Face attendance checking method, device and equipment and storage medium

Abstract

The invention discloses a face attendance checking method and device, equipment and a storage medium. The method comprises the following steps: acquiring a panoramic picture of a student and a preset number threshold; analyzing the panoramic picture to obtain character position distribution; according to the figure position distribution, based on a preset number of people threshold, dividing the panoramic picture to obtain sub-regions; controlling a close-up camera to carry out close-up shooting on the actual sub-regions corresponding to the sub-regions in the panoramic picture; and generating an attendance checking result according to a face recognition result in a picture obtained by close-up shooting. The shooting efficiency and the quality of the shot picture are both considered for shooting the close-up picture, and the system has certain maneuverability and is very suitable for classroom scenes.

Description

Technical Field

[0001] The present invention relates to the technical field of video image processing, and more specifically, to a method, device, equipment and storage medium for face attendance. Background Art

[0002] In the process of classroom teaching and work management, attendance and roll call is an important part. With the development of computer vision technology, the attendance system based on image face recognition is increasingly widely used in classroom roll call. In this system, the classroom situation is photographed to obtain the attendance image, and then the face in the attendance image is recognized to generate the attendance result, so as to achieve the purpose of non-inductive roll call.

[0003] However, the image intelligent recognition and analysis algorithms have high requirements for the original image data, and the application scenarios of remote interactive video teaching are mostly ordinary classrooms, tiered classrooms and other scenes, which have the characteristics of large space and dense population. Conventional shooting systems cannot provide attendance images that meet the image analysis requirements. For example, a wide-angle camera is used to shoot a panoramic view of the classroom area. The image of the people in the back row of the panoramic picture will not meet the minimum pixel requirement of the analysis algorithm, and there will be occlusions between the people. These factors will make the captured images unable to be fully analyzed.

[0004] To solve the above problems, the main solutions currently include: in a classroom environment, a pan-tilt camera is used to manually set shooting positions in advance for each area of ​​the classroom where students may appear, and the cruising action path of the pan-tilt camera is planned accordingly, and separate shooting is performed to obtain sufficiently clear character images. However, the disadvantages of this solution are obvious. Each shooting plan is pre-set and lacks flexibility. In some special scenes, such as changes in classroom seats, staircase environments, etc., when the positions of characters cannot be predicted, the above solution will have omissions. Therefore, this method is a passive recognition shooting system, which cannot independently complete character shooting and character recognition without human intervention. Summary of the invention

[0005] In order to overcome at least one of the defects of the above-mentioned prior art, the present invention provides a method, device, equipment and storage medium for face attendance. The technical solution adopted by the present invention is as follows.

[0006] In a first aspect, the present invention provides a method for face attendance, comprising:

[0007] Get a panoramic picture of students and the preset number threshold;

[0008] Analyze the panoramic picture to obtain the position distribution of people;

[0009] According to the distribution of the positions of the people, based on a preset threshold of the number of people, the panoramic picture is divided into sub-areas, wherein the number of people in each sub-area in the panoramic picture is required not to exceed the preset threshold of the number of people and the number of people in at least one sub-area is greater than 1;

[0010] Controlling the close-up camera to take close-up photos of the actual sub-areas corresponding to each sub-area in the panoramic image;

[0011] The attendance result is generated based on the face recognition result in the picture obtained by close-up shooting.

[0012] In one implementation, the process of controlling the close-up camera to take close-up photos of the actual sub-areas corresponding to the sub-areas in the panoramic picture includes:

[0013] Controlling the close-up camera to cruise through the actual sub-areas corresponding to the various sub-areas in the panoramic image to obtain close-up shooting images of the actual sub-areas;

[0014] Count the number of people who can perform face recognition in the close-up shots of each actual sub-area;

[0015] The close-up camera is controlled to perform a new round of cruise close-up photography, wherein the new round of cruise close-up photography is only directed to the actual sub-area where the faces of people cannot be recognized.

[0016] In one embodiment, before the process of performing a new round of cruise close-up photography, the process further includes:

[0017] According to the total number of people in each actual sub-area and the number of people who can perform face recognition, the number of close-up shooting pictures in each actual sub-area during a new round of cruise close-up shooting is determined.

[0018] In one embodiment, before the process of performing a new round of cruise close-up photography, the process further includes:

[0019] According to the sum of the total number of people in all actual sub-areas and the total number of attendances, the actual sub-areas to be photographed during a new round of cruise close-up photography are determined.

[0020] In one embodiment, before the process of performing a new round of cruise close-up photography, the process further includes:

[0021] According to the total number of people in each actual sub-area of ​​the current round of cruise and the total number of people in each actual sub-area of ​​the previous round of cruise, the actual sub-area to be photographed during the close-up photography of the new round of cruise is determined.

[0022] In one implementation, the process of dividing the panoramic image according to the person position distribution and based on a preset number threshold to obtain sub-areas includes:

[0023] Get the preset sub-area division style;

[0024] A target division pattern is selected from preset sub-region division patterns, wherein the target division pattern meets the requirement of the number of people in each sub-region and is the division pattern with the least number of sub-regions.

[0025] In one implementation, the aspect ratio of each of the sub-regions in the panoramic picture is the same as the aspect ratio of the panoramic picture.

[0026] In one embodiment, the aspect ratio of each of the sub-areas in the panoramic picture is the same as the aspect ratio of the panoramic picture, and the panoramic picture is divided into at least two rows, each row contains a number of sub-areas, the number of sub-areas contained in each row is no more than the number of sub-areas contained in the row above it, and the number of sub-areas contained in the bottom row is not less than 2.

[0027] In one implementation, the process of dividing the panoramic image according to the person position distribution and based on a preset number threshold to obtain sub-areas includes:

[0028] Dividing the panoramic picture into sub-areas from bottom to top until all the people in the panoramic picture are covered by the sub-areas, wherein the aspect ratio of each sub-area is the same as the aspect ratio of the panoramic picture;

[0029] When dividing the bottom row, first set the number of sub-regions of the row to 2, and then determine whether each sub-region meets the number requirement. If not, set the number of sub-regions of the row to 3, and then determine whether each sub-region meets the number requirement. This process is repeated until each sub-region meets the number requirement.

[0030] When dividing rows other than the bottom row, first set the number of sub-regions of the row to N, where N is the number of sub-regions of the next row, and determine whether each sub-region meets the number of people requirement. If not, set the number of sub-regions of the row to N+1, and then determine whether each sub-region meets the number of people requirement, and so on, until each sub-region meets the number of people requirement.

[0031] In a second aspect, the present invention provides a device for face attendance at a sending end, comprising:

[0032] The acquisition module is used to obtain the panoramic picture of students and the preset number threshold;

[0033] An analysis module, used for analyzing the panoramic picture to obtain the position distribution of people;

[0034] A division module, used to divide the panoramic picture into sub-areas according to the distribution of the positions of the people and based on a preset threshold of the number of people, wherein the number of people in each sub-area in the panoramic picture is required not to exceed the preset threshold of the number of people and the number of people in at least one sub-area is greater than 1;

[0035] A control module, used for controlling the close-up camera to take close-up photos of the actual sub-areas corresponding to the respective sub-areas in the panoramic picture;

[0036] The result module is used to generate attendance results based on the face recognition results in the pictures obtained by close-up shooting.

[0037] In a third aspect, the present invention provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any of the above-mentioned methods when executing the program.

[0038] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the method of any of the above-mentioned embodiments.

[0039] In the present invention, according to the position distribution of students in the classroom, the area is reasonably divided for close-up shooting, and then photos that meet the requirements of face recognition are taken, thereby completing face attendance. The present invention takes into account both the efficiency of shooting and the quality of the shot pictures for close-up shooting, and has a certain degree of mobility, which is very suitable for classroom scenes. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a schematic diagram of the overall process of an implementation method of Example 1 of the present invention.

[0041] Figure 2 It is a schematic diagram of a panoramic camera screen according to the first embodiment of the present invention.

[0042] Figure 3 It is a schematic diagram of sub-area division according to the first embodiment of the present invention.

[0043] Figure 4 It is a schematic diagram of the overall structure of the second embodiment of the present invention. DETAILED DESCRIPTION

[0044] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0045] It should be noted that the terms "first\second\..." involved in the embodiments of the present invention are only used to distinguish similar objects, and do not represent a specific order for the objects. It is understandable that the specific order or sequence of "first\second\..." can be interchanged where permitted. It should be understood that the objects distinguished by "first\second\..." can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein.

[0046] Embodiment 1

[0047] See also Figure 1 , Figure 1 This is a flow chart of a method for face attendance provided in Embodiment 1 of the present invention, the method comprising step S110, step S120, step S130, step S140 and step S150. It should be noted that step S110, step S120, step S130, step S140 and step S150 are only reference numerals for the purpose of clearly explaining the embodiment and the appendix. Figure 1 The corresponding relationship does not limit the order of the steps in this embodiment.

[0048] Step S110, obtaining a panoramic picture of students and a preset number threshold;

[0049] Step S120, analyzing the panoramic picture to obtain the position distribution of the characters;

[0050] Step S130, dividing the panoramic image into sub-areas according to the person position distribution and based on a preset number threshold, wherein the number of people in each sub-area in the panoramic image is required not to exceed the preset number threshold and the number of people in at least one sub-area is greater than 1;

[0051] Step S140, controlling the close-up camera to take close-up photos of the actual sub-areas corresponding to the sub-areas in the panoramic image;

[0052] Step S150, generating attendance results based on the face recognition results in the picture obtained by close-up shooting.

[0053] In a teaching scenario, there are generally cameras that are specifically used to shoot a panoramic view of the classroom area, cameras that are specifically used to shoot close-ups of students, and a host. The host is used to analyze the images captured by the camera and control the camera to shoot according to the analysis results. The camera that is specifically used to shoot a panoramic view of the classroom area is generally a wide-angle camera. The method of this embodiment is applied to this type of host.

[0054] In step S110, a panoramic picture of the student taken by the panoramic camera is obtained.

[0055] In step S120, the obtained student panoramic picture is analyzed, and the characters in the student panoramic picture are detected by using methods such as human body detection or head detection, and then the position distribution of the characters in the student panoramic picture is obtained.

[0056] As described in the background technology, the panoramic image of students may fail to obtain a face attendance image that meets the image analysis requirements due to resolution or occlusion between characters. In order to obtain an image that meets the requirements, this method obtains an image that meets the requirements by shooting close-up images. Before shooting a close-up image, it is necessary to plan the shooting area of ​​each close-up image. Step S130 is the process of how to plan the shooting area of ​​each close-up image.

[0057] In step S130, in order to obtain an image that meets the requirements, the number of people in the close-up image is limited to a certain number range, that is, within a preset number threshold. Figure 2 As shown, Figure 2 It is a panoramic picture of students. If there are densely populated people in an area, such as Figure 2 If the area on the left side of the close-up is too small, the characters in the close-up may be blocked, and the effect is not ideal. In addition, the appropriate number of characters is also conducive to improving the clarity of the face, so this area should be subdivided as much as possible, and the densely packed characters should be separated so that the close-up contains an appropriate number of characters to facilitate the subsequent face recognition process. Figure 2 As shown, Figure 2 The two dotted boxes in the figure are the images taken by two close-up shots. Figure 2 The number of people included in the close-up shot is controlled to be less than 3, which not only avoids occlusion between people, but also improves the clarity of the face, and better meets the subsequent face recognition process. The division method of step S130 is divided according to actual conditions, so it can meet the current division needs according to local conditions and has a certain degree of mobility and flexibility.

[0058] It should be noted here that in addition to capturing images that meet the requirements, step S130 also requires taking efficiency into consideration and taking as few shots as possible. Otherwise, if close-ups of each person are taken one by one, the captured images will definitely meet the requirements. However, this is too time-consuming and laborious, so it is required that the number of people in each sub-area cannot be 1, and the number of people in at least one sub-area must be greater than 1.

[0059] After the sub-areas are divided in step S130 and how to perform close-up shooting is arranged, shooting begins in step S140. Generally, there is a certain linkage mechanism between the panoramic camera and the close-up camera, through which the sub-areas in each picture corresponding to the sub-areas of the actual space can be found, and then close-up shooting can be performed.

[0060] It should be noted that the method for implementing the close-up picture shooting can be shooting one by one or by cruising, which is not limited in this embodiment. In addition, in order to obtain a more reliable image, the shooting process can be shot once or several times, which is not limited in this embodiment.

[0061] In step S150, after obtaining close-up shots of each actual sub-area, face recognition can be performed on these images, and the images with people's faces can be uploaded to a face recognition machine, such as a server, to obtain face recognition information or results, and compare them with a database of pre-stored student face information to obtain matching face information, and then the attendance results of this class can be obtained.

[0062] In this method, according to the distribution of students in the classroom, the area is reasonably divided for close-up shooting, and then photos that meet the requirements of face recognition are taken to complete face attendance. This method takes into account both the efficiency of shooting and the quality of the shooting picture for close-up shooting, and has a certain degree of mobility, which is very suitable for classroom scenes.

[0063] In one implementation, the process of step S140 includes: step S141, step S143 and step S145.

[0064] Step S141, controlling the close-up camera to cruise through the actual sub-areas corresponding to the sub-areas in the panoramic image, and obtaining close-up images of the actual sub-areas;

[0065] Step S143, counting the number of people who can perform face recognition in the close-up shooting picture of each actual sub-area;

[0066] Step S145, controlling the close-up camera to perform a new round of cruise close-up photography, wherein the new round of cruise close-up photography is only for the actual sub-area where there are human faces and face recognition cannot be performed.

[0067] This embodiment is to shoot close-up pictures by cruising. After dividing each sub-area in step S130, the cruising route is planned, and then each actual sub-area is traversed to obtain a close-up picture. In this embodiment, in order to obtain a picture with quality that meets the requirements, multiple cruises are implemented, but in order to improve the efficiency of the cruise, before a new round of cruises, it is first determined whether the previous close-up picture fully meets the requirements, and whether all the faces of the characters in the picture can perform face recognition. If they all meet the requirements and can perform face recognition, then the area is skipped, and the new round of cruise close-up shooting does not include this area.

[0068] It should be noted here that the number of rounds of cruising is not limited in the present embodiment and can be pre-set, such as pre-setting 3 rounds. After 3 rounds, the cruising will be stopped regardless of whether there are faces of people in the close-up picture that cannot perform face recognition. Alternatively, the number of rounds of cruising is not limited, and the cruising will be carried out until all faces can perform face recognition. Alternatively, the number of rounds and the number of sub-areas are pre-set, such as pre-setting 3 rounds and 3 sub-areas. If after 3 rounds, there are only 3 or less sub-areas with faces of people that cannot perform face recognition, the cruising can be stopped.

[0069] In one implementation, before step S145, the process includes: step S144.

[0070] Step S144, determining the number of close-up shooting images of each actual sub-area during a new round of cruise close-up shooting according to the total number of people in each actual sub-area and the number of people who can perform face recognition.

[0071] Since some students lower their heads or face sideways, they may not be recognized. Therefore, after each round of patrol, the total number of people in each actual sub-area and the number of people who can perform face recognition are counted. Through these two data, we can draw a conclusion about the difficulty of close-up shooting in each actual sub-area, for example, through the recognition rate X, let X = number of people who can be recognized / total number of people, the lower the recognition rate, the higher the difficulty, or through the number of people who cannot perform face recognition, the more people, the higher the difficulty. For actual sub-areas with higher difficulty, more attention should be given, and step S144 is the process of paying attention. Specifically, step S144 determines the number of close-up shots by multiple shooting or multiple captures, taking the recognition rate X as an example, and determines the number of captured pictures when a certain area is subsequently traversed according to the recognition rate X (the interval time for capturing pictures is T, which can be set): 0≤X<25%: the number of captured pictures is 4, 25%≤X<50%: the number of captured pictures is 3, 50%≤X<75%: the number of captured pictures is 2, 75%≤X<100%: the number of captured pictures is 1, 100%: that is, all are recognized and the area is skipped directly.

[0072] This embodiment reduces the number of cruises by acquiring more close-up images during the cruise.

[0073] In one implementation, before step S145, the following steps are included:

[0074] According to the sum of the total number of people in all actual sub-areas and the total number of attendances, the actual sub-areas to be photographed during a new round of cruise close-up photography are determined.

[0075] During the patrol process, some students may bend under the table, so that the close-up image cannot capture this student. If this area is skipped directly, the attendance result will be wrong. Therefore, in this embodiment, it is necessary to check with the total number of attendance. If it is different from the total number of attendance, the above situation may occur, so the area cannot be skipped directly.

[0076] It should be noted that the total number of students in attendance here is the total number of students currently in the classroom, not the total number of students who should theoretically take this class. This application does not limit how to obtain this total number.

[0077] In one implementation, before step S145, the following steps are included:

[0078] According to the total number of people in each actual sub-area of ​​the current round of cruise and the total number of people in each actual sub-area of ​​the previous round of cruise, the actual sub-area to be photographed during the close-up photography of the new round of cruise is determined.

[0079] The same as the previous implementation is to solve the problem that students cannot be photographed when bending over. In this implementation, the result of the cruise is compared with the previous one. If the number of students is consistent, it is considered that all students are photographed. If the number of students is inconsistent, it is considered that some students are not photographed, and the area cannot be skipped directly.

[0080] In one implementation, the process of step S130 includes: step S131 and step S132.

[0081] Step S131, obtaining a preset sub-region division pattern;

[0082] Step S132, selecting a target division pattern from the preset sub-region division patterns, wherein the target division pattern meets the number of people required for each sub-region and is the division pattern with the least number of sub-regions.

[0083] In this embodiment, the centralized division area styles are preset, and the close-up shooting parameters are preset and adjusted for these preset division styles, which makes the division process simpler and more efficient. In fact, this embodiment is to pick out the most suitable style from the preset sub-area division styles. This style needs to meet two conditions. The first is to meet the number of people in each sub-area, and the number of people in each sub-area does not exceed the preset number threshold. The second is to select the one with the least number of sub-areas on the basis of meeting the first condition, because the least one means the least close-up pictures, which is the most efficient.

[0084] In one implementation, the aspect ratio of each of the sub-regions in the panoramic picture is the same as the aspect ratio of the panoramic picture.

[0085] Generally speaking, the aspect ratio of the picture taken by a close-up camera is the same as that of the picture taken by a panoramic camera, for example, both are 16:9. Therefore, when dividing the sub-areas, try to make the aspect ratio of the sub-areas the same as that of the close-up camera to facilitate the shooting of the close-up camera.

[0086] It should be noted here that if the sub-areas are divided with a fixed aspect ratio as above, the following may occur: Figure 3 As shown in the figure, all sub-areas cannot fill the panoramic picture, such as Figure 3 The top and bottom of the image are not divided into any sub-areas. Since some areas are less important and have no people, it is not necessary to divide all the positions in the panoramic image. Some positions can be excluded from any sub-area, and these positions naturally do not need to be photographed in close-up. Similarly, even if the sub-areas are not divided with a fixed aspect ratio, some positions can be excluded from any sub-area.

[0087] In one embodiment, the aspect ratio of each of the sub-areas in the panoramic picture is the same as the aspect ratio of the panoramic picture, and the panoramic picture is divided into at least two rows, each row contains a number of sub-areas, the number of sub-areas contained in each row is no more than the number of sub-areas contained in the row above it, and the number of sub-areas contained in the bottom row is not less than 2.

[0088] like Figure 3 As shown, Figure 3 It is a schematic diagram of the sub-area division. In this embodiment, several rows are divided, and the distance between each row and the camera is almost the same. Because things that are close are larger and things that are far away are smaller, the bottom of the panoramic image is the closest scenery, and the characters will be larger. The farther up and farther away, the smaller the characters will be. Therefore, the area at the bottom should also be larger, that is, there will be fewer sub-areas in one row, and the higher up, the more sub-areas will be accommodated. Since this attendance method is used in the scene of class roll call, students are generally in their seats at this time, with a certain distance between each other, and they will not gather together, so if Figure 3 This division can divide the sub-areas that meet the requirements.

[0089] Specifically, the sub-areas are divided according to the following logic:

[0090] 1. Divide the area from near to far, with the bottom row as the first row. For the first row: first set the length of the sub-area to 1 / 2 of the length of the panoramic image, that is, the first row contains 2 sub-areas, and then determine whether each sub-area meets the preset number of people threshold condition; if it does, divide the second row downwards, if not, set the length of the sub-area to 1 / 3 of the length of the panoramic image, and so on, until each sub-area meets the preset number of people threshold condition;

[0091] 2. For the second row, first set the length of the sub-area of ​​the second row to be equal to that of the first row, and then determine whether the condition is met. If so, divide the area down to the third row. If not, refer to step 1 to refine the area and make another determination until the condition is met.

[0092] 3. The sub-area division of each subsequent row is repeated in this way until all the people in the panoramic image are covered by the area.

[0093] Embodiment 2

[0094] Corresponding to the method of embodiment 1, as Figure 4 As shown, the present invention also provides a sending-end face attendance device 4, comprising: an acquisition module 410, an analysis module 420, a division module 430, a control module 440 and a result module 450.

[0095] An acquisition module 410 is used to acquire a panoramic picture of students and a preset number threshold;

[0096] An analysis module 420 is used to analyze the panoramic picture to obtain the position distribution of people;

[0097] A division module 430 is used to divide the panoramic picture into sub-areas according to the distribution of the person positions and based on a preset number threshold, wherein the number of people in each sub-area in the panoramic picture is required not to exceed the preset number threshold and the number of people in at least one sub-area is greater than 1;

[0098] The control module 440 is used to control the close-up camera to take close-up photos of the actual sub-areas corresponding to each sub-area in the panoramic picture;

[0099] The result module 450 is used to generate attendance results according to the face recognition results in the pictures obtained by close-up shooting.

[0100] In one implementation, the control module executes the process of controlling the close-up camera to take close-up photos of the actual sub-areas corresponding to the sub-areas in the panoramic picture, including:

[0101] Controlling the close-up camera to cruise through the actual sub-areas corresponding to the various sub-areas in the panoramic image to obtain close-up shooting images of the actual sub-areas;

[0102] Count the number of people who can perform face recognition in the close-up shots of each actual sub-area;

[0103] The close-up camera is controlled to perform a new round of cruise close-up photography, wherein the new round of cruise close-up photography is only directed to the actual sub-area where the faces of people cannot be recognized.

[0104] In one embodiment, before the process of performing a new round of cruise close-up photography, the process further includes:

[0105] According to the total number of people in each actual sub-area and the number of people who can perform face recognition, the number of close-up shooting pictures in each actual sub-area during a new round of cruise close-up shooting is determined.

[0106] In one implementation, the division module performs the process of dividing the panoramic image according to the person position distribution and based on a preset number threshold to obtain sub-areas, including:

[0107] Get the preset sub-area division style;

[0108] A target division pattern is selected from preset sub-region division patterns, wherein the target division pattern meets the requirement of the number of people in each sub-region and is the division pattern with the least number of sub-regions.

[0109] In one implementation, the aspect ratio of each of the sub-regions in the panoramic picture is the same as the aspect ratio of the panoramic picture.

[0110] In one embodiment, the aspect ratio of each of the sub-areas in the panoramic picture is the same as the aspect ratio of the panoramic picture, and the panoramic picture is divided into at least two rows, each row contains a number of sub-areas, the number of sub-areas contained in each row is no more than the number of sub-areas contained in the row above it, and the number of sub-areas contained in the bottom row is not less than 2.

[0111] In one implementation, the division module performs the process of dividing the panoramic image according to the person position distribution and based on a preset number threshold to obtain sub-areas, including:

[0112] Dividing the panoramic picture into sub-areas from bottom to top until all the people in the panoramic picture are covered by the sub-areas, wherein the aspect ratio of each sub-area is the same as the aspect ratio of the panoramic picture;

[0113] When dividing the bottom row, first set the number of sub-regions of the row to 2, and then determine whether each sub-region meets the number requirement. If not, set the number of sub-regions of the row to 3, and then determine whether each sub-region meets the number requirement. This process is repeated until each sub-region meets the number requirement.

[0114] When dividing rows other than the bottom row, first set the number of sub-regions of the row to N, where N is the number of sub-regions of the next row, and determine whether each sub-region meets the number of people requirement. If not, set the number of sub-regions of the row to N+1, and then determine whether each sub-region meets the number of people requirement, and so on, until each sub-region meets the number of people requirement.

[0115] In this device, according to the distribution of students in the classroom, the area is reasonably divided for close-up shooting, and then photos that meet the requirements of face recognition are taken to complete face attendance. This device takes into account both the efficiency of shooting and the quality of the shooting picture for close-up shooting, and has a certain degree of mobility, which is very suitable for classroom scenes.

[0116] Embodiment 3

[0117] An embodiment of the present invention further provides a storage medium on which computer instructions are stored. When the instructions are executed by a processor, the method for face attendance in any of the above embodiments is implemented.

[0118] Those skilled in the art can understand that: all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: mobile storage devices, random access memories (RAM), read-only memories (ROM), magnetic disks or optical disks, etc., various media that can store program codes.

[0119] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the relevant technology. The computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, terminal, or network device, etc.) to execute all or part of the methods of each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, RAM, ROM, magnetic disks or optical disks.

[0120] Corresponding to the above-mentioned computer storage medium, in one embodiment, a computer device is also provided, which includes a memory, an encoder, and a computer program stored in the memory and executable on the encoder, wherein when the encoder executes the program, a facial attendance method as described in any one of the above-mentioned embodiments is implemented.

[0121] The computer device can reasonably divide the area for close-up shooting according to the position distribution of students in the classroom, and then take photos that meet the requirements of face recognition, so as to complete face attendance. The close-up shooting of the computer device takes into account both the shooting efficiency and the quality of the shooting picture, and has a certain degree of mobility, which is very suitable for classroom scenes.

[0122] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described 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.

[0123] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A facial attendance method, characterized in that: include: Get a panoramic picture of students and the preset number threshold; Analyze the panoramic picture to obtain the position distribution of people; According to the distribution of the positions of the people, based on a preset threshold of the number of people, the panoramic picture is divided into sub-areas, wherein the number of people in each sub-area in the panoramic picture is required not to exceed the preset threshold of the number of people and the number of people in at least one sub-area is greater than 1; Controlling the close-up camera to take close-up photos of the actual sub-areas corresponding to each sub-area in the panoramic image; The attendance result is generated based on the face recognition result in the picture obtained by close-up shooting.

2. The method for face attendance according to claim 1, characterized in that: The process of controlling the close-up camera to take close-up photos of the actual sub-areas corresponding to the sub-areas in the panoramic picture includes: Controlling the close-up camera to cruise through the actual sub-areas corresponding to the various sub-areas in the panoramic image to obtain close-up shooting images of the actual sub-areas; Count the number of people who can perform face recognition in the close-up shots of each actual sub-area; The close-up camera is controlled to perform a new round of cruise close-up photography, wherein the new round of cruise close-up photography is only directed to the actual sub-area where the faces of people cannot be recognized.

3. The method for face attendance according to claim 2, characterized in that: Before the process of taking a new round of cruise close-up photos, the following steps are also included: According to the total number of people in each actual sub-area and the number of people who can perform face recognition, the number of close-up shooting pictures in each actual sub-area during a new round of cruise close-up shooting is determined.

4. The method for face attendance according to claim 2, characterized in that: Before the process of taking a new round of cruise close-up photos, the following steps are also included: According to the sum of the total number of people in all actual sub-areas and the total number of attendances, the actual sub-areas to be photographed during a new round of cruise close-up photography are determined.

5. The method for face attendance according to claim 2, characterized in that: Before the process of taking a new round of cruise close-up photos, the following steps are also included: According to the total number of people in each actual sub-area of ​​the current round of cruise and the total number of people in each actual sub-area of ​​the previous round of cruise, the actual sub-area to be photographed during the close-up photography of the new round of cruise is determined.

6. The facial attendance method according to claim 1, characterized in that: The process of dividing the panoramic picture according to the person position distribution and based on a preset number threshold to obtain sub-areas includes: Get the preset sub-area division style; A target division pattern is selected from preset sub-region division patterns, wherein the target division pattern meets the requirement of the number of people in each sub-region and is the division pattern with the least number of sub-regions.

7. The facial attendance method according to any one of claims 1 to 6, characterized in that: The aspect ratio of each of the sub-areas in the panoramic picture is the same as the aspect ratio of the panoramic picture.

8. The facial attendance method according to any one of claims 1 to 6, characterized in that: The aspect ratio of each of the sub-areas in the panoramic picture is the same as that of the panoramic picture. The panoramic picture is divided into at least two rows, each row contains a number of sub-areas, the number of sub-areas contained in each row is no more than the number of sub-areas contained in the row above it, and the number of sub-areas contained in the bottom row is no less than 2.

9. The facial attendance method according to claim 1, characterized in that: The process of dividing the panoramic picture according to the person position distribution and based on a preset number threshold to obtain sub-areas includes: Dividing the panoramic picture into sub-areas from bottom to top until all the people in the panoramic picture are covered by the sub-areas, wherein the aspect ratio of each sub-area is the same as the aspect ratio of the panoramic picture; When dividing the bottom row, first set the number of sub-regions of the row to 2, and then determine whether each sub-region meets the number requirement. If not, set the number of sub-regions of the row to 3, and then determine whether each sub-region meets the number requirement. This process is repeated until each sub-region meets the number requirement. When dividing rows other than the bottom row, first set the number of sub-regions of the row to N, where N is the number of sub-regions of the next row, and determine whether each sub-region meets the number of people requirement. If not, set the number of sub-regions of the row to N+1, and then determine whether each sub-region meets the number of people requirement, and so on, until each sub-region meets the number of people requirement.

10. A device for face attendance at the sending end, characterized in that: include: The acquisition module is used to obtain the panoramic picture of students and the preset number threshold; An analysis module, used for analyzing the panoramic picture to obtain the position distribution of people; A division module, used to divide the panoramic picture into sub-areas according to the distribution of the positions of the people and based on a preset threshold of the number of people, wherein the number of people in each sub-area in the panoramic picture is required not to exceed the preset threshold of the number of people and the number of people in at least one sub-area is greater than 1; A control module, used for controlling the close-up camera to take close-up photos of the actual sub-areas corresponding to the respective sub-areas in the panoramic picture; The result module is used to generate attendance results based on the face recognition results in the pictures obtained by close-up shooting.

11. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method according to any one of claims 1 to 9 is implemented.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 9 is implemented.

Images