of the target environmental area, determines the detection area in the second target image, performs image recognition on the detection area, recognizes the first item and the second item in the detection area, and determines whether the first item and the second item intersect based on the position coordinates of the first item in the second target image and based on the position coordinates of the second item in the second target image. It should be noted that the detection area in the first target image is the same as the detection area in the second target image.

[0117] Step S203: When the first image acquisition device determines that the first item and the second item intersect, and the second image acquisition device determines that the first item and the second item intersect, an alarm instruction is output.

[0118] When the first image acquisition device determines that the first item and the second item intersect, and the second image acquisition device determines that the first item and the second item intersect, it indicates that there is an illegal operation behavior of the staff, and an alarm instruction is output.

[0119] In this embodiment, by installing two image acquisition devices, the target environmental area is photographed from different shooting angles respectively, item recognition is performed on the same detection area in the respectively obtained target images, and when it is recognized that both the first item and the second item intersect, it is determined that there is an illegal operation behavior of the staff, so as to improve the accuracy of identifying illegal operation behaviors, reduce the false alarm rate of recognition errors, and further improve the efficiency of store management.

[0120] In some embodiments, the first image acquisition device is a first camera, and the second image acquisition device is a second camera. The method further includes: when the first camera does not determine that the first item and the second item intersect, the second camera captures second target images at a first time interval; when the first camera determines that the first item and the second item intersect, the second camera captures second target images at a second time interval, where the second time interval is less than the first time interval.

[0121] When the first camera does not determine that the first item and the second item intersect, it indicates that there is no current violation operation behavior. At this time, there is no need to identify whether there is a violation operation behavior. Therefore, the second camera captures second target images at a relatively long first time interval to reduce device consumption. When the first camera determines that the first item and the second item intersect, it means that there is a current violation operation behavior, which belongs to the peak time period of the occurrence of violation operation behaviors. Therefore, the second camera captures second target images at a relatively short second time interval to ensure that the complete behavior monitoring process can be accurately monitored and the recognition accuracy can be improved. The first time interval and the second time interval can be set according to empirical values.

[0122] In some embodiments, if the first image acquisition device determines that the first item and the second item intersect, and the second image acquisition device determines that the first item and the second item intersect, an alarm instruction is output, including: when the first image acquisition device determines that the first item and the second item intersect, it outputs a first recognition result to the server; when the second image acquisition device determines that the first item and the second item intersect, it outputs a second recognition result to the server; the server outputs an alarm instruction in response to the first recognition result and the second recognition result. The first recognition result is used to indicate that the first item and the second item intersect, and the second recognition result is used to indicate that the first item and the second item intersect. In this embodiment, when the server receives the first recognition result sent by the first image acquisition device and simultaneously receives the second recognition result sent by the second image acquisition device, it indicates that the image acquisition device has detected a violation operation behavior, and outputs an alarm instruction to remind the staff that there is a violation operation, reducing the false alarm rate of the alarm information.

[0123] In some embodiments, as Figure 3 shown, the method further includes the steps:

[0124] Step S301: Obtain the false alarm information feedback by the user for the alarm instruction within a preset time period, determine the movable items with a false alarm rate higher than the preset false alarm rate threshold, and divide the movable items into at least a first part and a second part. Among them, in the false alarm information feedback by the user: the first part is determined that the number of intersections with the first item is greater than the first threshold, or the proportion of the number of images in which the first part is determined to intersect with the first item in the number of images of the false alarm feedback by the user when the first item intersects with the second item is greater than the second threshold.

[0125] Step S302: If it is determined that the first item intersects with the first part of the movable item, do not output the alarm instruction.

[0126] In one application scenario of this embodiment, the second item is a movable item. When the staff uses this movable item, they may accidentally touch the first item, resulting in a false alarm of the alarm information. For example, the movable item is a mop. When the staff is mopping the floor, the top of the mop handle may often touch the sink, resulting in being determined to intersect with the sink and prone to false alarms. Therefore, for items prone to false alarms, the items are segmented and identified to reduce the false alarm rate.

[0127] Collect the false alarm information feedback by the user for the alarm instruction, obtain the false alarm information within a preset time period, analyze the obtained false alarm information, calculate the false alarm rate of each movable item, and thus determine the movable items with a false alarm rate higher than the preset false alarm rate threshold. The preset time period can be set according to the actual situation. The false alarm rate threshold can be set according to experience values.

[0128] Based on the false alarm information feedback by the user, divide the movable item into at least a first part and a second part. The first part is determined that the number of intersections with the second item is greater than the first threshold, or the proportion of the number of images in which the first part is determined to intersect with the second item in the number of images of the false alarm feedback by the user when the first item intersects with the second item is greater than the second threshold.

[0129] Exemplarily, in the feedback false alarm information, if the number of intersections of the first part of the movable item with the first item is greater than the first threshold, it indicates that the first part of the movable item is prone to false alarms. In the subsequent processing, when it is determined that the first item intersects with the first part of the movable item, do not output the alarm instruction. The setting of the first threshold can be set according to experience values.

[0130] Exemplarily, obtain the number of false alarm images in which the first item intersects with the first item for which user feedback is obtained. Among these false alarm images, if the proportion of the number of images in which the first part of the movable item is determined to intersect with the first item in the number of false alarm images is greater than the second threshold, it indicates that the first part of the movable item is prone to false alarms. In subsequent processing, when it is determined that the first item intersects with the first part of the movable item, no alarm instruction is output. The setting of the second threshold can be set according to empirical values.

[0131] In this embodiment, for the movable item that is prone to false alarms, segmented or partial recognition is performed to determine a part of the movable item that is prone to false alarms. When this part intersects with the first item, no alarm is generated. For example, for the above-mentioned mop, the top of the mop rod is prone to false alarms. The mop is divided into a mop rod and a mop head. No alarm is generated when the mop rod intersects with the sink, and an alarm is generated only when the mop head intersects with the sink. Thereby, the correctness of identifying illegal operation behaviors can be improved, the false alarm rate can be reduced, and the efficiency of store management can be further enhanced.

[0132] In some embodiments, as Figure 4 shown, the method further includes the steps:

[0133] Step S401: Obtain a plurality of specific false alarm images, where the specific false alarm images are false alarm images in which the first part is determined to intersect with the first item;

[0134] Step S402: Mark the intersection positions of the first part and the first item in the plurality of specific false alarm images respectively;

[0135] Step S403: According to the marked intersection positions of the first part and the first item, divide the first part into at least a third sub-part and a fourth sub-part, where the number of times the third sub-part is determined to intersect with the first item is greater than the third threshold, or the proportion of the number of images in which the third sub-part is determined to intersect with the first item in the number of specific false alarm images is greater than the fourth threshold;

[0136] Step S404: If in the obtained target image, it is determined that the third sub-part intersects with the first item, no alarm instruction is output;

[0137] Step S405: If in the obtained target image, it is determined that the fourth sub-part intersects with the first item, an alarm instruction is output.

[0138] Obtain multiple specific false alarm images, where the specific false alarm images are false alarm images in which the first part is determined to intersect with the second part. Perform image analysis on the specific false alarm images to more accurately identify the position where the first part of the movable item intersects with the first item. Label the specific false alarm images, and use manual labeling or automatic labeling to label the intersection positions. According to the labeled intersection positions, use manual division or automatic division methods to divide the first part into at least a third sub-part and a fourth sub-part. Perform image recognition on the obtained target image. If it is recognized that the third sub-part in the detection area intersects with the first item, no alarm instruction is output. Perform image recognition on the obtained target image. If it is recognized that the fourth sub-part in the detection area intersects with the first item, an alarm instruction is output.

[0139] Exemplarily, based on the labeled position information, if the number of times the third sub-part of the movable item intersects with the first item is greater than a third threshold, it indicates that the third sub-part of the movable item is prone to false alarms. In subsequent processing, when it is determined that the first item intersects with the third sub-part of the movable item, no alarm instruction is output. The setting of the third threshold can be determined according to empirical values.

[0140] Exemplarily, obtain the number of specific false alarm images. Among these specific false alarm images, if the proportion of the number of images in which the third part of the movable item is determined to intersect with the first item in the total number of specific false alarm images is greater than a fourth threshold, it indicates that the third sub-part of the movable item is prone to false alarms. In subsequent processing, when it is determined that the first item intersects with the third sub-part of the movable item, no alarm instruction is output. The setting of the fourth threshold can be determined according to empirical values.

[0141] In this embodiment, by labeling the intersection positions of the movable item and the first item in the false alarm image, the first part of the movable item that is prone to false alarms is further segmented or divided for recognition. When the third sub-part intersects with the first item, no alarm is generated. When the fourth sub-part intersects with the first item, an alarm is generated. For example, the above-mentioned mop rod is divided into two parts. The part from the top of the mop rod to 1 / 3 of the mop rod is the third sub-part, and the remaining part of the mop rod is the fourth sub-part. Thus, it is possible to more accurately identify illegal operation behaviors, reduce both false alarms and missed alarms.

[0142] In some embodiments, as Figure 5 shown, the first item is the first cleaning container for cleaning food ingredients, and the second item is the floor cleaning tool. The method further includes the steps:

[0143] Step S501, determine the first detection area in the target image, and determine whether the first cleaning container and the floor cleaning tool exist simultaneously in the first detection area, where the first detection area includes at least one first cleaning container;

[0144] Step S502, if both the first cleaning container and the floor cleaning tool exist in the first detection area, determine whether the first cleaning container and the floor cleaning tool intersect, or whether the first cleaning container contains the floor cleaning tool. If they intersect or the first cleaning container contains the floor cleaning tool, output a first warning instruction.

[0145] In one application scenario of this embodiment, the first item is the first cleaning container for cleaning food ingredients. For example, the first cleaning container can be a sink or a pool. The second item is the floor cleaning tool. For example, the floor cleaning tool can be a mop or a broom. The floor cleaning tool is prohibited from being placed in the first cleaning container.

[0146] Obtain a target image of the target environment area in the back kitchen, determine the first detection area in the target image, perform image recognition on the first detection area to identify whether both the first cleaning container and the floor cleaning tool exist in the first detection area. The first detection area includes at least one cleaning container. A convolutional neural network for object detection can be used to identify the target objects in the image. If it is recognized that both the first cleaning container and the floor cleaning tool exist in the first detection area, based on the position coordinates of the first cleaning container in the target image and the position coordinates of the floor cleaning tool in the target image, determine whether the first cleaning container and the floor cleaning tool intersect, or whether the first cleaning container contains the floor cleaning tool. If they intersect or the first cleaning container contains the floor cleaning tool, it indicates that the floor cleaning tool is partially or fully placed in the first cleaning container, determine that there is a current violation operation behavior, and output a warning instruction to remind the staff.

[0147] In some embodiments, as Figure 6 shown, determining whether the first cleaning container and the floor cleaning tool intersect, or the first cleaning container contains the floor cleaning tool, includes the steps of:

[0148] Step S601, obtain the recognition frame of the first cleaning container in the first detection area;

[0149] Step S602, obtain the recognition frame of the floor cleaning tool in the first detection area;

[0150] Step S603, according to the position coordinates of the recognition frame of the first cleaning container in the target image and the position coordinates of the recognition frame of the floor cleaning tool in the target image, determine whether there is an overlapping area between the recognition frame of the first cleaning container and the recognition frame of the floor cleaning tool;

[0151] Step S604, if the area ratio of the overlapping area to the recognition frame of the first cleaning container is greater than or equal to a preset first ratio threshold, determine that the first cleaning container contains the floor cleaning tool, otherwise determine that the first cleaning container and the floor cleaning tool intersect.

[0152] Exemplarily, a convolutional neural network for object detection, such as the YOLO network, can be used to detect objects in the first detection area to obtain rectangular frames of object targets. Each object corresponds to a rectangular frame of an object target. Based on the rectangular frame of the object target, the corresponding target object is determined, that is, the recognition frame of the first cleaning container and the recognition frame of the floor cleaning tool in the first detection area are determined. Alternatively, other object detection algorithms can also be used to determine the target object in the image. The specific detection method is not limited. According to the position coordinates of the recognition frame of the first cleaning container in the target image and the position coordinates of the recognition frame of the floor cleaning tool in the target image, it is determined whether there is an overlapping area between the recognition frame of the first cleaning container and the recognition frame of the floor cleaning tool. If the area ratio of the overlapping area to the area of the recognition frame of the first cleaning container is greater than or equal to the first ratio threshold, it is determined that the first cleaning container contains the floor cleaning tool; otherwise, it is determined that the first cleaning container and the floor cleaning tool intersect. Based on this embodiment, it is determined whether the floor cleaning tool is placed in the first cleaning container to provide a judgment basis for subsequent processing. The first ratio threshold can be set according to empirical values.

[0153] In some embodiments, as Figure 7 shown, the first item includes a second cleaning container and a third cleaning container, the second item includes tableware and food ingredients, the second cleaning container is used to clean tableware, and the third cleaning container is used to clean food ingredients. The method further includes the steps:

[0154] Step S701, determining a second detection area in the target image, where the second detection area includes the second cleaning container and the third cleaning container;

[0155] Step S702, if it is recognized that the second cleaning container intersects with the food ingredients in the second detection area, or the third cleaning container intersects with the tableware, output a second warning instruction.

[0156] In one application scenario of this embodiment, the first item is a second cleaning container and a third cleaning container, the second item includes tableware and food ingredients, the second cleaning container is used to clean tableware, the third cleaning container is used to clean food ingredients, the second cleaning container is prohibited from cleaning food ingredients, and the third cleaning container is prohibited from cleaning tableware, so as to separate the containers for cleaning tableware and cleaning food ingredients. Based on this embodiment, when the cleaning container for cleaning tableware cleans food ingredients, or the cleaning container for cleaning food ingredients cleans tableware, it is recognized that there is a current illegal operation behavior and a warning message is output. A convolutional neural network for object detection can be used to recognize the second cleaning container and the third cleaning container in the second detection area, as well as the food ingredients and tableware. Details are not described herein again.

[0157] In some embodiments, as Figure 8As shown, the first article includes a first placement board, the second article includes a first type of food ingredient and a second type of food ingredient, and the first placement board is used for processing the first type of food ingredient. The method further includes the steps:

[0158] Step S801: Determine a third detection area in the target image, where the third detection area at least includes the first placement board;

[0159] Step S802: If both the first placement board and the second type of food ingredient are recognized in the third detection area and the first placement board and the second type of food ingredient intersect, output a third warning instruction.

[0160] In one application scenario of this embodiment, the first article is the first placement board, the second article includes a first type of food ingredient and a second type of food ingredient, and the first placement board is used for processing the first type of food ingredient. For example, the first placement board is a meat processing board, the first food ingredient is meat or aquatic products, and the second type of food ingredient is fruit or vegetables. Fruit or vegetables are prohibited from being processed on the meat processing board, and the boards for food ingredient processing need to be separated. Based on this implementation, when the first placement board for processing the first type of food ingredient processes the second type of food ingredient, it is recognized that there is an illegal operation behavior currently, and a warning message is output.

[0161] In some embodiments, as Figure 9 shown, recognizing both the first placement board and the second type of food ingredient in the third detection area and the first placement board and the second type of food ingredient intersecting includes the steps:

[0162] Step S901: Obtain the recognition frame of the first placement board in the third detection area;

[0163] Step S902: Obtain the recognition frame of the second type of food ingredient in the third detection area;

[0164] Step S903: According to the position coordinates of the recognition frame of the first placement board in the target image and the position coordinates of the recognition frame of the second type of food ingredient in the target image, determine whether there is an overlapping area between the recognition frame of the first placement board and the recognition frame of the second type of food ingredient;

[0165] Step S904: If the area ratio of the overlapping area to the recognition frame of the first placement board is greater than or equal to a preset second ratio threshold, determine that the first placement board and the second type of food ingredient intersect.

[0166] Exemplarily, a convolutional neural network for object detection can be employed to detect objects in the third detection area, and identify the recognition frames of the first placement board and the second type of food ingredients in the third detection area. Based on the position coordinates of the recognition frame of the first placement board in the target image and the position coordinates of the recognition frame of the second type of food ingredients in the target image, it is determined whether there is an overlapping area between the recognition frame of the first placement board and the recognition frame of the second type of food ingredients. If the area ratio of the overlapping area to the recognition frame of the first placement board is greater than or equal to the second ratio threshold, it is determined that the first placement board and the second type of food ingredients intersect. Based on this embodiment, it is determined whether the second type of food ingredients is placed on the first placement board, providing a basis for subsequent processing. The second ratio threshold can be set according to empirical values.

[0167] In some embodiments, as Figure 10 shown, the first object includes a second placement board, the second object includes the first type of food ingredients and the second type of food ingredients, and the second placement board is used for processing the second type of food ingredients. The method further includes the steps:

[0168] Step S1001: Determine the fourth detection area in the target image, where the fourth detection area at least includes the second placement board;

[0169] Step S1002: If both the second placement board and the first type of food ingredients are recognized in the fourth detection area and the second placement board and the first type of food ingredients intersect, output a fourth warning instruction.

[0170] In one application scenario of this embodiment, the first object is the second placement board, the second object includes the first type of food ingredients and the second type of food ingredients, and the second placement board is used for processing the second type of food ingredients. For example, the second placement board is a vegetable processing board, the first food ingredient is meat or aquatic products, and the second type of food ingredients is fruits or vegetables. It is prohibited to process food ingredients on the vegetable processing board, and the boards for food ingredient processing need to be separated. Based on this embodiment, when the second placement board for processing the second type of food ingredients processes the first type of food ingredients, it is recognized that there is a current illegal operation behavior and a warning message is output.

[0171] In some embodiments, as Figure 11 shown, if both the second placement board and the first type of food ingredients are recognized in the fourth detection area and the second placement board and the first type of food ingredients intersect, it includes the steps:

[0172] Step S1101: Obtain the recognition frame of the second placement board in the fourth detection area;

[0173] Step S1102: Obtain the recognition frame of the first type of food ingredients in the fourth detection area;

[0174] Step S1103: Determine whether there is an overlapping area between the recognition frame of the second placement board and the recognition frame of the first type of food ingredients according to the position coordinates of the recognition frame of the second placement board in the target image and the position coordinates of the recognition frame of the first type of food ingredients in the target image.

[0175] Step S1104: If the area ratio of the overlapping area to the recognition frame of the second placement board is greater than or equal to a preset third ratio threshold, determine that the second placement board and the first type of food ingredients intersect.

[0176] Exemplarily, a convolutional neural network for object detection can be used to detect objects in the fourth detection area to determine the recognition frame of the second placement board and the recognition frame of the first type of food ingredients in the fourth detection area. According to the position coordinates of the recognition frame of the second placement board in the target image and the position coordinates of the recognition frame of the first type of food ingredients in the target image, determine whether there is an overlapping area between the recognition frame of the second placement board and the recognition frame of the first type of food ingredients. If the area ratio of the overlapping area to the recognition frame of the second placement board is greater than or equal to the third ratio threshold, it is determined that the second placement board and the first type of food ingredients intersect. Based on this embodiment, it is determined whether the first type of food ingredients is placed on the second placement board to provide a judgment basis for subsequent processing. The third ratio threshold can be set according to empirical values. The third detection area and the fourth detection area can be the same or different.

[0177] In some embodiments, as Figure 12 shown, the first item is a food ingredient processing board and the second item is a cleaning cloth. The method further includes the steps of:

[0178] Step S1201: Determine the fifth detection area in the target image, where the fifth detection area includes the food ingredient processing board.

[0179] Step S1202: If both the food ingredient processing board and the cleaning cloth are recognized in the fifth detection area and the food ingredient processing board and the cleaning cloth intersect, output a fifth warning instruction.

[0180] In one application scenario of this embodiment, the first item is a food ingredient processing board and the second item includes a cleaning cloth. It is prohibited to place the cleaning cloth on the food ingredient processing board. Based on this embodiment, when it is recognized that the cleaning cloth is placed on the food ingredient processing board, it is determined that there is a violation operation behavior currently, and a warning message is output.

[0181] In some embodiments, the food processing board is configured with a first color, and the cleaning cloth is configured with a second color, where the first color is different from the second color. If both the food processing board and the cleaning cloth are simultaneously recognized in the fifth detection area and they intersect, a fifth warning instruction is output, including: performing color recognition on the target objects in the fifth detection area, identifying the target object with the first color as the food processing board, and identifying the target object with the second color as the cleaning cloth; if the food processing board and the cleaning cloth intersect, output a fifth warning instruction.

[0182] In this embodiment, by configuring different colors for the food processing board and the cleaning cloth, it is possible to identify whether the cleaning cloth is placed on the food processing board, providing a basis for subsequent processing.

[0183] Based on the same inventive concept, an embodiment of the present application further provides a device for implementing the above-mentioned identification of abnormal operations in the back kitchen. The solution provided by this device to solve the problem is similar to the solution described in the above method. Therefore, the specific limitations in the embodiment of the device for identifying abnormal operations in the back kitchen provided below can refer to the limitations on the method for identifying abnormal operations in the back kitchen above, and will not be repeated here.

[0184] Please refer to Figure 13 , the present application provides a device for identifying abnormal operations in the back kitchen, and this device includes:

[0185] An acquisition module 1301, configured to acquire a target image of a target environment area in the back kitchen, where the target environment area includes at least one detection area;

[0186] An identification module 1302, configured to identify a first item and a second item in the detection area, and determine whether the first item and the second item intersect according to the position information of the first item and the second item; where the first item is a tool for processing food, and the second item is a movable item, and the second item is prohibited from being placed on the first item;

[0187] An alarm module 1303, configured to output an alarm instruction if the first item intersects with the second item.

[0188] In this embodiment, by acquiring the target image of the target environment area in the back kitchen, identifying the detection area in the target image, identifying the first item and the second item in the detection area, and determining whether the first item and the second item intersect, it is determined that the staff has a violation operation behavior, thereby solving the problem of difficult management of the violation operation behavior of the staff in the back kitchen, reducing the management cost, being able to automatically and effectively identify the violation operation behavior, timely discover the violation operation behavior of the staff, improving the efficiency of store management, reducing the management difficulty and management cost of the staff, and improving the supervision effect.

[0189] Please refer toFigure 14 , this application provides a system for identifying abnormal operations in the back kitchen. The system includes:

[0190] An image acquisition device 1401, configured to obtain a target image of a target environment area in the back kitchen. The target environment area includes at least one detection area. Identify a first item and a second item in the detection area, and determine whether the first item and the second item intersect according to the position information of the first item and the second item. If they intersect, output an identification result; wherein, the first item is a tool for processing food ingredients, the second item is a movable item, and the second item is prohibited from being placed on the first item;

[0191] A server 1402, configured to output an alarm instruction in response to the identification result.

[0192] An embodiment of this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements any of the above methods for identifying abnormal operations in the back kitchen.

[0193] Figure 15 It is a schematic diagram of the hardware structure of a computer device provided by an embodiment of this application. Figure 15 The illustrated computer device includes: a processor 1501, a communication interface 1502, a memory 1503, and a communication bus 1504. The processor 1501, the communication interface 1502, and the memory 1503 complete mutual communication through the communication bus 1504. Among them, Figure 15 The connection manners among the illustrated processor 1501, communication interface 1502, and memory 1503 are merely exemplary. During implementation, the processor 1501, communication interface 1502, and memory 1503 may also communicate and connect with each other using other connection manners besides the communication bus 1504.

[0194] The memory 1503 may be used to store a computer program. The computer program may include instructions and data, and implement the steps of any of the above methods for identifying abnormal operations in the back kitchen. In an embodiment of this application, the memory 1503 may be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), flash memory, optical memory, and registers, etc. The memory 1503 may include a hard disk and / or memory.

[0195] The processor 1501 can be a general-purpose processor, which can be a processor that executes specific steps and / or operations by reading and executing a computer program (such as a computer program) stored in a memory (such as the memory 1503). The general-purpose processor may use data stored in the memory (such as the data in the memory 1503) during the execution of the steps and / or operations. The general-purpose processor can be, for example but not limited to, a central processing unit (CPU). In addition, the processor 1501 can also be a dedicated processor, which can be a processor specifically designed to execute specific steps and / or operations. The dedicated processor can be, for example but not limited to, ASIC and FPGA, etc. In addition, the processor 1501 can also be a combination of multiple processors, such as a multi-core processor.

[0196] The communication interface 1502 can include interfaces such as input / output (I / O) interfaces, physical interfaces, and logical interfaces for implementing the interconnection of components inside the network device, as well as interfaces for implementing the interconnection between the network device and other devices (such as network devices). The communication network can be Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc. The communication interface 1502 can be a module, a circuit, a transceiver, or any device capable of implementing communication.

[0197] In the implementation process, each step of the above method can be completed by the integrated logic circuit in the hardware of the processor 1501 or by instructions in software form. The method for identifying abnormal operations in the back kitchen according to the embodiments of the present application can be directly embodied as being executed and completed by the hardware processor, or by a combination of the hardware and software modules in the processor. The software module can be located in a mature storage medium in the art such as random access memory flash, read-only memory, programmable read-only memory, or electrically erasable programmable memory, registers, etc. This storage medium is located in the memory 1503, and the processor 1501 reads the information in the memory 1503 and combines its hardware to complete the steps of the above method. To avoid repetition, it will not be described in detail here.

[0198] Although the preferred embodiments of the present application have been disclosed for illustrative purposes, those of ordinary skill in the art will realize that various improvements, additions, and substitutions are possible without departing from the scope and spirit of the present application disclosed by the appended claims.

Claims

1. A method for identifying abnormal operations in the back kitchen, characterized in that, The method includes: Obtaining a target image of a target environment area in the back kitchen, where the target environment area includes at least one detection area; Identifying a first item and a second item in the detection area, and determining whether the first item and the second item intersect according to the position information of the first item and the second item; wherein, the first item is a tool for processing food ingredients, and the second item is a movable item, and the second item is prohibited from being placed on the first item; If the first item intersects with the second item, output an alarm instruction; Wherein, the method further includes: Obtaining false alarm information fed back by a user for the alarm instruction within a preset time period, determining a movable item with a false alarm rate higher than a preset false alarm rate threshold, and dividing the movable item into at least a first part and a second part, wherein, in the false alarm information fed back by the user: the first part is determined to have a number of intersections with the first item greater than a first threshold, or, the proportion of the number of images in which the first part is determined to intersect with the first item in the number of images of the false alarm in which the user feeds back the intersection of the first item and the second item is greater than a second threshold; If it is determined that the first item intersects with the first part of the movable item, do not output an alarm instruction; The method further includes: Obtaining a plurality of specific false alarm images, where the specific false alarm images are false alarm images in which the first part is determined to intersect with the first item; Marking the intersection positions of the first part and the first item in the plurality of specific false alarm images; According to the marked intersection positions of the first part and the first item, dividing the first part into at least a third sub-part and a fourth sub-part, wherein, the third sub-part is determined to have a number of intersections with the first item greater than a third threshold, or, the proportion of the number of images in which the third sub-part is determined to intersect with the first item in the number of specific false alarm images is greater than a fourth threshold; If it is determined that the third sub-part intersects with the first item in the obtained target image, do not output an alarm instruction; If it is determined that the fourth sub-part intersects with the first item in the obtained target image, output an alarm instruction.

2. The method for identifying abnormal operations in the back kitchen according to claim 1, wherein A first image acquisition device is configured to be able to identify the first item and the second item, a second image acquisition device is configured to be able to identify the first item and the second item, and the installation positions and shooting angles of the first image acquisition device and the second image acquisition device are different. The method further includes: The first image acquisition device acquires and obtains a first target image of the target environment area, identifies the first item and the second item in the detection area of the first target image, and determines whether the first item and the second item intersect; The second image acquisition device acquires and obtains a second target image of the target environment area, identifies the first item and the second item in the detection area of the second target image, and determines whether the first item and the second item intersect; If the first image acquisition device determines that the first item and the second item intersect, and the second image acquisition device determines that the first item and the second item intersect, output the warning instruction.

3. The method for identifying abnormal operations in the back kitchen according to claim 2, wherein The first image acquisition device is a first camera, and the second image acquisition device is a second camera. The method further includes: When the first camera does not determine that the first item and the second item intersect, the second camera captures the second target image at a first time interval. When the first camera determines that the first item and the second item intersect, the second camera captures the second target image at a second time interval, where the second time interval is less than the first time interval.

4. The method for identifying abnormal operations in the back kitchen according to claim 2, wherein The step of "If the first image acquisition device determines that the first item and the second item intersect, and the second image acquisition device determines that the first item and the second item intersect, output the warning instruction" includes: The first image acquisition device determines that the first item and the second item intersect, and outputs a first recognition result to the server. The second image acquisition device determines that the first item and the second item intersect, and outputs a second recognition result to the server. The server outputs a warning instruction in response to the first recognition result and the second recognition result.

5. The method for identifying abnormal operations in the back kitchen according to claim 1, characterized in that, The first item is a first cleaning container for cleaning food ingredients, and the second item is a floor cleaning tool. The method further includes: Determine a first detection area in the target image, and determine whether the first cleaning container and the floor cleaning tool exist simultaneously in the first detection area, where the first detection area includes at least one first cleaning container. If the first cleaning container and the floor cleaning tool exist simultaneously in the first detection area, determine whether the first cleaning container and the floor cleaning tool intersect, or whether the first cleaning container contains the floor cleaning tool. If they intersect or contain, output a first warning instruction. and / or The first item includes a second cleaning container and a third cleaning container, the second item includes tableware and food ingredients, the second cleaning container is used for cleaning tableware, and the third cleaning container is used for cleaning food ingredients. The method further includes: Determine a second detection area in the target image, where the second detection area includes the second cleaning container and the third cleaning container. If it is recognized that the second cleaning container in the second detection area intersects with the food ingredients, or the third cleaning container intersects with the tableware, output a second warning instruction. and / or The first item includes a first placement board, the second item includes a first type of food ingredient and a second type of food ingredient, and the first placement board is used for processing the first type of food ingredient. The method further includes: Determine a third detection area in the target image, where the third detection area includes at least the first placement board. If the first placement board and the second type of food ingredient are simultaneously recognized in the third detection area, and the first placement board and the second type of food ingredient intersect, output a third warning instruction. and / or The first article includes a second placement board, the second article includes a first type of food ingredient and a second type of food ingredient, the second placement board is used for processing the second type of food ingredient, and the method further includes: Determine a fourth detection area in the target image, where the fourth detection area at least includes the second placement board; If both the second placement board and the first type of food ingredient are recognized in the fourth detection area, and the second placement board and the first type of food ingredient intersect, output a fourth warning instruction; and / or, The first article is a food processing board, the second article is a cleaning cloth, and the method further includes: Determine a fifth detection area in the target image, where the fifth detection area includes the food processing board; If both the food processing board and the cleaning cloth are recognized in the fifth detection area, and the food processing board and the cleaning cloth intersect, output a fifth warning instruction.

6. The method for identifying abnormal operations in the back kitchen according to claim 5, wherein, The determination of whether the first cleaning container and the floor cleaning tool intersect, or whether the first cleaning container contains the floor cleaning tool, includes: Obtain the recognition frame of the first cleaning container in the first detection area; Obtain the recognition frame of the floor cleaning tool in the first detection area; According to the position coordinates of the recognition frame of the first cleaning container in the target image and the position coordinates of the recognition frame of the floor cleaning tool in the target image, determine whether there is an overlapping area between the recognition frame of the first cleaning container and the recognition frame of the floor cleaning tool; If the area ratio of the overlapping area to the recognition frame of the first cleaning container is greater than or equal to a preset first ratio threshold, determine that the first cleaning container contains the floor cleaning tool, otherwise determine that the first cleaning container and the floor cleaning tool intersect; and / or, The step of if both the first placement board and the second type of food ingredient are recognized in the third detection area, and the first placement board and the second type of food ingredient intersect, includes: Obtain the recognition frame of the first placement board in the third detection area; Obtain the recognition frame of the second type of food ingredient in the third detection area; According to the position coordinates of the recognition frame of the first placement board in the target image and the position coordinates of the recognition frame of the second type of food ingredient in the target image, determine whether there is an overlapping area between the recognition frame of the first placement board and the recognition frame of the second type of food ingredient; If the area ratio of the overlapping area to the recognition frame of the first placement board is greater than or equal to a preset second ratio threshold, determine that the first placement board and the second type of food ingredient intersect; and / or, The step of if both the second placement board and the first type of food ingredient are recognized in the fourth detection area, and the second placement board and the first type of food ingredient intersect, includes: Obtain the recognition frame of the second placement board in the fourth detection area; Obtain the recognition frame of the first type of food ingredient in the fourth detection area; Determine whether there is an overlapping area between the recognition frame of the second placement board and the recognition frame of the first type of food ingredients based on the position coordinates of the recognition frame of the second placement board in the target image and the position coordinates of the recognition frame of the first type of food ingredients in the target image; If the area ratio of the overlapping area to the recognition frame of the second placement board is greater than or equal to a preset second ratio threshold, determine that the second placement board and the first type of food ingredients intersect; and / or, The food processing board is configured with a first color, and the cleaning cloth is configured with a second color, and the first color is different from the second color. Wherein, if the food processing board and the cleaning cloth are simultaneously recognized in the fifth detection area and the food processing board and the cleaning cloth intersect, output a fifth warning instruction, including: Perform color recognition on the target objects in the fifth detection area, recognize the target object with the first color as the food processing board, and recognize the target object with the second color as the cleaning cloth; If the food processing board and the cleaning cloth intersect, output a fifth warning instruction.

7. An apparatus for identifying abnormal operations in the back kitchen, characterized in that, The device: An acquisition module, configured to acquire a target image of a target environment area in the back kitchen, where the target environment area includes at least one detection area; An identification module, configured to identify a first item and a second item in the detection area, and determine whether the first item and the second item intersect according to the position information of the first item and the second item; wherein, the first item is a tool for processing food ingredients, and the second item is a movable item, and the second item is prohibited from being placed on the first item; An alarm module, configured to output an alarm instruction if the first item intersects the second item; Wherein, the recognition module is specifically configured to: acquire false alarm information fed back by the user for the alarm instruction within a preset time period, determine a movable item with a false alarm rate higher than a preset false alarm rate threshold, and divide the movable item into at least a first part and a second part. Among the false alarm information fed back by the user: the first part is determined to intersect the first item more than a first threshold, or the proportion of the number of images in which the first part is determined to intersect the first item in the number of images of false alarms when the user feeds back that the first item intersects the second item is greater than a second threshold; if it is determined that the first item intersects the first part of the movable item, no alarm instruction is output; The recognition module is further specifically configured to: obtain a plurality of specific false alarm images, where the specific false alarm images are false alarm images in which the first part is determined to intersect with the first item; mark the intersection positions of the first part and the first item in the plurality of specific false alarm images; according to the marked intersection positions of the first part and the first item, divide the first part into at least a third sub - part and a fourth sub - part, where the number of times the third sub - part is determined to intersect with the first item is greater than a third threshold, or the proportion of the number of images in which the third sub - part is determined to intersect with the first item in the number of specific false alarm images is greater than a fourth threshold; if in the obtained target image, it is determined that the third sub - part intersects with the first item, no alarm instruction is output; if in the obtained target image, it is determined that the fourth sub - part intersects with the first item, an alarm instruction is output.

8. A system for identifying abnormal operations in the back kitchen, characterized in that, The system includes: An image acquisition device, configured to obtain a target image of a target environment area in the back kitchen, where the target environment area includes at least one detection area, identify a first item and a second item in the detection area, determine whether the first item and the second item intersect according to the position information of the first item and the second item, and if they intersect, output an identification result; where the first item is a tool for processing food ingredients, and the second item is a movable item, and the second item is prohibited from being placed on the first item; A server, configured to output an alarm instruction in response to the identification result; The image acquisition device is further configured to obtain false alarm information fed back by a user for the alarm instruction within a preset time period, determine a movable item with a false alarm rate higher than a preset false alarm rate threshold, and divide the movable item into at least a first part and a second part, where in the false alarm information fed back by the user: the number of times the first part is determined to intersect with the first item is greater than a first threshold, or the proportion of the number of images in which the first part is determined to intersect with the first item in the number of false alarm images in which the user feeds back that the first item intersects with the second item is greater than a second threshold; if it is determined that the first item intersects with the first part of the movable item, no alarm instruction is output; The image acquisition device is further configured to obtain a plurality of specific false alarm images, where the specific false alarm images are false alarm images in which the first part is determined to intersect with the first item; mark the intersection positions of the first part and the first item in the plurality of specific false alarm images respectively; divide the first part into at least a third sub-part and a fourth sub-part according to the marked intersection positions of the first part and the first item, where the third sub-part is determined to intersect with the first item more than a third threshold value, or the proportion of the number of images in which the third sub-part is determined to intersect with the first item in the number of specific false alarm images is greater than a fourth threshold value; if in the obtained target image, it is determined that the third sub-part intersects with the first item, no alarm instruction is output; if in the obtained target image, it is determined that the fourth sub-part intersects with the first item, an alarm instruction is output.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method for identifying abnormal operations in the back kitchen according to any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for identifying abnormal operations in the back kitchen according to any one of claims 1 to 6.

Images