Object recognition method and apparatus

Abstract

The present specification provides an object recognition method and device, wherein the object recognition method comprises: acquiring a to-be-recognized depth image and preset key point information; extracting depth information of at least two to-be-recognized key points of a gate opening and closing area in the to-be-recognized depth image according to the preset key point information; comparing the depth information of each to-be-recognized key point with the preset key point information respectively to obtain a key point comparison result of each to-be-recognized key point; and recognizing whether a target object exists in the gate opening and closing area according to the key point comparison result of each to-be-recognized key point. On the basis of improving the accuracy of object recognition, the safety of the target object in passing through the gate opening and closing area can be further improved, and the risk of injuring pedestrians or damaging other objects when the gate door is opened or closed can be avoided.

Description

TECHNICAL FIELD

[0001] The present specification relates to the technical field of computer vision, and in particular, to an object recognition method. The present specification also relates to an object recognition apparatus, a computing device, a computer-readable storage medium, and a computer program product. BACKGROUND

[0002] Urban rail transit modes such as subways and light rails have become an important way for urban people to travel. During the peak period of people entering and leaving the station, the personnel passing efficiency and passing judgment accuracy of the gate machine affect the passing experience of people. The actions of the gate machine door include opening, closing and emergency stopping. Before opening or closing, the gate machine door needs to detect whether there are other objects in the opening and closing area of the gate machine door to prevent injury to pedestrians or damage to objects in the opening and closing area of the gate machine door during the opening or closing process. In the current actual application, in the gate machine based on image recognition for gate passing logic judgment, the image during the operation of the gate machine is usually collected, and a target detection model is used to detect the target in the image to determine whether there are pedestrians or other objects in the gate machine area. However, this implementation method has the risk of causing injury to pedestrians or other objects when the gate machine door is opened or closed, because the target detection model does not detect pedestrians or other objects in the image, but there are actually pedestrians or other objects. SUMMARY

[0003] Therefore, an object recognition method is provided in the embodiments of the present specification. One or more embodiments of the present specification also relate to an object recognition apparatus, a computing device, a computer-readable storage medium, and a computer program product to solve the technical defects in the prior art.

[0004] According to a first aspect of the embodiments of the present specification, an object recognition method is provided, including:

[0005] obtaining a to-be-recognized depth image and preset key point information;

[0006] extracting depth information of at least two to-be-recognized key points of a gate opening and closing area in the to-be-recognized depth image according to the preset key point information;

[0007] comparing the depth information of each to-be-recognized key point with the preset key point information respectively to obtain a key point comparison result of each to-be-recognized key point;

[0008] recognizing whether there is a target object in the gate opening and closing area according to the key point comparison result of each to-be-recognized key point.

[0009] According to a second aspect of the embodiments of the present specification, an object recognition apparatus is provided, including:

[0010] an acquisition module configured to acquire a to-be-identified depth image and preset key point information;

[0011] an extraction module configured to extract, according to the preset key point information, depth information of at least two to-be-identified key points of a gate opening and closing region in the to-be-identified depth image;

[0012] a comparison module configured to compare the depth information of each to-be-identified key point with the preset key point information respectively, to obtain a key point comparison result of each to-be-identified key point;

[0013] an identification module configured to identify, according to the key point comparison result of each to-be-identified key point, whether a target object exists in the gate opening and closing region.

[0014] According to a third aspect of an embodiment of the present specification, a computing device is provided, comprising:

[0015] a memory and a processor;

[0016] The memory is configured to store computer programs / instructions, and the processor is configured to execute the computer programs / instructions, which realize the steps of the above object identification method when executed by the processor.

[0017] According to a fourth aspect of an embodiment of the present specification, a computer readable storage medium is provided, which stores computer programs / instructions, which realize the steps of the above object identification method when executed by the processor.

[0018] According to a fifth aspect of an embodiment of the present specification, a computer program product is provided, comprising computer programs / instructions, which realize the steps of the above object identification method when executed by the processor.

[0019] An embodiment of the specification realizes determining a gate opening and closing area in a gate passing area, setting a plurality of key points in the gate opening and closing area in advance, and acquiring preset key point information of the plurality of key points, so as to determine whether a target object exists in the gate opening and closing area by acquiring a to-be-recognized depth image, extracting to-be-recognized key points corresponding to the preset key points in the to-be-recognized depth image, and comparing depth information of each to-be-recognized key point in the to-be-recognized depth image with the preset key point information. By determining the gate opening and closing area in the gate passing area, the identification area for identifying the target object is reduced, the calculation workload is reduced, and the accuracy of identifying the target object is improved. By acquiring the to-be-recognized depth image and setting the preset key points for object identification, the detection error caused by using a target detection model to detect the target in the image is avoided, and the accuracy of object identification is further improved. Then, on the basis of improving the accuracy of object identification, the safety of the target object passing through the gate opening and closing area can be further improved, and the pedestrians or other objects are prevented from being injured or damaged when the gate door is opened or closed. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 is an application scenario of an object identification method provided by an embodiment of the specification;

[0021] Figure 2 is a flowchart of an object identification method provided by an embodiment of the specification;

[0022] Figure 3 is a schematic diagram of a gate opening and closing area and a preset key point provided by an embodiment of the specification;

[0023] Figure 4 is a process flowchart of an object identification method provided by an embodiment of the specification;

[0024] Figure 5 is a structural schematic diagram of an object identification device provided by an embodiment of the specification;

[0025] Figure 6 is a structural block diagram of a computing device provided by an embodiment of the specification. DETAILED DESCRIPTION

[0026] In the following description, a large number of specific details are set forth in order to facilitate a thorough understanding of the specification. However, the specification can be implemented in many different ways from those described herein, and those skilled in the art can make similar extensions without departing from the connotation of the specification, so the specification is not limited to the specific implementation disclosed below.

[0027] The terminology used in this disclosure, one or more embodiments of the present specification, is for the purpose of describing particular embodiments only and is not intended to be limiting of one or more embodiments of the present specification. As used in this disclosure and the appended claims herein, the singular forms "a," "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "and / or," as used in this disclosure, refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0028] It will be understood that, although the terms first, second, etc. can be employed in this disclosure, one or more embodiments of the present specification, to describe various information, these information should not be limited to these terms. These terms are only used to differentiate one piece of information from another piece of information. For example, a first can also be referred to as a second, and similarly, a second can also be referred to as a first, without departing from the scope of one or more embodiments of the present specification. Depending on the context, the word "if' as used herein can be interpreted to mean "when" or "in response to determining".

[0029] In addition, it should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in one or more embodiments of the present specification are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation portal for user to choose authorization or refusal.

[0030] First, the nomenclature involved in one or more embodiments of the present specification is explained.

[0031] Structured light camera: a 3D imaging device based on the principle of active stereo vision, which uses structured light technology to obtain the three-dimensional geometric information of an object.

[0032] RGB image: a color image representation method based on the combination of red (Red), green (Green), and blue (Blue) color channels. In an RGB image, the color of each pixel is determined by the intensity values of the three channels, and the intensity range of each channel is usually 0 to 255. In digital image processing and computer graphics, the data of an RGB image is usually organized into a three-dimensional array, where two dimensions represent the width and height of the image, and the third dimension represents the three color channels of red, green, and blue.

[0033] Subways, light rail, and other urban rail transit systems have become essential modes of transportation for city dwellers. During peak hours, the efficiency and accuracy of turnstiles in determining passenger flow significantly impact the overall experience. Turnstile operations include opening, closing, and emergency stopping. Before opening or closing, turnstiles must detect the presence of objects within the opening / closing area to prevent accidental injury to pedestrians or damage to objects in the area. In current applications, turnstiles that rely on image recognition for their logic typically use image capture during operation and object detection models to determine the presence of pedestrians or other objects. However, this approach has limitations. There are instances where the detection model fails to detect pedestrians or other objects in the image, even when such objects are present. In such cases, there is a risk of accidental injury to pedestrians or other objects when the turnstile opens or closes.

[0034] This specification provides an object recognition method, and also relates to an object recognition device, a computing device, a computer-readable storage medium, and a computer program product, which will be described in detail in the following embodiments.

[0035] See Figure 1 , Figure 1 This diagram illustrates an application scenario of an object recognition method according to an embodiment of this specification. The application scenario of a subway turnstile is used as an example for illustration. Figure 1 As shown, passengers are moving through the turnstile's passage area along the direction of travel. In practical applications, passengers need to swipe their cards at the turnstile's sensor area before entering to open the gate. Only after the gate is open can the passenger pass through. During this process, the turnstile needs to identify whether there are pedestrians or other objects in the passage area before opening or closing to avoid accidental injury to pedestrians or damage to objects. However, during the passenger's passage through the turnstile, misidentification may occur, potentially resulting in accidental injury to pedestrians or damage to objects within the turnstile's opening and closing area.

[0036] In actual application, if the passenger is injured by the gate door, it indicates that the passenger is currently located in the operation range of the gate door during actual operation, i.e., the gate opening and closing area. Therefore, in order to improve the safety of the passenger passing through the gate passage area, i.e., to improve the accurate identification of the passenger in the gate passage area during actual operation of the gate, the present specification can specifically obtain the to-be-identified depth image and the preset key point information during actual operation of the gate, accurately determine the gate opening and closing area in the to-be-identified depth image through the preset key point information, and extract the to-be-identified key point corresponding to the preset key point and the depth information of the to-be-identified key point in the gate opening and closing area of the to-be-identified depth image. Further, the depth information of each to-be-identified key point extracted is compared with the preset key point information, so as to identify whether there is a target object (passenger or object) in the gate opening and closing area after obtaining the key point comparison result of each to-be-identified key point.

[0037] In one embodiment of the present specification, the to-be-identified depth image is obtained, and the to-be-identified depth image is processed in the subsequent object identification process, without the need to collect a color image for target detection, thereby improving the accuracy of target detection by using a target detection model; the gate opening and closing area is determined in the to-be-identified depth image through the preset key point information, the identification area for object identification is reduced, and the workload of subsequent object identification is reduced; after the depth information of the to-be-identified key point is extracted, the depth information of each to-be-identified key point is compared with the preset key point information, and whether there is a target object in the gate opening and closing area is identified through the key point comparison result, thereby improving the accuracy of object identification in the gate opening and closing area.

[0038] It should be noted that the above subway gate application scenario is only used to explain and describe the object identification method provided by the present specification, and the object identification method provided by the present specification is not limited to the subway gate application scenario, but can also be applied to intelligent automatic door scenarios, access control gate scenarios, etc.

[0039] Referring to Figure 2 , Figure 2 A flowchart of an object identification method according to one embodiment of the present specification is shown, which specifically includes the following steps:

[0040] Step 202: Obtain a to-be-identified depth image and preset key point information.

[0041] The to-be-identified depth image refers to the gate depth image obtained by the image acquisition device during actual operation of the gate. The image acquisition device can be specifically a structured light camera (which can be specifically a monocular structured light camera or a binocular structured light camera) or other depth cameras.

[0042] Preferably, in one or more embodiments provided in the specification, the image acquisition device adopts a top-down perspective, i.e., the device lens faces the ground, and the line connecting the physical center of the camera lens and the focal point of the lens is perpendicular to the ground. Based on this, the to-be-identified depth image obtained by the image acquisition device is also a top-down perspective gate depth image.

[0043] Since the gate opening and closing area cannot be determined in the obtained to-be-identified depth image, the gate opening and closing area can be determined in the to-be-identified depth image by obtaining preset key point information. The preset key point information specifically includes pixel position information and depth range information of at least two preset key points determined in advance. The pixel position information can specifically be coordinate position information of the preset key points in a pixel coordinate system; the depth range information refers to a depth range corresponding to the preset key points, and the depth range information can be one or multiple. If the depth range information is one, multiple preset key points correspond to the same depth range information. If the depth range information is multiple, the number of the depth range information is the same as the number of the preset key points.

[0044] Since the preset key point information is a key for object recognition of the to-be-identified depth image, it is crucial to determine and generate the preset key point information before object recognition in the actual operation of the gate.

[0045] In a specific embodiment provided in the specification, before obtaining the to-be-identified depth image and the preset key point information, the method further includes:

[0046] obtaining a static color image of a target gate and a static depth image corresponding to the static color image;

[0047] determining a gate opening and closing area of the target gate and at least two preset key points of the gate opening and closing area in the static color image;

[0048] determining color space position information of the at least two preset key points, and determining depth space position information of the at least two preset key points in the static depth image according to the color space position information;

[0049] determining pixel position information and pixel depth information of the at least two preset key points according to the depth space position information;

[0050] generating preset key point information of the at least two preset key points according to the pixel position information and the pixel depth information.

[0051] The target gate is any gate in the gate passage area of the target object or any gate not in the running state. The static color image is a static color image in which the target object does not exist in the gate passage area of the target gate or is not in the running state. The static depth image is a static depth image in which the target object does not exist in the gate passage area of the target gate or is not in the running state. The static color image and the static depth image are obtained at the same time.

[0052] The gate opening and closing area is specifically the area in which the gate door can reach when actually running. The color space position information is the coordinate position information of the color pixel point in the color camera space coordinate system, and here specifically the coordinate position information of the preset key point in the color camera space coordinate system. The depth space position information is the coordinate position information of the depth pixel point in the depth camera space coordinate system, and here specifically the coordinate position information of the preset key point in the depth camera space coordinate system. The pixel position information is the coordinate position information of the depth pixel point in the pixel coordinate system, and here specifically the coordinate position information of the preset key point in the pixel coordinate system. The pixel depth information is the depth value of the depth pixel point, and here specifically the depth value of the preset key point.

[0053] Specifically, the static color image in which the target object does not exist or is not in the running state in the gate passage area of the target gate is obtained using the color camera, and the static depth image in which the target object does not exist or is not in the running state in the gate passage area of the target gate is obtained using the depth camera. The gate opening and closing area of the target gate is identified in the static color image, and a plurality of preset key points in the gate opening and closing area are determined. The preset key points can be obtained by manual labeling in actual application, or can be obtained by training a key point labeling model, and the key point labeling model can be specifically used to label a plurality of preset key points in the static color image. The preset key point acquisition method can be determined according to actual application, which is not limited in the present specification.

[0054] After obtaining a plurality of preset key points in the static color image, the coordinate position information of each preset key point in the color camera space coordinate system, i.e., the color space position information, is determined, which can be specifically determined by solving the intrinsic and extrinsic parameters of the color camera.

[0055] Since the object recognition method provided in the specification is based on a to-be-recognized depth image to recognize a target object, pixel position information of the plurality of preset key points needs to be further determined. Specifically, depth space position information of the preset key points in the static depth image can be determined according to the determined color space position information, and then the pixel position information of the preset key points is calculated using the depth space position information of the preset key points, and pixel depth information corresponding to the preset key points is determined. Specifically, the depth space position information of the preset key points in the static depth image can be determined according to the following formula 1:

[0056] Formula 1

[0057] wherein, is the depth space position information, is the color space position information, is a rotation conversion matrix, is a translation vector. The depth space position information of each preset key point can be calculated according to the color space position information of each preset key point by using the above formula 1. In actual application, the pixel position information of the preset key points can be calculated according to the conversion relationship between the depth camera space coordinate system and the pixel coordinate system.

[0058] Further, referring to Figure 3 , Figure 3 a schematic diagram of a gate opening and closing area and a preset key point according to one embodiment of the specification is shown. As Figure 3 shown, the dashed box area is specifically a static color image of a target gate, the solid box area is specifically a region where the target gate is located, Figure 3 the shaded part area shown in is a gate passing area of the target gate, Figure 3 the dashed line part area shown in is a gate opening and closing area of the target gate, and points A, B, C, D, E, F, G, H, and K are preset preset key points, and the coordinate position information of each preset key point in the color camera space coordinate system, i.e., the color space position information, is A(x1, y4), B((x1+x2) / 2, y4), C(x2, y4), D(x1, (y3+y4) / 2), E((x1+x2) / 2, (y3+y4) / 2), F(x2, (y3+y4) / 2), G(x1, y3), H((x1+x2) / 2, y3), and K(x2, y3), respectively. After determining the color space position information of each preset key point, the pixel position information and the pixel depth information of each preset key point can be determined based on the above method, so that the preset key point information of the preset key point can be generated according to the pixel position information and the pixel depth information in the subsequent process.

[0059] It should be noted that the number of preset key points can be adjusted according to the actual application, such as 9, 12, 15, etc. The arrangement of the preset key points can also be adjusted according to the actual application, such as rectangle, star shape, etc.

[0060] In one specific embodiment provided in this specification, the preset key point information of the at least two preset key points is generated based on the pixel position information and the pixel depth information, including:

[0061] Determine the target key point from the at least two preset key points, and determine the target pixel depth information of the target key point from the pixel depth information;

[0062] Based on the target pixel depth information, calculate the target static mean feature and target static difference feature of the target key points;

[0063] Based on the target static mean feature and the target static difference feature, the target depth range information of the target key points is generated;

[0064] The target pixel position information of the target key point is determined from the pixel position information, and the target pixel position information and the target depth range information are determined as the preset key point information of the target key point.

[0065] Here, "target keypoint" refers to any one of the preset keypoints. "Target pixel depth information" refers to the pixel depth information of the target keypoint. "Target static mean feature" refers to the depth mean feature of the target keypoint, specifically the mean depth value of the target keypoint in each static depth image. "Target static difference feature" refers to the depth standard deviation feature of the target keypoint, specifically the standard deviation of the depth value of the target keypoint in each static depth image. "Target depth range information" refers to the depth range information of the target keypoint. "Target pixel position information" refers to the pixel position information of the target keypoint.

[0066] Specifically, taking any one of the preset key points (i.e., the target key point) as an example, the target pixel depth information of the target key point is determined from the pixel depth information of each preset key point, and the target static mean feature of the target key point is calculated based on the target pixel depth information. The calculation method is detailed in Formula 2 below:

[0067] Formula 2

[0068] in, For the target static mean feature, j The value is determined based on the number of preset key points. The number of static depth images and static color images. For the firstj a preset key point in the first i depth information in the static depth image.

[0069] In actual applications, n sets of static color images and static depth images are collected, and the static mean features of each preset key point are calculated through the above formula 2. The number of static color images and static depth images can be determined according to actual application conditions, for example ≥1000, etc. This specification does not limit it here.

[0070] Still taking any one of the preset key points as an example, after obtaining the target static mean feature of the target key point, the target static difference feature of the target key point can be calculated according to the target static mean feature, as shown in the following formula 3:

[0071] Formula 3

[0072] Wherein, S is the target static difference feature. The target static difference feature of the target key point is calculated based on the above formula 3. Since the target object does not exist in the static depth image, the depth information corresponding to each preset key point is the floor depth value of the gate opening and closing area. Since the gate door bottom of the gate opening and closing area is often a certain distance from the floor, the depth range from the gate door bottom to the floor does not need to be considered in the subsequent object recognition process. Therefore, in order to further improve the accuracy of subsequent object recognition, the depth range from the gate door bottom to the floor can be ignored.

[0073] Based on this, after obtaining the target static mean feature and the target static difference feature of the target key point, the distance between the gate door bottom and the floor of the gate opening and closing area is obtained, so as to generate the target depth range information of the target key point according to the target static mean feature, the target static difference feature and the distance between the gate door bottom and the floor of the gate opening and closing area. The target depth range information can be specifically expressed as , wherein, is the distance between the gate door bottom and the floor of the gate opening and closing area. The value range of the target depth range information can also be fine-tuned according to the depth information distribution of each preset key point.

[0074] The target pixel position information of the target key point is determined in the pixel position information of each preset key point. The target pixel position information and the target depth range information determined above are the preset key point information of the target key point.

[0075] The generation process of the preset key point information of any one of the preset key points is as described above, and the preset key point information of other preset key points is also generated based on the above method to obtain the preset key point information of each preset key point. Thus, in the actual operation process of the gate, the object recognition of the gate opening and closing area can be performed according to the preset key point information of each preset key point using the to-be-recognized depth image.

[0076] In one or more embodiments provided in the specification, a plurality of sets of static color images and static depth images are obtained, the gate opening and closing area is determined using the static color images, a plurality of preset key points are determined in the static color images, and color space position information of each preset key point is determined. According to the color space position information, depth space position information of each preset key point in the static depth image is determined, pixel position information and pixel depth information of each preset key point are determined according to the depth space position information, and after static mean features and static difference features of each preset key point are generated according to the pixel depth information, depth range information of each preset key point is generated, so as to determine the pixel position information and the depth range information as the preset key point information of each preset key point. The gate opening and closing area is determined based on the static color image, the preset key point information of each preset key point is determined using the static depth image, so as to determine the corresponding to-be-recognized key point in the to-be-recognized depth image according to the preset key point information in the subsequent process, thereby completing the object recognition process.

[0077] Step 204: According to the preset key point information, depth information of at least two to-be-recognized key points of the gate opening and closing area in the to-be-recognized depth image is extracted.

[0078] After obtaining the to-be-recognized depth image and the preset key point information, since the gate opening and closing area cannot be determined based on only the to-be-recognized depth image, the gate opening and closing area of the gate can be determined in the to-be-recognized depth image according to the preset key point information, specifically, the key points in the gate opening and closing area are determined.

[0079] The to-be-recognized key point refers to a key point extracted in the to-be-recognized depth image. After a plurality of to-be-recognized key points are extracted in the to-be-recognized depth image, depth information of each to-be-recognized key point is obtained. The to-be-recognized key point and the depth information of the to-be-recognized key point are extracted based on the preset key point information, and the implementation manner is specifically as follows:

[0080] In a specific embodiment provided in the specification, according to the preset key point information, depth information of at least two to-be-recognized key points of the gate opening and closing area in the to-be-recognized depth image is extracted, including:

[0081] The target pixel position information of the target key point is obtained from the preset key point information, wherein the target key point is any one of the preset key points.

[0082] The target key point is determined from the target pixel position information in the to-be-recognized depth image, and the to-be-processed depth information of the target key point is extracted.

[0083] The to-be-processed key point refers to a to-be-recognized key point with the same target pixel position information as the target key point. The to-be-processed depth information refers to the depth information of the to-be-processed key point.

[0084] Specifically, still taking any one of the preset key points (target key point) as an example, the target pixel position information of the target key point is obtained from the preset key point information, and the to-be-recognized key point with the same target pixel position information as the target key point in the to-be-recognized depth image is determined as the to-be-processed key point corresponding to the target key point according to the target pixel position information, and the to-be-processed depth information of the to-be-processed key point is extracted. According to the same method, the to-be-recognized key point corresponding to each preset key point in the to-be-recognized depth image is extracted, and the depth information of each to-be-recognized key point is extracted. For example, the preset key points specifically include points A, B, C, D, E, F, G, H, and K, and A', B', C', D', E', F', G', H', and K' corresponding to A, B, C, D, E, F, G, H, and K can be determined in the to-be-recognized depth image according to the pixel position information of A, B, C, D, E, F, G, H, and K.

[0085] In one embodiment of the present specification, the to-be-recognized key points are extracted from the to-be-recognized depth image by the pixel position information of each preset key point in the preset key point information, and the depth information of each to-be-recognized key point is determined. Thus, in the actual operation of the gate, only the to-be-recognized depth image of the gate needs to be obtained, and the gate opening and closing area of the gate and the to-be-recognized key points in the gate opening and closing area can be determined in the to-be-recognized depth image, without the need to obtain a color image to determine the gate opening and closing area, avoiding the error recognition of object recognition based on the color image, improving the accuracy of subsequent object recognition, and reducing the computational workload.

[0086] Step 206: Comparing the depth information of each to-be-recognized key point with the preset key point information respectively to obtain the key point comparison result of each to-be-recognized key point.

[0087] The depth information of the to-be-identified key point is extracted from the to-be-identified depth image, and the depth information of the to-be-identified key point is determined. The depth information of the to-be-identified key point can be compared with the preset key point information to obtain the key point comparison result of the to-be-identified key point. In actual application, the depth information of each to-be-identified key point is compared with the depth range information in the preset key point information, so as to determine whether the depth information of each to-be-identified key point is located in the depth range information.

[0088] Based on this, in a specific embodiment provided in the specification, the depth information of each to-be-identified key point is compared with the preset key point information respectively to obtain the key point comparison result of each to-be-identified key point, including:

[0089] Obtaining target depth range information of the target key point in the preset key point information;

[0090] Judging whether the to-be-processed depth information is located in the target depth range information to obtain the to-be-processed key point comparison result of the to-be-processed key point.

[0091] Wherein, the to-be-processed key point comparison result refers to the key point comparison result of the to-be-processed key point.

[0092] Specifically, taking any one of the to-be-identified key points (i.e. the to-be-processed key point) as an example, the target depth range information of the target key point corresponding to the to-be-processed key point is obtained in the preset key point information, it is judged whether the depth information of the to-be-processed key point (i.e. the to-be-processed depth information) is located in the target depth range information, and the to-be-processed key point comparison result of the to-be-processed key point is obtained.

[0093] For example, the depth information of the to-be-identified key point A' is , the target depth range information of the preset key point A is , it is judged whether is located in , and the key point comparison result of the to-be-identified key point A' is generated.

[0094] In addition, in order to reduce the complexity of subsequent object identification on the opening and closing area of the gate, the unified depth range information corresponding to each preset key point can also be generated according to the static mean value feature and the static difference value feature of each preset key point, and the depth information of each to-be-identified key point is compared with the unified depth range information respectively, and the key point comparison result corresponding to each to-be-identified key point is generated. The generation mode of the depth range information and the mode of comparing the depth range information with the depth information of each to-be-identified key point can be determined according to actual application, which is not limited in the specification.

[0095] One embodiment provided in the specification realizes that the depth information of each to-be-identified key point determined in the to-be-identified depth image is compared with the depth range information of the corresponding preset key point, a key point comparison result is generated, and whether a target object exists in the opening and closing area of the gate can be identified according to the key point comparison result in a subsequent process, thereby improving the accuracy of object identification in the opening and closing area of the gate.

[0096] Step 208: identifying whether a target object exists in the opening and closing area of the gate according to the key point comparison result of each to-be-identified key point.

[0097] After obtaining the key point comparison result of each to-be-identified key point, object identification can be performed on the opening and closing area of the gate according to the key point comparison result of each to-be-identified key point, so as to identify whether a target object exists in the opening and closing area of the gate. The target object can be a pedestrian, an object, or the like.

[0098] Since the number of preset key points is often large, and since each to-be-identified key point corresponds to a preset key point, the number of to-be-identified key points extracted from the to-be-identified depth image is also large. In order to ensure the accuracy of object identification, object identification can be performed on the opening and closing area of the gate according to the key point comparison result of each to-be-identified key point, as follows:

[0099] In a specific embodiment provided in the specification, identifying whether a target object exists in the opening and closing area of the gate according to the key point comparison result of each to-be-identified key point includes:

[0100] In the case where the to-be-processed depth information is located in the target depth range information, it is determined that a target object exists in the opening and closing area of the gate.

[0101] Specifically, for any one of the to-be-identified key points (i.e., a to-be-processed key point), if the to-be-processed depth information of the to-be-processed key point is located in the target depth range information of the target key point corresponding thereto, it is determined that a target object exists in the opening and closing area of the gate. If the to-be-processed depth information of the to-be-processed key point is not located in the target depth range information, it can be determined whether the depth information of the next to-be-identified key point is located in the depth range information of the preset key point corresponding thereto, so as to determine whether a target object exists in the opening and closing area of the gate.

[0102] Based on this, in a specific embodiment provided in the specification, the method further includes:

[0103] In the case where the to-be-processed depth information is not located in the target depth range information, a new to-be-processed key point is determined as a to-be-traversed key point among the to-be-identified key points, and a to-be-traversed key point comparison result of the to-be-traversed key point is obtained.

[0104] In a case where the to-be-traversed depth information of the to-be-traversed key point is located in the to-be-traversed depth range information, it is determined that the target object exists in the opening and closing area of the gate.

[0105] In a case where the to-be-traversed depth information is not located in the to-be-traversed depth range information, the step of determining a new to-be-processed key point as a to-be-traversed key point in each to-be-identified key point and obtaining the to-be-traversed key point comparison result of the to-be-traversed key point is returned until the traversal of each to-be-identified key point is completed.

[0106] Specifically, if the to-be-processed depth information is not located in the target depth range information, a new to-be-processed key point is determined as a to-be-traversed key point (i.e., a next to-be-identified key point) in each to-be-identified key point, and the to-be-traversed key point comparison result of the to-be-traversed key point is obtained. If the to-be-traversed depth information of the to-be-traversed key point is located in the to-be-traversed depth range information of the preset key point corresponding thereto, it is determined that the target object exists in the opening and closing area of the gate. If the to-be-traversed depth information of the to-be-traversed key point is not located in the to-be-traversed depth range information, a new to-be-processed key point is determined again until the traversal of each to-be-identified key point is completed. If the depth information of each to-be-identified key point is not located in the depth range information corresponding thereto, it is determined that the target object does not exist in the opening and closing area of the gate.

[0107] Further, to improve the object identification efficiency of the opening and closing area of the gate, the key point comparison result of each to-be-identified key point can also be obtained, and whether the target object exists in the opening and closing area of the gate is identified according to the key point comparison result of each to-be-identified key point.

[0108] Based on this, in another specific embodiment provided in the specification, whether the target object exists in the opening and closing area of the gate is identified according to the key point comparison result of each to-be-identified key point, including:

[0109] In the key point comparison result of each to-be-identified key point, it is judged whether the depth information of the to-be-identified key point is located in the depth range information corresponding thereto.

[0110] In a case where the depth information of the to-be-identified key point is located in the depth range information corresponding thereto, it is determined that the target object exists in the opening and closing area of the gate.

[0111] Specifically, after obtaining the key point comparison result of each to-be-identified key point, in the key point comparison result of each to-be-identified key point, it is judged whether the depth information of the to-be-identified key point is located in the depth range information corresponding thereto. If the depth information of the to-be-identified key point is located in the depth range information corresponding thereto, it is determined that the target object exists in the opening and closing area of the gate.

[0112] In one or more embodiments provided in this specification, the presence of a target object in the gate opening / closing area can be identified sequentially by using the key point comparison results of each key point to be identified. Alternatively, the key point comparison results of each key point to be identified can be obtained simultaneously. Within these results, it can be determined whether the depth information of any key point to be identified falls within its corresponding depth range to identify the presence of a target object in the gate opening / closing area. By employing multiple object recognition methods to identify the presence of a target object in the gate opening / closing area, the flexibility and accuracy of object recognition are improved.

[0113] The object recognition method provided in this specification includes: acquiring a depth image to be recognized and preset key point information; extracting depth information of at least two key points to be recognized in the gate opening and closing area from the depth image to be recognized based on the preset key point information; comparing the depth information of each key point to be recognized with the preset key point information to obtain a key point comparison result for each key point to be recognized; and identifying whether a target object exists in the gate opening and closing area based on the key point comparison result for each key point to be recognized.

[0114] This specification implements one or more embodiments that define a gate opening / closing area within the gate's passage zone. Multiple key points are pre-set within this area, and preset key point information is acquired. During actual gate operation, a depth image is acquired, and key points corresponding to the preset key points are extracted. The depth information of each key point in the depth image is compared with the preset key point information to determine if a target object exists within the gate's opening / closing area. By defining the gate opening / closing area within the passage zone, the recognition area for target objects is reduced, improving accuracy while decreasing computational workload. Acquiring depth images and setting preset key points for object recognition avoids detection errors that occur when using target detection models, further improving object recognition accuracy. Furthermore, this improved accuracy enhances the safety of target objects passing through the gate's opening / closing area, preventing accidental injury to pedestrians or damage to other objects when the gate opens or closes.

[0115] The following is in conjunction with the appendix Figure 4 Taking the application of the object recognition method provided in this specification in a subway turnstile scenario as an example, the object recognition method will be further explained. Figure 4 This specification illustrates a flowchart of the processing procedure for an object recognition method according to one embodiment. Figure 4 As shown, the object recognition method provided in this specification can be specifically divided into a preprocessing stage and an actual processing stage.

[0116] First, the pre-processing stage is explained. The static color image of the subway gate is obtained by the color camera, the static depth image of the subway gate is obtained by the depth camera, a plurality of preset key points are determined in the static color image, and the color space position information of each preset key point in the color camera space coordinate system is calculated. According to the color space position information of each preset key point, the depth space position information of each preset key point in the depth camera space coordinate system is determined. Through the conversion relationship between the color camera space coordinate system and the depth camera space coordinate system, the pixel position information and the pixel depth information of each preset key point are determined, so as to realize the data alignment between the static color image and the static depth image. Further, according to the pixel depth information of each preset key point, the static mean value feature and the static difference value feature of each preset key point are calculated, and after the distance between the bottom of the subway gate and the floor of the gate opening and closing area is obtained, the depth range information of each preset key point is generated according to the static mean value feature and the static difference value feature, which can be specifically represented as .

[0117] It should be noted that the above pre-processing stage is realized when there is no passenger passing through the subway gate, and the pre-processing stage does not need to be executed again during the actual operation of the subway gate. The pixel position information and the depth range information of each preset key point can be directly obtained and used.

[0118] Next, the actual processing stage is explained. The to-be-identified depth image of the subway gate in the actual running process is obtained by the depth camera, and the to-be-identified key points corresponding to each preset key point and the depth information of each to-be-identified key point are extracted in the to-be-identified depth image according to the pixel position information of each preset key point. The depth information of each to-be-identified key point is compared with the depth range information of the corresponding preset key point to determine whether the depth information of the to-be-identified key point is within the depth range information of the corresponding preset key point. If the depth information of any to-be-identified key point is within the depth range information of the corresponding preset key point among the to-be-identified key points, it is determined that there is a passenger or other passing object in the gate opening and closing area of the subway gate, and the gate opening and closing area is in an unsafe state at this time. If the depth information of the to-be-identified key point is not within the depth range information of the corresponding preset key point among the to-be-identified key points, it is determined that there is no passenger or other passing object in the gate opening and closing area of the subway gate, and the gate opening and closing area is in a safe state at this time.

[0119] An embodiment of the present specification realizes that, when there is no passenger passing through the subway gate, a plurality of key points are set in the opening and closing area of the gate by acquiring a static color image and a static depth image of the subway gate, and preset key point information of the plurality of key points is generated, so that in the actual operation process of the gate, a to-be-identified depth image is acquired, to-be-identified key points corresponding to the preset key points are extracted from the to-be-identified depth image, and the depth information of each to-be-identified key point in the to-be-identified depth image is compared with the depth range information, so as to determine whether there is a passenger or other passing object in the opening and closing area of the gate. By determining the opening and closing area of the gate in the passing area of the gate, the identification area is reduced, the calculation workload is reduced, and the accuracy of identifying passengers or other passing objects is improved; by acquiring the to-be-identified depth image and setting the preset key points for object identification, the detection error caused by using the target detection model to detect the target in the image is avoided, and the accuracy of object identification is further improved; and then, on the basis of improving the accuracy of object identification, the safety of passengers or other passing objects passing through the opening and closing area of the gate can be further improved, and the passengers or other passing objects are prevented from being injured or damaged when the gate door is opened or closed.

[0120] Corresponding to the method embodiments described above, the present specification also provides object identification device embodiments, Figure 5 A structure schematic diagram of an object identification device provided by an embodiment of the present specification is shown. As shown in the figure, Figure 5 The device comprises:

[0121] The acquisition module 502 is configured to acquire a to-be-identified depth image and preset key point information;

[0122] The extraction module 504 is configured to extract the depth information of at least two to-be-identified key points of the gate opening and closing area in the to-be-identified depth image according to the preset key point information;

[0123] The comparison module 506 is configured to compare the depth information of each to-be-identified key point with the preset key point information respectively, and obtain the key point comparison result of each to-be-identified key point;

[0124] The identification module 508 is configured to identify whether there is a target object in the gate opening and closing area according to the key point comparison result of each to-be-identified key point.

[0125] Optionally, the device further comprises a generation module configured to:

[0126] acquire a static color image of a target gate and a static depth image corresponding to the static color image;

[0127] determine a gate opening and closing area of the target gate and at least two preset key points of the gate opening and closing area in the static color image;

[0128] determine color space position information of the at least two preset key points, and determine depth space position information of the at least two preset key points in the static depth image according to the color space position information;

[0129] determine pixel position information and pixel depth information of the at least two preset key points according to the depth space position information;

[0130] generate preset key point information of the at least two preset key points according to the pixel position information and the pixel depth information.

[0131] Optionally, the generating module is further configured to:

[0132] determine a target key point in the at least two preset key points, and determine target pixel depth information of the target key point in the pixel depth information;

[0133] calculate target static mean value features and target static difference value features of the target key point according to the target pixel depth information;

[0134] generate target depth range information of the target key point according to the target static mean value features and the target static difference value features;

[0135] determine target pixel position information of the target key point in the pixel position information, and determine the target pixel position information and the target depth range information as preset key point information of the target key point.

[0136] Optionally, the extracting module 504 is further configured to:

[0137] obtain target pixel position information of a target key point in the preset key point information, wherein the target key point is any one of the preset key points;

[0138] determine a to-be-processed key point in the to-be-identified depth image according to the target pixel position information, and extract to-be-processed depth information of the to-be-processed key point.

[0139] Optionally, the comparing module 506 is further configured to:

[0140] obtain target depth range information of the target key point in the preset key point information;

[0141] determine whether the to-be-processed depth information is located in the target depth range information, and obtain a to-be-processed key point comparison result of the to-be-processed key point.

[0142] Optionally, the identification module 508 is further configured to:

[0143] In a case where the to-be-processed depth information is located in the target depth range information, it is determined that the target object exists in the gate opening and closing area.

[0144] Optionally, the apparatus further includes a traversal module configured to:

[0145] In a case where the to-be-processed depth information is not located in the target depth range information, a new to-be-processed key point in each to-be-identified key point is determined as a to-be-traversed key point, and a to-be-traversed key point comparison result of the to-be-traversed key point is obtained;

[0146] In a case where the to-be-traversed depth information of the to-be-traversed key point is located in to-be-traversed depth range information, it is determined that the target object exists in the gate opening and closing area.

[0147] In a case where the to-be-traversed depth information is not located in the to-be-traversed depth range information, the step of determining a new to-be-processed key point in each to-be-identified key point as a to-be-traversed key point and obtaining a to-be-traversed key point comparison result of the to-be-traversed key point is returned to be executed until traversal of each to-be-identified key point is completed.

[0148] Optionally, the identification module 508 is further configured to:

[0149] In the key point comparison result of each to-be-identified key point, it is determined whether depth information of the to-be-identified key point is located in corresponding depth range information.

[0150] In a case where the depth information of the to-be-identified key point is located in the corresponding depth range information, it is determined that the target object exists in the gate opening and closing area.

[0151] The object identification apparatus provided in the specification includes an acquisition module configured to acquire a to-be-identified depth image and preset key point information; an extraction module configured to extract depth information of at least two to-be-identified key points of a gate opening and closing area in the to-be-identified depth image according to the preset key point information; a comparison module configured to compare the depth information of each to-be-identified key point with the preset key point information respectively, and obtain a key point comparison result of each to-be-identified key point; and an identification module configured to identify whether a target object exists in the gate opening and closing area according to the key point comparison result of each to-be-identified key point.

[0152] The one or more embodiments of the specification achieve that a gate opening and closing area is determined in a gate passing area, a plurality of key points are preset in the gate opening and closing area, preset key point information of the plurality of key points is acquired, in actual operation of the gate, a to-be-recognized depth image is collected, to-be-recognized key points corresponding to the preset key points are extracted from the to-be-recognized depth image, and then depth information of each to-be-recognized key point in the to-be-recognized depth image is compared with the preset key point information, so as to determine whether a target object exists in the gate opening and closing area. By determining the gate opening and closing area in the gate passing area, the recognition area for recognizing the target object is reduced, the accuracy of recognizing the target object is improved on the basis of reducing the calculation workload, the object recognition is performed by acquiring the to-be-recognized depth image and setting the preset key points, the detection error caused by using the target detection model to perform target detection on the image is avoided, and the accuracy of the object recognition is further improved. Then, on the basis of improving the accuracy of the object recognition, the safety of the target object in passing through the gate opening and closing area can be further improved, and the pedestrians or other objects are prevented from being injured or damaged when the gate door is opened or closed.

[0153] The above is a schematic scheme of the object recognition device according to the embodiment of the specification. It should be noted that the technical scheme of the object recognition device belongs to the same concept as the technical scheme of the object recognition method described above, and the details of the technical scheme of the object recognition device that are not described in detail can be referred to the description of the technical scheme of the object recognition method.

[0154] Figure 6 A structural block diagram of a computing device 600 according to an embodiment of the specification is shown. The components of the computing device 600 include but are not limited to a memory 610 and a processor 620. The processor 620 is connected to the memory 610 through a bus 630, and a database 650 is used to save data.

[0155] The computing device 600 also includes an access device 640 that enables the computing device 600 to communicate via one or more networks 660. Examples of such networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or combinations of such networks, such as the Internet. The access device 640 can include one or more of any type of network interface (for example, a network interface card (NIC)) such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a Worldwide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a Bluetooth interface, a Near Field Communication (NFC) interface, or the like.

[0156] In one embodiment of the present specification, the above-mentioned components of the computing device 600 and other components not shown in the Figure 6 may be connected to each other, for example, through a bus. It should be understood that Figure 6 The computing device structure diagram shown is only for the purpose of example, and is not a limitation on the scope of the present specification. Those skilled in the art can add or replace other components as needed.

[0157] The computing device 600 can be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (for example, a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, and the like), a mobile phone (for example, a smartphone), a wearable computing device (for example, a smart watch, smart glasses, and the like), or other types of mobile devices, or a stationary computing device such as a desktop computer or a personal computer (PC). The computing device 600 can also be a mobile or stationary server.

[0158] wherein the processor 620 implements the steps of the object recognition method when executing the computer program / instructions.

[0159] The above is a schematic scheme of the computing device of the embodiment. It should be noted that the technical scheme of the computing device and the technical scheme of the object recognition method described above belong to the same concept, and details of the technical scheme of the computing device that are not described in detail can be referred to the description of the technical scheme of the object recognition method.

[0160] An embodiment of the present specification also provides a computer readable storage medium storing computer programs / instructions, which, when executed by a processor, implement the steps of the object recognition method described above.

[0161] The above is a schematic scheme of the computer readable storage medium of the embodiment. It should be noted that the technical scheme of the storage medium and the technical scheme of the object recognition method described above belong to the same concept, and details of the technical scheme of the storage medium that are not described in detail can be referred to the description of the technical scheme of the object recognition method.

[0162] An embodiment of the present specification also provides a computer program product comprising computer programs / instructions, which, when executed by a processor, implement the steps of the object recognition method described above.

[0163] The above is a schematic scheme of the computer program product of the embodiment. It should be noted that the technical scheme of the computer program product and the technical scheme of the object recognition method described above belong to the same concept, and details of the technical scheme of the computer program product that are not described in detail can be referred to the description of the technical scheme of the object recognition method.

[0164] The specific embodiments of the present specification are described above. Other embodiments are within the scope of the appended claims. In some cases, acts or steps recited in the claims can be performed in an order other than that in which they are recited in the embodiments, and still achieve desirable results. Also, the process depicted in the figures can not necessarily require the particular order shown, or sequential order to achieve the results desired. In certain implementations, multitasking and parallel processing can be advantageous.

[0165] The computer programs / instructions include computer program code, which can be in the form of source code, object code, executable code, or some intermediate form. The computer readable medium can include any entity or apparatus capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0166] It should be noted that, for the aforementioned method embodiments, the sequences of the actions are described for ease of description. However, it is to be understood that the sequences of the actions can be changed according to the embodiments of the present application, and more actions can be added, or existing actions can be removed, depending on the actual conditions. Moreover, the embodiments described in the specification are preferred embodiments only and do not limit the scope of the application.

[0167] In the above embodiments, the description of each embodiment is focused on the description of the embodiment. The parts not described in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0168] The preferred embodiments of the present application disclosed above are only used to clarify the present application. The alternative embodiments do not describe all the details and limit the application to the specific embodiments described. Obviously, according to the content of the present application, many modifications and changes can be made. The present application selects and describes these embodiments in order to better explain the principles and practical applications of the present application, so that those skilled in the art can well understand and use the present application. The present application is limited by the claims and their full scope and equivalents.

Claims

1. A method of object recognition, characterized by, The method comprises the following steps: obtaining a to-be-identified depth image and preset key point information, wherein the preset key point information comprises pixel position information and depth range information of the preset key points, the depth range information is generated according to static mean value characteristics and static difference value characteristics of the preset key points, the static mean value characteristics and the static difference value characteristics are determined according to pixel depth information of the preset key points, the preset key points are preset key points in a gate opening and closing area of a target gate, the pixel position information and the pixel depth information are determined according to the preset key points in a corresponding static depth image of the target gate, and the static difference value characteristics are standard deviations of the pixel depth information of the preset key points in at least two static depth images; extracting depth information of at least two to-be-identified key points in the gate opening and closing area from the to-be-identified depth image according to the preset key point information; comparing the depth information of each to-be-identified key point with the preset key point information respectively to obtain key point comparison results of each to-be-identified key point; identifying whether a target object exists in the gate opening and closing area according to the key point comparison results of each to-be-identified key point.

2. The method of claim 1, wherein, Before obtaining the to-be-identified depth image and the preset key point information, the method further comprises the following steps: obtaining a static color image of a target gate and a static depth image corresponding to the static color image; determining a gate opening and closing area of the target gate and at least two preset key points in the gate opening and closing area in the static color image; determining color space position information of the at least two preset key points and depth space position information of the at least two preset key points in the static depth image according to the color space position information; determining pixel position information and pixel depth information of the at least two preset key points according to the depth space position information; generating preset key point information of the at least two preset key points according to the pixel position information and the pixel depth information.

3. The method of claim 1, wherein, Extracting depth information of at least two to-be-identified key points in the gate opening and closing area from the to-be-identified depth image according to the preset key point information comprises the following steps: obtaining target pixel position information of a target key point in the preset key point information, wherein the target key point is any one of the preset key points; determining a to-be-processed key point in the to-be-identified depth image according to the target pixel position information and extracting to-be-processed depth information of the to-be-processed key point.

4. The method of claim 3, wherein, Comparing the depth information of each to-be-identified key point with the preset key point information respectively to obtain key point comparison results of each to-be-identified key point comprises the following steps: obtaining target depth range information of the target key point in the preset key point information; judging whether the to-be-processed depth information is located in the target depth range information to obtain a to-be-processed key point comparison result of the to-be-processed key point.

5. The method of claim 4, wherein, Identifying whether a target object exists in the gate opening and closing area according to the key point comparison results of each to-be-identified key point comprises the following steps: in the case that the to-be-processed depth information is located in the target depth range information, it is determined that a target object exists in the gate opening and closing area.

6. The method of claim 5, wherein, The method further comprises: In a case where the to-be-processed depth information is not located in the target depth range information, determining a new to-be-processed key point in each to-be-identified key point as a to-be-traversed key point, and obtaining a to-be-traversed key point comparison result of the to-be-traversed key point; In a case where the to-be-traversed depth information of the to-be-traversed key point is located in to-be-traversed depth range information, it is determined that there is a target object in the gate opening and closing area; In a case where the to-be-traversed depth information is not located in the to-be-traversed depth range information, returning to execute the step of determining a new to-be-processed key point in each to-be-identified key point as a to-be-traversed key point, and obtaining a to-be-traversed key point comparison result of the to-be-traversed key point, until each to-be-identified key point is traversed.

7. The method of claim 4, wherein, According to the key point comparison result of each to-be-identified key point, whether there is a target object in the gate opening and closing area is identified, comprising: In the key point comparison result of each to-be-identified key point, it is judged whether there is to-be-identified key point depth information located in the corresponding depth range information; In a case where there is to-be-identified key point depth information located in the corresponding depth range information, it is determined that there is a target object in the gate opening and closing area.

8. An object recognition apparatus characterized by comprising: Comprise: The acquisition module is configured to acquire a to-be-identified depth image and preset key point information, wherein the preset key point information comprises pixel position information and depth range information corresponding to a preset key point, the depth range information is generated according to static mean value characteristics and static difference value characteristics of the preset key point, the static mean value characteristics and the static difference value characteristics are determined according to pixel depth information of the preset key point, the pixel position information and the pixel depth information are determined according to the preset key point in the target gate corresponding static depth image, and the static difference value characteristics are standard deviations of the pixel depth information of the preset key point in at least two static depth images; The extraction module is configured to extract depth information of at least two to-be-identified key points of the gate opening and closing area in the to-be-identified depth image according to the preset key point information; The comparison module is configured to compare the depth information of each to-be-identified key point with the preset key point information respectively, and obtain a key point comparison result of each to-be-identified key point; The identification module is configured to identify whether there is a target object in the gate opening and closing area according to the key point comparison result of each to-be-identified key point.

9. A computing device, comprising: a memory and a processor; The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions, characterized in that the computer programs / instructions are executed by the processor to realize the steps of the method of any one of claims 1-7.

10. A computer readable storage medium storing computer programs / instructions, characterized in that, The computer programs / instructions are executed by the processor to realize the steps of the method of any one of claims 1-7.

11. A computer program product comprising computer programs / instructions, characterized in that, The computer programs / instructions are executed by the processor to realize the steps of the method of any one of claims 1-7.

Images