Attribute information detection method and device and storage medium

Abstract

The invention discloses an attribute information detection method and device and a storage medium. The attribute information detection method comprises the steps of extracting back sign information of a to-be-detected object from a shot image of a first camera; wherein the first camera is one of the two cameras, and the two cameras are arranged back to back; matching the back sign information of the to-be-detected object with the first data set to determine a first matching object; wherein the first data set comprises front sign information and front attribute information of a plurality of first objects extracted from a shot image of the second camera, the first matching object is one of the plurality of first objects, and the second camera is the other one of the two cameras; and taking the front attribute information of the first matching object as the front attribute information of the to-be-detected object. In this way, the front attribute information of the to-be-detected object can be rapidly and accurately obtained.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to an attribute information detection method, device, and storage medium. Background Art

[0002] In practical applications, detecting the attributes of objects under inspection is of great significance. For example, in criminal investigation scenarios, based on the attributes of the objects under inspection, a basic profile of the objects under inspection can be obtained, providing effective clues for relevant departments and narrowing the scope of investigation. In another example, in safe passage scenarios, the objects under inspection can be identified as violating safety regulations.

[0003] Currently, image recognition algorithms (e.g., facial recognition, body recognition, etc.) are typically used to process images captured by a camera to obtain various attribute information of the object to be detected in the captured image, thereby achieving attribute information detection of the object to be detected. However, due to deployment issues of video security systems, when the object to be detected is facing away from the camera's captured image, only information related to the back of the object to be detected can be obtained, and attribute information of the front of the object to be detected cannot be directly obtained.

[0004] Therefore, when the object to be detected is facing away from the camera's shooting screen, how to obtain the front attribute information of the object to be detected becomes a technical problem that needs to be solved urgently. Summary of the Invention

[0005] The main technical problem solved by the present application is to provide an attribute information detection method, device and computer-readable storage medium, which can quickly and accurately obtain the positive attribute information of the object to be detected.

[0006] In order to solve the above technical problems, a technical solution adopted in this application is: to provide an attribute information detection method, the method comprising: extracting the back physical sign information of the object to be detected from the image captured by the first camera; wherein, the first camera is one of the two cameras, and the two cameras are arranged back to back; matching the back physical sign information of the object to be detected with the first data set to determine the first matching object; wherein, the first data set includes the front physical sign information and front attribute information of several first objects extracted from the image captured by the second camera, the first matching object is one of the several first objects, and the second camera is the other of the two cameras; using the front attribute information of the first matching object as the front attribute information of the object to be detected.

[0007] Optionally, the two cameras are integrated into one camera device; or, the two cameras are two independent camera devices, and the installation distance between the two cameras is less than a preset distance.

[0008] Optionally, matching the back physical sign information of the object to be detected with the first data set to determine the first matching object includes: respectively determining the first similarity between the back physical sign information of the object to be detected and the front physical sign information of each first object; and taking the first object whose first similarity meets the first similarity condition as the first matching object.

[0009] Optionally, the step of obtaining the first data set includes: in response to detecting that the first object enters the shooting screen of the second camera from the front, extracting the front physical sign information and front attribute information of the first object from the shooting image of the second camera; and adding the front physical sign information and front attribute information of the first object to the first data set.

[0010] Optionally, adding the frontal physical sign information and frontal attribute information of the first object to the first data set includes: matching the frontal physical sign information of the first object with the second data set to determine whether there is a second matching object that matches the first object; wherein the frontal information of the first object includes at least one of the frontal physical sign information and frontal attribute information of the first object, the second data set includes the frontal physical sign information and frontal attribute information of several second objects extracted from the image captured by the first camera, and the second matching object is one of the several second objects; in response to the absence of the second matching object, adding the frontal physical sign information and frontal attribute information of the first object to the first data set; in response to the existence of the second matching object, deleting the frontal physical sign information and frontal attribute information of the second matching object from the second data set, and adding the frontal physical sign information and frontal attribute information of the first object to the first data set.

[0011] Optionally, after taking the front attribute information of the first matching object as the front attribute information of the object to be detected, the method further includes: deleting the front physical sign information and the front attribute information of the first matching object from the first data set.

[0012] Optionally, for the positive physical sign information and positive attribute information of each first object in the first data set, in response to the storage time of the positive physical sign information and positive attribute information of the first object being greater than the elimination time, the positive physical sign information and positive attribute information of the first object are deleted from the first data set.

[0013] Optionally, the step of determining the elimination time includes: determining the farthest detection distance of the two cameras; determining a first ratio between the farthest detection distance and a preset speed; and taking the product of the first ratio and a preset multiple as the elimination time.

[0014] Optionally, in response to the number of the first subjects corresponding to the first data set being greater than a target number, the positive physical sign information and positive attribute information of some of the first subjects in the first data set are deleted.

[0015] Optionally, the step of determining the target number includes: counting the total number of first objects passing through the second camera within a preset time period; determining a second ratio between the total number and the preset time period; and taking the product of the second ratio and the elimination time period as the target number.

[0016] In order to solve the above technical problems, another technical solution adopted in this application is: to provide an attribute information detection device, which includes a first acquisition module, a matching module and a second acquisition module. The first acquisition module is used to extract the back physical sign information of the object to be detected from the image captured by the first camera; wherein, the first camera is one of the two cameras, and the two cameras are set back to back. The matching module is used to match the back physical sign information of the object to be detected with the first data set to determine the first matching object; wherein, the first data set includes the front physical sign information and front attribute information of several first objects extracted from the image captured by the second camera, the first matching object is one of the several first objects, and the second camera is the other of the two cameras. The second acquisition module is used to use the front attribute information of the first matching object as the front attribute information of the object to be detected.

[0017] Optionally, the two cameras are integrated into one camera device; or, the two cameras are two independent camera devices, and the installation distance between the two cameras is less than a preset distance.

[0018] Optionally, the matching module is used to respectively determine the first similarity between the back physical sign information of the object to be detected and the front physical sign information of each first object; and use the first object whose first similarity meets the first similarity condition as the first matching object.

[0019] Optionally, the attribute information detection device further includes a third acquisition module. The third acquisition module is configured to acquire the first data set. In response to detecting that the first subject's front face enters the capture image of the second camera, the third acquisition module is configured to extract frontal vital sign information and frontal attribute information of the first subject from the image captured by the second camera; and to add the frontal vital sign information and frontal attribute information of the first subject to the first data set.

[0020] Optionally, the third acquisition module is used to match the front information of the first object with the second data set to determine whether there is a second matching object that matches the first object; wherein the front information of the first object includes at least one of the front physical sign information and front attribute information of the first object, the second data set includes the front physical sign information and front attribute information of several second objects extracted from the captured image of the first camera, and the second matching object is one of the several second objects; in response to the absence of the second matching object, the front physical sign information and front attribute information of the first object are added to the first data set; in response to the existence of the second matching object, the front physical sign information and front attribute information of the second matching object are deleted from the second data set, and the front physical sign information and front attribute information of the first object are added to the first data set.

[0021] Optionally, the attribute information detection module further includes an elimination module configured to delete the first matching object's frontal physical sign information and frontal attribute information from the first data set after the second acquisition module uses the first matching object's frontal attribute information as the frontal attribute information of the object to be detected.

[0022] Optionally, for the positive physical sign information and positive attribute information of each first object in the first data set, the elimination module is also used to delete the positive physical sign information and positive attribute information of the first object from the first data set in response to the storage time of the positive physical sign information and positive attribute information of the first object being greater than the elimination time.

[0023] Optionally, the elimination module is used to determine the farthest detection distance of the two cameras; determine a first ratio between the farthest detection distance and a preset speed; and use the product of the first ratio and a preset multiple as the elimination time.

[0024] Optionally, the elimination module is configured to delete the positive physical sign information and positive attribute information of some of the first subjects in the first data set in response to the number of the first subjects corresponding to the first data set being greater than a target number.

[0025] Optionally, the elimination module is used to count the total number of first objects passing through the second camera within a preset time period; determine a second ratio between the total number and the preset time period; and use the product of the second ratio and the elimination time period as the target number.

[0026] To solve the above technical problems, another technical solution adopted in this application is: to provide an electronic device, including a memory and a processor coupled to each other, the memory storing program instructions; the processor is used to execute the program instructions stored in the memory to implement the above attribute information detection method.

[0027] In order to solve the above technical problems, another technical solution adopted in this application is: providing a computer-readable storage medium, which is used to store program instructions, and the program instructions can be executed by a processor to implement the above attribute information detection method.

[0028] The above scheme designs two cameras that are set up to face each other, with the first camera being one of the two cameras and the second camera being the other of the two cameras. Since the two cameras are set up to face each other, if the object to be detected enters the shooting picture of one of the cameras from the front, the object to be detected will leave the shooting picture of the other camera with its back facing away. When the object to be detected leaves the shooting picture of the first camera, only the back physical sign information of the object to be detected can be extracted from the shooting picture of the first camera, but the front attribute information of the object to be detected can be obtained from the shooting picture obtained by the second camera that faces away. Therefore, the back physical sign information of the object to be detected extracted from the shooting picture of the first camera is matched with the first data set to determine the first matching object. The first data set includes the front physical sign information and front attribute information of several first objects extracted from the shooting picture of the second camera. The first matching object is one of the several first objects. Afterwards, the front attribute information of the first matching object can be used as the front attribute information of the object to be detected. In this way, the front attribute information of the object to be detected can be obtained quickly and accurately. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 This is a flow chart of an embodiment of the attribute information detection method provided by this application;

[0030] Figure 2 This is a flowchart of an embodiment of a method for obtaining a first data set provided by this application;

[0031] Figure 3 is a schematic diagram of the first data set and the second data set provided in this application;

[0032] Figure 4 This is a structural diagram of an embodiment of the attribute information detection system provided by this application;

[0033] Figure 5 This is a schematic diagram of the framework of an embodiment of the attribute information detection device provided by the present application;

[0034] Figure 6 This is a schematic diagram of the framework of an embodiment of an electronic device provided by the present application;

[0035] Figure 7 It is a schematic diagram of a framework of an embodiment of a computer-readable storage medium provided by this application. DETAILED DESCRIPTION

[0036] The following is a clear and complete description of the technical solutions in the embodiments of the present application in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0037] It should be noted that the term "several" in this article means at least one; the terms "first", "second", etc. are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0038] See also Figure 1 , Figure 1 This is a flow chart of an embodiment of the attribute information detection method provided by this application. The method is applied to an attribute information detection system, which includes two cameras arranged in back-to-back directions, the first camera being one of the two cameras, and the second camera being the other of the two cameras. In addition, the attribute information detection system can be deployed in a one-way road or a two-way road, and this embodiment does not specifically limit this. It should be noted that if there are substantially the same results, the method of this application is not limited to Figure 1 The process sequence shown is limited.

[0039] like Figure 1 As shown, the method includes the following steps:

[0040] S11: Extracting back physical sign information of the object to be detected from the image captured by the first camera.

[0041] When the object to be detected turns away from the shooting picture of the first camera, the back physical sign information of the object to be detected can be extracted from the shooting image of the first camera.

[0042] For example, the back physical sign information of the object to be detected can be extracted from the captured image by using a relevant image recognition algorithm. For details, reference can be made to the methods in the prior art, which will not be described in detail in this embodiment.

[0043] For example, the object to be detected is a person, and the back physical sign information of the object to be detected may include information such as the back head-to-shoulder ratio, the back top color, and the back bottom color of the object to be detected.

[0044] S12: Matching the back physical sign information of the object to be detected with the first data set to determine a first matching object.

[0045] The first data set includes frontal physical sign information and frontal attribute information of a plurality of first objects extracted from images captured by the second camera, and the first matching object is one of the plurality of first objects.

[0046] When the front of each first object enters the shooting picture of the second camera, the front physical sign information and front attribute information of each first object can be obtained from the image captured by the second camera. The front attribute information of each first object can only be obtained from the front detection of each first object.

[0047] Exemplarily, the frontal physical sign information and frontal attribute information of the first object can be extracted from the captured image by using a relevant image recognition algorithm. For details, reference can be made to the methods in the prior art, which will not be described in detail in this embodiment.

[0048] For example, the first object is a person, and the front physical sign information of the first object may include information such as the front head-to-shoulder ratio, the front top color, and the front bottom color of the first object.

[0049] For example, the frontal attribute information of the first subject includes frontal body attribute information and facial attribute information of the first subject. The frontal body attribute information may include information such as whether the subject is wearing an ID card, a tie, any handheld objects (such as a bag or folder), whether the subject is wearing glasses, whether the subject is wearing a mask, and age (such as children, teenagers, young adults, middle-aged people, and the elderly). The facial attribute information may include information such as facial expressions, face shape, and special facial marks (such as dimples and moles between the eyebrows).

[0050] In one embodiment, considering that the front and back physical sign information of the same subject have similarities (slight differences or substantially the same), similarity matching can be performed between the back physical sign information of the subject to be detected and the front physical sign information of each first subject in the first data set to quickly and accurately determine the first matching subject that matches the subject to be detected. This step can further include the following sub-steps:

[0051] Sub-step 1: determining first similarities between the back physical sign information of the subject to be detected and the front physical sign information of each first subject.

[0052] Exemplarily, the back physical sign information of the object to be detected and the front physical sign information of each first object can be vectorized respectively to obtain a back physical sign feature vector corresponding to the back physical sign information of the object to be detected and a front physical sign feature vector corresponding to the front physical sign information of each first object. Then, the similarity between the back physical sign feature vector and each front physical sign feature vector is determined respectively, such as cosine similarity, as the first similarity between the back physical sign information of the object to be detected and the front physical sign information of each first object.

[0053] Sub-step 2: taking the first object whose first similarity meets the first similarity condition as the first matching object.

[0054] Exemplarily, the first similarity condition is the highest similarity, that is, the first object with the highest first similarity is used as the first matching object.

[0055] S13: Using the front attribute information of the first matching object as the front attribute information of the object to be detected.

[0056] In the present embodiment, two cameras are designed to be arranged in a back-to-back manner, wherein the first camera is one of the two cameras and the second camera is the other of the two cameras. Since the two cameras are arranged in a back-to-back manner, if the object to be detected enters the shooting picture of one of the cameras from the front, the object to be detected will leave the shooting picture of the other camera with its back to the camera. When the object to be detected leaves the shooting picture of the first camera, only the back physical sign information of the object to be detected can be extracted from the shooting picture of the first camera, but the front attribute information of the object to be detected can be obtained from the shooting picture obtained by the second camera facing back. Therefore, the back physical sign information of the object to be detected extracted from the shooting picture of the first camera is matched with the first data set to determine the first matching object. The first data set includes the front physical sign information and front attribute information of several first objects extracted from the shooting picture of the second camera. The first matching object is one of the several first objects. Afterwards, the front attribute information of the first matching object can be used as the front attribute information of the object to be detected. In this way, the front attribute information of the object to be detected can be obtained quickly and accurately.

[0057] In this embodiment, when the attribute information detection system is deployed in a one-way passageway, if the object to be detected enters the second camera's camera view from the front and leaves the first camera's camera view with its back facing away, the object's movement direction is the same as the one-way passageway's direction of travel. In this case, only the first dataset can be constructed for querying.

[0058] When the attribute information detection system is deployed in a two-way passageway, the subject to be detected can enter the second camera's shooting frame from the front and exit the first camera's shooting frame with its back facing away. Alternatively, the subject can enter the first camera's shooting frame from the front and exit the second camera's shooting frame with its back facing away. In this case, a first dataset and a second dataset can be constructed simultaneously for querying. The first dataset includes frontal vital signs and frontal attribute information of several first subjects extracted from the images captured by the second camera. The second dataset includes frontal vital signs and frontal attribute information of several second subjects extracted from the images captured by the first camera.

[0059] See also Figure 2 , Figure 2 FIG. 1 is a flow chart of an embodiment of a method for obtaining a first data set provided by this application. Figure 2 As shown, the method includes the following steps:

[0060] S21: In response to detecting that the front face of the first object enters the shooting picture of the second camera, extracting frontal physical sign information and frontal attribute information of the first object from the shooting picture of the second camera.

[0061] S22: Adding the positive physical sign information and positive attribute information of the first subject to the first data set.

[0062] In one embodiment, without considering the situation where the first object rewinds (i.e., the first object enters the shooting screen of the second camera with its front facing forward and leaves the shooting screen of the first camera with its back facing away, and then the first object enters the shooting screen of the second camera with its front facing forward again), the frontal vital sign information and frontal attribute information of the first object can be directly added to the first data set.

[0063] In another embodiment, in consideration of the first subject's retraction, adding the first subject's positive vital sign information and positive attribute information to the first dataset further comprises the following steps:

[0064] Step 1: Use the front information of the first object to match the second data set to determine whether there is a second matching object that matches the first object.

[0065] The frontal information includes at least one of frontal physical sign information and frontal attribute information. Exemplarily, the frontal information includes only frontal physical sign information, or only frontal attribute information, or both. The second dataset includes frontal physical sign information and frontal attribute information of a plurality of second objects extracted from images captured by the first camera, and the second matching object is one of the plurality of second objects.

[0066] In step 1, the front information of the first object can be similarly matched with the front information of each second object in the second data set to determine a second matching object that matches the first object from the plurality of second objects. This step can further include the following sub-steps:

[0067] Sub-step 1: determining the second similarity between the positive information of the first object and the positive information of each second object respectively.

[0068] Exemplarily, the positive information of the first object and the positive information of each second object can be vectorized respectively to obtain the positive feature vector corresponding to the positive information of the first object and the positive feature vector corresponding to the positive information of each second object. Then, the similarity between the positive feature vector of the first object and the positive feature vector of each second object is determined respectively, such as the cosine similarity, as the second similarity between the positive information of the first object and the positive information of each second object.

[0069] Sub-step 2: taking the second object whose second similarity meets the second similarity condition as the second matching object that matches the first object.

[0070] Illustratively, the second similarity condition is that the similarity is greater than a preset similarity threshold and the similarity is the highest. The preset similarity threshold can be set according to actual needs.

[0071] When there is no second object that meets the second similarity condition, it is determined that there is no second matching object that matches the first object. When there is a second object that meets the second similarity condition, it is determined that there is a second matching object that matches the first object.

[0072] In step 1, if there is no second matching object that matches the first object, then step 2 is executed; if there is a second matching object that matches the first object, then step 3 is executed.

[0073] Step 2: Add the positive physical sign information and positive attribute information of the first subject to the first data set.

[0074] When there is no second matching object matching the first object in the second data set, it indicates that the first object is entering the shooting screen of the second camera from the front for the first time. At this time, the front physical sign information and front attribute information of the first object are directly added to the first data set.

[0075] Step three: delete the positive physical sign information and positive attribute information of the second matching object from the second data set, and add the positive physical sign information and positive attribute information of the first object to the first data set.

[0076] When there is a second matching object that matches the first object in the second data set, it indicates that this is not the first time that the first object enters the shooting screen of the second camera from the front. At this time, in order to ensure that the attribute information detection method of this application can be performed normally when the first object leaves the shooting screen of the first camera, and at the same time avoid data duplication and reduce data redundancy, the front physical sign information and front attribute information of the second matching object can be deleted from the second data set.

[0077] It should be noted that this embodiment only illustrates the process of obtaining the first data set. The principle of obtaining the second data set is the same as that of the first data set, and will not be described in detail herein.

[0078] See also Figure 3 , Figure 3 is a schematic diagram of the first data set and the second data set provided by this application. Figure 3As shown, the specific forms of the first data set and the second data set are both queues. The first data set corresponds to the first cache queue, and the second data set corresponds to the second cache queue. The first cache queue caches the positive vital signs and positive attribute information of n first objects, from first object 1 to first object n. The second cache queue caches the positive vital signs and positive attribute information of m second objects, from second object 1 to second object m. The greater the number of n, the larger the cache size occupied by the first cache queue. The greater the number of m, the larger the cache size occupied by the second cache queue.

[0079] It should be noted that, in other embodiments, the first data set and the second data set may also be in the form of arrays. This embodiment does not specifically limit the form of the first data set and the second data set.

[0080] In this embodiment, the back physical sign information of the object to be detected is extracted from the image captured by the first camera, and the back physical sign information of the object to be detected is matched with the first data set to determine a first matching object among several first objects that matches the object to be detected, and the front attribute information of the first matching object is used as the front attribute information of the object to be detected. Afterwards, the front physical sign information and front attribute information of the first matching object can be deleted from the first data set.

[0081] Similarly, the back physical sign information of the object to be detected is extracted from the image captured by the second camera, and the back physical sign information of the object to be detected is matched with the second data set to determine a third matching object among several second objects that matches the object to be detected, and the front attribute information of the third matching object is used as the front attribute information of the object to be detected. Afterwards, the front physical sign information and front attribute information of the third matching object can be deleted from the second data set.

[0082] After the match is successful and the positive attribute information of the object to be detected is obtained, the data cache size can be reduced by deleting the positive physical sign information and positive attribute information of the matching object from the first data set or the second data set. At the same time, the matching efficiency of other objects to be detected can be improved, thereby improving the query efficiency of the positive attribute information of other objects to be detected.

[0083] In this embodiment, for the positive physical sign information and positive attribute information of each first object in the first data set, if the storage time of the positive physical sign information and positive attribute information of the first object is greater than the elimination time, the positive physical sign information and positive attribute information of the first object will be deleted from the first data set.

[0084] Similarly, for the positive physical sign information and positive attribute information of each second object in the second data set, if the storage time of the positive physical sign information and positive attribute information of the second object is greater than the elimination time, the positive physical sign information and positive attribute information of the second object will be deleted from the second data set.

[0085] The storage duration refers to the length of time since the object's information was stored in the corresponding data set. When the storage duration of the first object's frontal vital sign and positive attribute information exceeds the elimination duration, the first object is considered to have turned its back and left the first camera's detection line of sight. When the storage duration of the second object's frontal vital sign and positive attribute information exceeds the elimination duration, the second object is considered to have turned its back and left the second camera's detection line of sight.

[0086] By eliminating data in the first data set and the second data set whose storage time is longer than the elimination time, redundant and useless data can be reduced, and the cache size occupied by data can be reduced. At the same time, the matching efficiency of the object to be detected can be improved, thereby improving the query efficiency of the positive attribute information of the object to be detected.

[0087] In one embodiment, the elimination time period is a fixed value that is directly preset.

[0088] In another embodiment, the elimination period is determined based on the actual scenario. Specifically, the elimination period is determined by the following steps:

[0089] Step 1: Determine the maximum detection distance between the two cameras.

[0090] The maximum detection distance of the two cameras refers to the maximum distance that an object (such as the first object or the second object) moves from the time it enters the shooting picture of one camera from the front to the time it disappears from the shooting picture of the other camera from the back.

[0091] For example, the farthest detection distance of the two cameras can be detected and set in advance.

[0092] Step 2: Determine a first ratio between the farthest detection distance and the preset speed.

[0093] Exemplarily, the preset speed is the average speed of the object, for example, 60 m / min.

[0094] Step three: multiply the first ratio by the preset multiple as the elimination time.

[0095] Exemplarily, the preset multiple is 2. In other examples, the preset multiple may also be other values.

[0096] In a specific application, the elimination time can be expressed as N / 30 (minutes), where N is the aforementioned maximum detection distance.

[0097] In this embodiment, when the number of first objects corresponding to the first data set is greater than the target number (ie, the cache size occupied by the first data set exceeds the cache limit), the positive physical sign information and positive attribute information of some first objects in the first data set are deleted.

[0098] Similarly, when the number of second objects corresponding to the second data set is greater than the target number (ie, the cache size occupied by the second data set exceeds the cache limit), the positive physical sign information and positive attribute information of some second objects in the second data set are deleted.

[0099] For example, the positive vital sign information and positive attribute information of the first subject whose storage time exceeds the time threshold is deleted from the first data set, and the positive vital sign information and positive attribute information of the second subject whose storage time exceeds the time threshold is deleted from the second data set. The time threshold may be the same as or different from the aforementioned elimination time.

[0100] In one embodiment, the target quantity is a preset fixed value.

[0101] In another embodiment, the target number is dynamically determined based on the size of the current scene's pedestrian flow, so that the target number is adapted to the size of the current scene's pedestrian flow. Specifically, the target number is determined by the following steps:

[0102] Step 1: Count the total number of first objects passing by the second camera within a preset time period.

[0103] Exemplarily, the preset duration is 5 minutes.

[0104] Step 2: Determine a second ratio between the total number and the preset time length.

[0105] Step three: take the product of the second ratio and the elimination time as the target number.

[0106] In a specific application, the target number can be expressed as Q*N / 30 (persons), where Q is the second ratio, i.e., the size of the human flow, and N / 30 represents the elimination time.

[0107] See also Figure 4 , Figure 4 This is a schematic diagram of the structure of an embodiment of the attribute information detection system provided by this application. Figure 4 As shown, the attribute information detection system includes two cameras arranged in back-to-back positions, namely a first camera and a second camera. The two cameras are arranged in back-to-back positions, which means that the lenses of the two cameras are facing back to back and the shooting directions are opposite. Since the two cameras are arranged in back-to-back positions, if the object to be detected enters the shooting picture of one of the cameras from the front, the object to be detected will leave the shooting picture of the other camera from the back. Figure 4As shown, the object to be detected enters the shooting picture of the first camera from the front, and then the object to be detected leaves the shooting picture of the second camera from the back.

[0108] In this embodiment, the two cameras can be two independent camera devices, and the installation distance between the two cameras is less than the preset distance. The preset distance is a shorter distance, such as 0.5m, 1m, 2m, etc. The specific value can be set according to actual needs. Or, Figure 4 As shown, two cameras are integrated into one camera device.

[0109] Furthermore, the attribute information detection system further includes a data processing unit ( Figure 2 Not shown in the figure), used to implement the attribute information detection method in this application.

[0110] In one embodiment, the data processing unit includes a first processor corresponding to the first camera, a second processor corresponding to the second camera, and a server.

[0111] In a specific example, the first processor is used to upload the image captured by the first camera to the server in real time, and the second processor is used to upload the image captured by the second camera to the server in real time. The server is used to obtain the first data set and the second data set based on the image captured by the first camera and the image captured by the second camera, and to execute the attribute information detection method in this application.

[0112] In this example, since the first processor and the second processor push data to the backend server for processing, the computation of the first processor and the second processor can be reduced, so that the first processor and the second processor can use low-cost processors.

[0113] In another specific example, the first processor is used to upload the second data set corresponding to the first camera to the server, and the second processor is used to upload the first data set corresponding to the second camera to the server. After extracting the back vital signs information of the object to be detected from the image captured by the first camera, the first processor obtains the first data set from the server, matches the back vital signs information of the object to be detected with the first data set, and then obtains the front attribute information of the object to be detected. After extracting the back vital signs information of the object to be detected from the image captured by the second camera, the second processor obtains the second data set from the server, matches the back vital signs information of the object to be detected with the second data set, and then obtains the front attribute information of the object to be detected.

[0114] In this example, the first processor and the second processor perform attribute information detection, but need to obtain a data set from the backend server.

[0115] In another embodiment, the data processing unit only includes a first processor corresponding to the first camera and a second processor corresponding to the second camera, and the first processor and the second processor can communicate with each other.

[0116] In this embodiment, the first processor is used to obtain and store a second data set corresponding to the first camera, and the second processor is used to obtain and store a first data set corresponding to the second camera. After the first processor extracts the back-face vital signs information of the object to be detected from the image captured by the first camera, it communicates with the second processor to obtain the first data set, matches the back-face vital signs information of the object to be detected with the first data set, and then obtains the front-face attribute information of the object to be detected. After the second processor extracts the back-face vital signs information of the object to be detected from the image captured by the second camera, it communicates with the first processor to obtain the second data set, matches the back-face vital signs information of the object to be detected with the second data set, and then obtains the front-face attribute information of the object to be detected.

[0117] In this embodiment, only the processor of the front-end camera device is required to participate in data processing, and the back-end server is not required to participate in data processing. When the object to be detected deviates from the shooting screen of the first camera or the shooting screen of the second camera, if the object to be detected generates a target event (such as running a red light) and it is necessary to obtain the front attribute information of the object to be detected, the front-end processor does not need to interact with the back-end server for data, so that it can further quickly match and obtain the front attribute information of the object to be detected.

[0118] In another embodiment, the first camera and the second camera are integrated into the same camera device. In this case, the data processing unit may include only one processor, that is, the first camera and the second camera share one processor, which is used to acquire and store the first data set, the second data set, and execute the attribute information detection method in this application.

[0119] In this embodiment, since only one processor is included, hardware resources are saved and the time for data interaction with other processors is reduced, so that the positive attribute information of the object to be detected can be further quickly matched and obtained.

[0120] In this embodiment, since the installation distance between the first camera and the second camera is short or the first camera and the second camera are integrated into the same camera device, the first data set and the second data set can only store query data for a short period of time before the current time (for example, 10 minutes, 5 minutes), and the time and space range is small, so that the front attribute information of the object to be detected can be obtained quickly and accurately. In the aforementioned second and third embodiments, since the time and space range of the first and second data sets is small, their calculation and storage requirements can be met by the processor of the front-end camera device. Therefore, the attribute information detection method of the present application can be implemented directly by the processor of the front-end camera device without the need for the back-end server to participate in data processing, and without the need to equip a complex and expensive back-end computing center or storage center, thereby reducing the amount of calculation and improving the accuracy and real-time performance of the obtained front attribute information of the object to be detected.

[0121] See also Figure 5 , Figure 5 It is a framework diagram of an embodiment of the attribute information detection device provided in the present application. In this embodiment, the attribute information detection device 50 includes a first acquisition module 51, a matching module 52 and a second acquisition module 53. The first acquisition module 51 is used to extract the back physical sign information of the object to be detected from the image captured by the first camera; wherein the first camera is one of the two cameras, and the two cameras are set back to back. The matching module 52 is used to match the back physical sign information of the object to be detected with the first data set to determine the first matching object; wherein the first data set includes the front physical sign information and front attribute information of several first objects extracted from the image captured by the second camera, the first matching object is one of the several first objects, and the second camera is the other of the two cameras. The second acquisition module 53 is used to use the front attribute information of the first matching object as the front attribute information of the object to be detected.

[0122] Optionally, the two cameras are integrated into one camera device; or, the two cameras are two independent camera devices, and the installation distance between the two cameras is less than a preset distance.

[0123] Optionally, the matching module 52 is configured to respectively determine first similarities between the back physical sign information of the object to be detected and the front physical sign information of each first object; and use the first object whose first similarity meets the first similarity condition as the first matching object.

[0124] Optionally, the attribute information detection device 50 further includes a third acquisition module 54. The third acquisition module 54 is configured to acquire the first data set. In response to detecting that the first subject's front face enters the capture image of the second camera, the third acquisition module 54 is configured to extract the first subject's frontal vital sign information and frontal attribute information from the image captured by the second camera; and to add the first subject's frontal vital sign information and frontal attribute information to the first data set.

[0125] Optionally, the third acquisition module 54 is used to match the front information of the first object with the second data set to determine whether there is a second matching object that matches the first object; wherein the front information of the first object includes at least one of the front physical sign information and front attribute information of the first object, the second data set includes the front physical sign information and front attribute information of several second objects extracted from the captured image of the first camera, and the second matching object is one of the several second objects; in response to the absence of the second matching object, the front physical sign information and front attribute information of the first object are added to the first data set; in response to the existence of the second matching object, the front physical sign information and front attribute information of the second matching object are deleted from the second data set, and the front physical sign information and front attribute information of the first object are added to the first data set.

[0126] Optionally, the attribute information detection module further includes an elimination module 55. After the second acquisition module 53 uses the front attribute information of the first matching object as the front attribute information of the object to be detected, the elimination module 55 is configured to delete the front physical sign information and front attribute information of the first matching object from the first data set.

[0127] Optionally, for the positive physical sign information and positive attribute information of each first object in the first data set, the elimination module 55 is also used to delete the positive physical sign information and positive attribute information of the first object from the first data set in response to the storage time of the positive physical sign information and positive attribute information of the first object being greater than the elimination time.

[0128] Optionally, the elimination module 55 is used to determine the farthest detection distance of the two cameras; determine a first ratio between the farthest detection distance and a preset speed; and use the product of the first ratio and a preset multiple as the elimination time.

[0129] Optionally, the elimination module 55 is configured to delete the positive physical sign information and positive attribute information of some of the first subjects in the first dataset in response to the number of the first subjects corresponding to the first dataset being greater than the target number.

[0130] Optionally, the elimination module 55 is used to count the total number of first objects passing through the second camera within a preset time period; determine a second ratio between the total number and the preset time period; and use the product of the second ratio and the elimination time period as the target number.

[0131] It should be noted that the device of this embodiment can execute the steps in the above method. For detailed description of the relevant content, please refer to the above method part, which will not be repeated here.

[0132] It should be noted that if the technical solution of this application involves personal information, the product using the technical solution of this application has clearly informed the personal information processing rules and obtained the individual's voluntary consent before processing personal information. If the technical solution of this application involves sensitive personal information, the product using the technical solution of this application has obtained the individual's separate consent before processing sensitive personal information, and at the same time meets the "explicit consent" requirement. For example, on personal information collection devices such as cameras, a clear and prominent sign is set to inform that the personal information collection scope has been entered and personal information will be collected. If the individual voluntarily enters the collection scope, it is deemed that they agree to the collection of their personal information; or on the personal information processing device, when the personal information processing rules are notified by obvious signs / information, the individual's authorization is obtained through pop-up information or by asking the individual to upload their personal information; among which, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the type of personal information processed.

[0133] See also Figure 6 , Figure 6 6 is a schematic diagram of a framework of an embodiment of an electronic device provided by the present application. In this embodiment, the electronic device 60 includes a memory 61 and a processor 62.

[0134] The processor 62 may also be referred to as a CPU (Central Processing Unit). The processor 62 may be an integrated circuit chip having signal processing capabilities. The processor 62 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component. The general-purpose processor may be a microprocessor, or the processor 62 may be any conventional processor 62.

[0135] The memory 61 in the electronic device 60 is used to store program instructions required for the processor 62 to run.

[0136] The processor 62 is configured to execute program instructions to implement the attribute information detection method of the present application.

[0137] See also Figure 7 , Figure 7: It is a schematic diagram of a framework of an embodiment of a computer-readable storage medium provided by the present application. The computer-readable storage medium 70 of the embodiment of the present application stores program instructions 71, and when the program instructions 71 are executed, the attribute information detection method provided by the present application is implemented. Among them, the program instructions 71 can form a program file and be stored in the above-mentioned computer-readable storage medium 70 in the form of a software product, so that a computer device (which can be a personal computer, a server, or a network device, etc.) executes all or part of the steps of the various implementation methods of the present application. The aforementioned computer-readable storage medium 70 includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or terminal devices such as a computer, a server, a mobile phone, and a tablet.

[0138] The above scheme designs two cameras that are set up to face each other, with the first camera being one of the two cameras and the second camera being the other of the two cameras. Since the two cameras are set up to face each other, if the object to be detected enters the shooting picture of one of the cameras from the front, the object to be detected will leave the shooting picture of the other camera with its back facing away. When the object to be detected leaves the shooting picture of the first camera, only the back physical sign information of the object to be detected can be extracted from the shooting picture of the first camera, but the front attribute information of the object to be detected can be obtained from the shooting picture obtained by the second camera that faces away. Therefore, the back physical sign information of the object to be detected extracted from the shooting picture of the first camera is matched with the first data set to determine the first matching object. The first data set includes the front physical sign information and front attribute information of several first objects extracted from the shooting picture of the second camera. The first matching object is one of the several first objects. Afterwards, the front attribute information of the first matching object can be used as the front attribute information of the object to be detected. In this way, the front attribute information of the object to be detected can be obtained quickly and accurately.

[0139] In some embodiments, the functions or modules included in the device provided by the embodiments of the present disclosure can be used to execute the method described in the above method embodiment. Its specific implementation can refer to the description of the above method embodiment. For the sake of brevity, it will not be repeated here.

[0140] The above description of the various embodiments tends to emphasize the differences between the various embodiments. The same or similar aspects can be referenced with each other and will not be repeated herein for the sake of brevity.

[0141] In the several embodiments provided in this application, it should be understood that the disclosed methods, devices and systems can be implemented in other ways. For example, the device implementation methods described above are only schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0142] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0143] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0144] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of each embodiment method of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0145] The above description is only an implementation method of the present application and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the description and drawings of this application, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for detecting attribute information, characterized in that: The method comprises: Extracting back physical sign information of the object to be detected from an image captured by a first camera; wherein the first camera is one of two cameras, and the two cameras are arranged to face each other in reverse; Matching the back physical sign information of the object to be detected with a first data set to determine a first matching object; wherein the first data set includes front physical sign information and front attribute information of a plurality of first objects extracted from images captured by a second camera, the first matching object is one of the plurality of first objects, and the second camera is the other of the two cameras; The front attribute information of the first matching object is used as the front attribute information of the object to be detected.

2. The method according to claim 1, characterized in that The two cameras are integrated into one camera device; Alternatively, the two cameras are two independent camera devices, and the installation distance between the two cameras is less than a preset distance.

3. The method according to claim 1, characterized in that The step of matching the back physical sign information of the subject to be detected with the first data set to determine a first matching subject includes: respectively determining first similarities between the back physical sign information of the subject to be detected and the front physical sign information of each of the first subjects; The first object whose first similarity meets the first similarity condition is used as the first matching object.

4. The method according to claim 1, wherein The step of obtaining the first data set includes: In response to detecting that the first object enters the shooting picture of the second camera from the front, extracting the front physical sign information and front attribute information of the first object from the image captured by the second camera; The positive physical sign information and positive attribute information of the first subject are added to the first data set.

5. The method according to claim 4, characterized in that The adding the positive physical sign information and positive attribute information of the first subject to the first data set includes: matching the frontal information of the first object with a second data set to determine whether there is a second matching object that matches the first object; wherein the frontal information of the first object includes at least one of frontal physical sign information and frontal attribute information of the first object, the second data set includes frontal physical sign information and frontal attribute information of a plurality of second objects extracted from images captured by the first camera, and the second matching object is one of the plurality of second objects; In response to the absence of the second matching object, adding the positive physical sign information and positive attribute information of the first object to the first data set; In response to the existence of the second matching object, the positive physical sign information and positive attribute information of the second matching object are deleted from the second data set, and the positive physical sign information and positive attribute information of the first object are added to the first data set.

6. The method according to claim 1, wherein After using the front attribute information of the first matching object as the front attribute information of the object to be detected, the method further includes: The positive physical sign information and positive attribute information of the first matching object are deleted from the first data set.

7. The method according to claim 1, characterized in that The method further comprises: For the positive physical sign information and positive attribute information of each of the first objects in the first data set, in response to the storage time of the positive physical sign information and positive attribute information of the first object being greater than the elimination time, the positive physical sign information and positive attribute information of the first object are deleted from the first data set.

8. The method according to claim 7, characterized in that The steps of determining the elimination time include: Determine the maximum detection distance between the two cameras; determining a first ratio between the farthest detection distance and a preset speed; The product of the first ratio and the preset multiple is used as the elimination time.

9. The method according to claim 1, characterized in that The method further includes: in response to the number of the first subjects corresponding to the first data set being greater than a target number, deleting positive physical sign information and positive attribute information of part of the first subjects in the first data set.

10. The method according to claim 9, characterized in that The steps of determining the target quantity include: Counting the total number of the first objects passing by the second camera within a preset time period; determining a second ratio between the total number and the preset duration; The product of the second ratio and the elimination time is used as the target number.

11. An electronic device, characterized in that: comprising a memory and a processor coupled to each other, The memory stores program instructions; The processor is configured to execute program instructions stored in the memory to implement the method according to any one of claims 1 to 10.

12. A computer-readable storage medium, characterized in that The computer-readable storage medium is used to store program instructions, and the program instructions can be executed by a processor to implement the method according to any one of claims 1 to 10.

Images