Hand region recognition method, device, equipment and storage medium

Abstract

The application discloses a hand region recognition method, device and equipment and a storage medium. The method comprises the following steps: acquiring a to-be-recognized hand image collected by an image collection device in a cockpit; the to-be-recognized hand image comprises a hand region of a to-be-recognized human body; performing hand region segmentation processing on the to-be-recognized hand image to obtain a hand region segmentation image; performing hand region key point extraction on the to-be-recognized hand image to obtain a hand region key point; determining a sliding region radius; performing image filtering on the hand region segmentation image according to the hand region key point and the sliding region radius to obtain a target hand region image. The technical scheme of the embodiment of the application improves the hand recognition accuracy.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method, apparatus, device, and storage medium for hand area recognition. Background Technology

[0002] With the rapid development of the smart cockpit ecosystem, gesture recognition, as a crucial function for in-cabin passenger interaction, is increasingly attracting the attention of consumers and OEMs. Hand segmentation is a key step in achieving gesture recognition, and accurate segmentation results can provide excellent assistance for gesture recognition.

[0003] However, the precision of hand segmentation in the complex environment of an in-vehicle vehicle is often constrained by various factors, leading to inconsistent recognition results. This is especially true when a person is holding other objects such as a steering wheel, water bottle, or mobile phone, where parts of these objects are frequently missegmented as hands, significantly impacting the consumer experience. Therefore, achieving accurate hand recognition has become a key challenge. Summary of the Invention

[0004] This invention provides a method, apparatus, device, and storage medium for hand area recognition to improve the accuracy of hand recognition.

[0005] According to one aspect of the present invention, a method for recognizing a hand region is provided, the method comprising:

[0006] Acquire an image of the hand to be identified from an image acquisition device inside the cockpit; the image of the hand to be identified includes the hand region of the human body to be identified;

[0007] The hand region of the image to be identified is segmented to obtain a segmented hand region image; and,

[0008] The key points of the hand region are extracted from the hand image to be identified.

[0009] Determine the radius of the sliding area;

[0010] Based on the key points of the hand region and the radius of the sliding region, the segmented image of the hand region is filtered to obtain the target hand region image.

[0011] According to another aspect of the present invention, a hand region recognition device is provided, the device comprising:

[0012] A hand image acquisition module is used to acquire a hand image to be identified obtained by an image acquisition device in the cockpit; the hand image to be identified includes the hand region of the human body to be identified;

[0013] A hand region processing module is used to perform hand region segmentation processing on the hand image to be identified, to obtain a hand region segmentation image; and,

[0014] The key point determination module is used to extract key points of the hand region from the hand image to be identified, and obtain the key points of the hand region.

[0015] The sliding radius determination module is used to determine the radius of the sliding area;

[0016] The image filtering module is used to perform image filtering on the segmented image of the hand region based on the key points of the hand region and the radius of the sliding region to obtain the target hand region image.

[0017] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0018] At least one processor; and

[0019] A memory communicatively connected to the at least one processor; wherein,

[0020] The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the hand region recognition method according to any embodiment of the present invention.

[0021] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the hand region recognition method according to any embodiment of the present invention.

[0022] This embodiment employs a pre-trained keypoint recognition model to extract keypoints from the hand region of the hand image to be identified, thereby improving the accuracy of keypoint recognition. By connecting the keypoints based on their location information, at least one keyline is obtained; each keyline includes at least one keypoint from the hand region. A filtering region corresponding to each keypoint is generated based on the radius of the sliding region. The filtered region of any keypoint on each keyline is used to filter the segmented hand region image, resulting in the target hand region image. By using a defined filtering region to filter the image pixels on the connecting lines, the accuracy of determining the target hand region image is further improved.

[0023] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1A This is a flowchart of a hand region recognition method provided in Embodiment 1 of the present invention;

[0026] Figure 1A This is a schematic diagram of key hand point markings provided in Embodiment 1 of the present invention;

[0027] Figure 2 This is a flowchart of a hand region recognition method according to Embodiment 2 of the present invention;

[0028] Figure 3 This is a schematic diagram of the structure of a hand area recognition device according to Embodiment 3 of the present invention;

[0029] Figure 4 This is a schematic diagram of the structure of an electronic device that implements the hand region recognition method of the present invention. Detailed Implementation

[0030] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0031] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0032] Example 1

[0033] Figure 1A This is a flowchart of a hand region recognition method provided in Embodiment 1 of the present invention. This embodiment is applicable to situations where drivers need to recognize their hand regions during driving. The method can be executed by a hand region recognition device, which can be implemented in hardware and / or software and can be configured in an electronic device. Figure 1A As shown, the method includes:

[0034] S110. Acquire the image of the hand to be identified obtained by the image acquisition device in the cockpit; the image of the hand to be identified includes the hand area of ​​the human body to be identified.

[0035] The image acquisition device can be a DMS (Driver Monitoring System) or OMS (Object Monitoring System) camera deployed in the vehicle cabin.

[0036] For example, the image acquisition device can acquire images of the upper body area of ​​the driver in the cockpit to obtain an image of the hand to be identified. It should be noted that the image acquisition device can acquire images of the person to be photographed based on a preset acquisition frequency until an image containing the hand area is acquired, and then that image is used as the hand image to be identified. The image to be identified includes the hand area of ​​the person to be identified.

[0037] S120. Perform hand region segmentation processing on the hand image to be recognized to obtain a hand region segmentation image.

[0038] For example, a pre-trained object detection model can be used to segment the hand region of the hand image to be identified, resulting in a segmented hand region image. The object detection model can be YOLOv8 (the eighth version of the YOLO model), or a detection model such as FasterNet (Fast Efficient Neural Network) or CenterNet (center point detection network). Specifically, the hand image to be identified is input into the pre-defined object detection model, and the model outputs a segmented hand region image.

[0039] The object detection model can be pre-trained. For example, a large number of hand region images can be pre-acquired as a sample training set, and the hand regions can be labeled to train the object detection model until the model converges, thus obtaining a fully trained object detection model.

[0040] S130. Extract key points of the hand region from the hand image to be recognized to obtain key points of the hand region.

[0041] Optionally, before extracting key points, hand region segmentation can be performed on the hand image to be recognized. Specifically, after obtaining the hand image to be recognized, semantic segmentation models such as HRNet (High-Resolution Network) can be used to segment the hand region. The segmentation method can employ either traditional image processing methods or deep learning-based model inference.

[0042] One method for extracting key points from the hand region of an image to be recognized is based on detection using deep learning networks. Different key points are used to represent different parts of the hand. For example, ... Figure 1B The diagram illustrates the annotation of key points on a hand. It includes 21 key points, each with a different meaning depending on its number. For example, 0 represents the palm, 9 represents the first joint of the middle finger, and 12 represents the fingertip of the middle finger. The meanings of the other key point numbers will not be elaborated upon in this embodiment.

[0043] S140. Determine the radius of the sliding area.

[0044] For example, the contour of the hand region to be identified can be extracted to obtain the hand region contour; the radius of the sliding region can be determined based on the size of the hand region contour. Specifically, the radius of the sliding region can be determined based on the area of ​​the hand region contour. For example, the area of ​​the sliding region can be 1 / 15 of the area of ​​the hand region contour. The radius of the sliding region can be determined based on the area of ​​the sliding region.

[0045] To further improve the accuracy of determining the radius of the sliding region, thereby achieving accurate image filtering of hand region segmentation images, the radius of the sliding region can also be determined by combining the relationship between human body and hand features.

[0046] In one optional embodiment, the hand image to be identified further includes a human body region of the human body to be identified; determining the sliding region radius includes: performing region detection on the human body region of the hand image to be identified to determine the human body height and width; and determining the sliding region radius based on the human body height and width.

[0047] For example, region detection can be performed on the human body region within the hand image to be identified, specifically using an existing object detection model. Since the human body image captured in a driving scenario is a half-body image, the size of the full-body image region can be determined based on the size of the half-body image region, and then the height and width of the human body can be determined based on the size of the full-body image region. It should be noted that the height and width of the human body determined here are relative to the image.

[0048] The method for determining the radius R of the sliding area based on the human body height H and width W is as follows:

[0049] R = k * H * W

[0050] Where k is a constant term preset by relevant technical personnel based on actual experience or experimental values; for example, k is 0.0023.

[0051] S150. Based on the key points of the hand region and the radius of the sliding region, perform image filtering on the segmented image of the hand region to obtain the target hand region image.

[0052] For example, using key points of each hand region as centers, a sliding filter region is determined based on the radius of the sliding region. The hand region segmentation image is then filtered according to the sliding filter region to obtain the target hand region image. For instance, regions outside the sliding filter region can be filtered to obtain the target hand region image. Alternatively, a pre-defined sliding path or range can be set, controlling the sliding filter region to slide within the set path or range, filtering out regions not traversed by the sliding filter region to obtain the target hand region image.

[0053] The technical solution of this invention acquires an image of a hand to be identified by an image acquisition device in the cockpit; the image includes the hand region of the person to be identified; the hand region is segmented from the image to obtain a segmented hand region image; key points are extracted from the hand region to obtain key points; the radius of the sliding region is determined; and image filtering is performed on the segmented hand region image based on the key points and the radius of the sliding region to obtain the target hand region image. This technical solution achieves accurate identification of the driver's hand region in a cockpit setting by combining hand region extraction and key point localization; further improving the accuracy of hand region recognition by combining the sliding region radius with secondary filtering of the segmented hand region image.

[0054] Example 2

[0055] Figure 2This is a flowchart of a hand region recognition method provided in Embodiment 2 of the present invention. This embodiment is an optimization and improvement based on the above technical solutions.

[0056] Furthermore, the step "extracting key points of the hand region from the hand image to be identified, and obtaining key points of the hand region" is refined to "using a pre-trained key point recognition model to extract key points of the hand region from the hand image to be identified, and obtaining key points of the hand region." This improves the method for extracting key points of the hand region.

[0057] Furthermore, the step "based on the key points of the hand region and the radius of the sliding region, perform image filtering on the segmented image of the hand region to obtain the target hand region image" is refined to "based on the point identification information of the key points of the hand region, connect the key points of the hand region to obtain at least one key line; each key line includes at least one key point of the hand region; based on the radius of the sliding region, generate the filtering region corresponding to the key point of the hand region; using the filtering region of any key point of the hand region on each key line, perform image filtering on the segmented image of the hand region to obtain the target hand region image." This improves the image filtering method for the segmented image of the hand region.

[0058] It should be noted that for parts not described in detail in the embodiments of the present invention, please refer to the descriptions in other embodiments. For example... Figure 2 As shown, the method includes the following specific steps:

[0059] S210. Acquire an image of the hand to be identified from an image acquisition device inside the cockpit; the image of the hand to be identified includes the hand region of the human body to be identified.

[0060] S220. Perform hand region segmentation processing on the hand image to be recognized to obtain a hand region segmentation image.

[0061] S230. Using a pre-trained key point recognition model, extract key points of the hand region from the hand image to be recognized, and obtain the key points of the hand region.

[0062] For example, the hand image to be recognized can be input into a pre-trained keypoint recognition model to obtain the keypoints of the hand region output by the model, specifically the keypoint coordinates and keypoint identifiers. It should be noted that keypoints with different identifiers have different meanings. For example, 0 can represent the palm, 9 can represent the first joint of the middle finger, and 12 can represent the fingertip of the middle finger. The meanings of other keypoint numbers will not be elaborated in this embodiment.

[0063] In one optional embodiment, the keypoint recognition model is trained as follows: historical hand region images over a historical time period are acquired; the historical hand region images are labeled with standard keypoint coordinates; the historical hand region images labeled with standard keypoint coordinates are input into a pre-built detection network model to obtain the predicted keypoint coordinates output by the model; the detection network model is trained based on the standard keypoint coordinates and the predicted keypoint coordinates until the model training termination condition is met, thus obtaining the keypoint recognition model.

[0064] The model training termination condition can be reaching a preset iteration threshold, or the loss value stabilizing. The annotation method for key points in the hand region of historical hand region images can be manual or semi-automated.

[0065] For example, the detection network model can be trained based on the predicted keypoint coordinates and standard keypoint coordinates output by the model, using a preset loss function, until the loss value no longer changes, thus obtaining a keypoint recognition model.

[0066] S240, Determine the radius of the sliding area.

[0067] S250. Based on the location identification information of key points in the hand area, connect the key points in the hand area to obtain at least one key line; each key line includes at least one key point in the hand area.

[0068] Among them, the point identification information can be the key point identifiers output by the model, that is, the key point number, and different numbers represent different hand areas.

[0069] For example, based on the location markers of key points in the hand area, key points are connected to obtain at least one key line. The connection method can be based on preset location markers. For example, ... Figure 1B The key points of the hand area shown can be obtained by connecting the point markers 0, 1, 2, 3, and 4; or by connecting the point markers 0, 5, 6, 7, and 8. The connection rules can be pre-set according to the different hand areas represented by the point markers.

[0070] S260. Generate the filter area corresponding to the key points of the hand area based on the radius of the sliding area.

[0071] For example, a filtering region corresponding to the key points of the hand region can be generated based on the radius of the sliding region, with the key points of the hand region as the center.

[0072] S270. Using the filtering region of any key point in the hand region on each key line, perform image filtering on the hand region segmentation image to obtain the target hand region image.

[0073] For example, the filtering area of ​​any key point of the hand region on each key line can be used to slide and erase image pixels outside the area on the key line, thereby filtering the image and obtaining the target hand region image.

[0074] It should be noted that due to the special nature of the key points on the palm in the hand region, which occupy a large area, a separate sliding area can be generated for each key point on the palm to achieve pixel filtering of the area within the palm region, thereby further improving the image filtering accuracy of the hand region image.

[0075] In an optional embodiment, if the point identification information of the key point in the hand region is the palm region, then at least one target region key point connected to the key point in the hand region of the palm region is determined; the point distance between the key point in the hand region of the palm region and each target region key point is determined; based on the point distance, the palm sliding area radius of the key point in the hand region of the palm region is determined; based on the palm sliding area radius, the target hand region image is image filtered to obtain the filtered target hand region image.

[0076] Among them, the key points of the hand area in the palm region can be connected with the key points of the five fingers. Therefore, each key line contains key points of the palm region.

[0077] For example, the positional distances between key points in the hand region of the palm area and key points in each target region are determined. The target region key points are those connected to the key points in the hand region of the palm area. Based on these positional distances, the radius of the palm sliding region for each key point in the hand region of the palm area is determined. For example, the radius of the palm sliding region can be the one with the largest positional distance. Based on the radius of the palm sliding region, the target hand region image is filtered to obtain the filtered target hand region image. The target hand region image can be an image within the area where the palm is located.

[0078] This embodiment employs a pre-trained keypoint recognition model to extract keypoints from the hand region of the hand image to be identified, thereby improving the accuracy of keypoint recognition. By connecting the keypoints based on their location information, at least one keyline is obtained; each keyline includes at least one keypoint from the hand region. A filtering region corresponding to each keypoint is generated based on the radius of the sliding region. The filtered region of any keypoint on each keyline is used to filter the segmented hand region image, resulting in the target hand region image. By using a defined filtering region to filter the image pixels on the connecting lines, the accuracy of determining the target hand region image is further improved.

[0079] Example 3

[0080] Figure 3 This is a schematic diagram of a hand area recognition device provided in Embodiment 3 of the present invention. The hand area recognition device provided in this embodiment of the present invention is applicable to situations where drivers need to recognize their hand areas during driving. This hand area recognition device can be implemented in hardware and / or software, such as... Figure 3 As shown, the device specifically includes: a hand image acquisition module 301, a hand region processing module 302, a key point determination module 303, a sliding radius determination module 304, and an image filtering module 305. Among them,

[0081] The hand image acquisition module 301 is used to acquire the hand image to be identified obtained by the image acquisition device in the cockpit; the hand image to be identified includes the hand area of ​​the human body to be identified;

[0082] The hand region processing module 302 is used to perform hand region segmentation processing on the hand image to be identified, to obtain a hand region segmentation image; and,

[0083] The key point determination module 303 is used to extract key points of the hand region from the hand image to be identified, and obtain key points of the hand region.

[0084] The sliding radius determination module 304 is used to determine the radius of the sliding area;

[0085] The image filtering module 305 is used to perform image filtering on the segmented image of the hand region based on the key points of the hand region and the radius of the sliding region to obtain the target hand region image.

[0086] This embodiment employs a pre-trained keypoint recognition model to extract keypoints from the hand region of the hand image to be identified, thereby improving the accuracy of keypoint recognition. By connecting the keypoints based on their location information, at least one keyline is obtained; each keyline includes at least one keypoint from the hand region. A filtering region corresponding to each keypoint is generated based on the radius of the sliding region. The filtered region of any keypoint on each keyline is used to filter the segmented hand region image, resulting in the target hand region image. By using a defined filtering region to filter the image pixels on the connecting lines, the accuracy of determining the target hand region image is further improved.

[0087] Optionally, the hand image to be identified further includes the human body region to be identified; the sliding radius determination module 304 includes:

[0088] The human body height and width determination unit is used to perform region detection on the human body region of the hand image to be identified, and determine the human body height and width.

[0089] The sliding radius determining unit is used to determine the radius of the sliding area based on the height and width of the human body.

[0090] Optionally, the hand area processing module 302 includes:

[0091] The hand region processing unit is used to extract key points of the hand region from the hand image to be identified using a pre-trained key point recognition model.

[0092] Optionally, the key point recognition model can be trained in the following way:

[0093] Acquire historical hand region images over a historical time period; the historical hand region images are labeled with standard key point coordinates;

[0094] Historical hand region images labeled with standard keypoint coordinates are input into a pre-built detection network model to obtain the predicted keypoint coordinates output by the model.

[0095] The detection network model is trained based on the standard keypoint coordinates and the predicted keypoint coordinates until the model training termination condition is met, thus obtaining the keypoint recognition model.

[0096] Optionally, the image filtering module 305 includes:

[0097] The keyline determination unit is used to connect the key points of the hand region according to the point identification information of the key points of the hand region to obtain at least one keyline; each keyline includes at least one key point of the hand region.

[0098] A filter area generation unit is used to generate a filter area corresponding to the key points of the hand area based on the radius of the sliding area.

[0099] An image filtering unit is used to perform image filtering on the hand region segmentation image by using the filtering area of ​​any hand region key point on each of the key lines, so as to obtain a target hand region image.

[0100] Optionally, the image filtering module 305 further includes:

[0101] The target region determination unit is used to determine at least one target region key point connected to the hand region key point of the palm region if the point identification information of the key point of the hand region is the palm region.

[0102] The point distance determination unit is used to determine the point distance between key points in the hand area of ​​the palm region and key points in each of the target areas;

[0103] The palm sliding area radius determination unit is used to determine the palm sliding area radius of the key points of the palm area based on the distance between each of the points.

[0104] The palm region filtering unit is used to filter the target hand region image based on the radius of the palm sliding area to obtain a filtered target hand region image.

[0105] The hand region recognition device provided in this embodiment of the invention can execute the hand region recognition method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0106] Example 4

[0107] Figure 4 A schematic diagram of an electronic device 40 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0108] like Figure 4 As shown, the electronic device 40 includes at least one processor 41 and a memory, such as a read-only memory (ROM) 42 or a random access memory (RAM) 43, communicatively connected to the at least one processor 41. The memory stores computer programs executable by the at least one processor. The processor 41 can perform various appropriate actions and processes based on the computer program stored in the ROM 42 or loaded into the RAM 43 from storage unit 48. The RAM 43 may also store various programs and data required for the operation of the electronic device 40. The processor 41, ROM 42, and RAM 43 are interconnected via a bus 44. An input / output (I / O) interface 45 is also connected to the bus 44.

[0109] Multiple components in electronic device 40 are connected to I / O interface 45, including: input unit 46, such as keyboard, mouse, etc.; output unit 47, such as various types of monitors, speakers, etc.; storage unit 48, such as disk, optical disk, etc.; and communication unit 49, such as network card, modem, wireless transceiver, etc. Communication unit 49 allows electronic device 40 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0110] Processor 41 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 41 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 41 performs the various methods and processes described above, such as hand region recognition methods.

[0111] In some embodiments, the hand region recognition method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 48. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 40 via ROM 42 and / or communication unit 49. When the computer program is loaded into RAM 43 and executed by processor 41, one or more steps of the hand region recognition method described above may be performed. Alternatively, in other embodiments, processor 41 may be configured to perform the hand region recognition method by any other suitable means (e.g., by means of firmware).

[0112] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0113] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0114] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0115] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0116] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0117] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0118] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0119] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for recognizing a hand region, characterized in that, include: Acquire the image of the hand to be identified from the image acquisition device inside the cockpit; The image of the hand to be identified includes the hand region of the human body to be identified; The hand region of the image to be identified is segmented to obtain a segmented hand region image; and, The key points of the hand region are extracted from the hand image to be identified. Determine the radius of the sliding area; Based on the key points of the hand region and the radius of the sliding region, the segmented image of the hand region is filtered to obtain the target hand region image; The hand image to be identified also includes the human body region of the human body to be identified. Determining the radius of the sliding region includes: Region detection is performed on the human body region of the hand image to be identified to determine the human body height and width; The radius of the sliding area is determined based on the height and width of the human body.

2. The method according to claim 1, characterized in that, The step of extracting key points of the hand region from the hand image to be identified, to obtain key points of the hand region, includes: A pre-trained key point recognition model is used to extract key points in the hand region of the hand image to be recognized, thereby obtaining the key points of the hand region.

3. The method according to claim 2, characterized in that, The training method for the key point recognition model is as follows: Acquire historical hand region images over a historical time period; the historical hand region images are labeled with standard key point coordinates; Historical hand region images labeled with standard keypoint coordinates are input into a pre-built detection network model to obtain the predicted keypoint coordinates output by the model. The detection network model is trained based on the standard keypoint coordinates and the predicted keypoint coordinates until the model training termination condition is met, thus obtaining the keypoint recognition model.

4. The method according to claim 1, characterized in that, The step of performing image filtering on the segmented image of the hand region based on the key points of the hand region and the radius of the sliding region to obtain the target hand region image includes: Based on the location identification information of the key points in the hand area, the key points in the hand area are connected to form a key line, and each key line includes at least one key point in the hand area. Based on the radius of the sliding area, a filtering area corresponding to the key points of the hand area is generated; By using the filtering region of any key point in the hand region on each of the key lines, the segmented image of the hand region is filtered to obtain the target hand region image.

5. The method according to claim 4, characterized in that, The method further includes: If the point identification information of the key point in the hand region is the palm region, then at least one target region key point connected to the key point in the hand region of the palm region is determined. Determine the positional distances between key points in the hand region of the palm area and key points in each of the target regions; Based on the distances between the points, determine the radius of the palm sliding area of ​​the key points in the palm region; Based on the radius of the palm sliding area, the target hand region image is filtered to obtain the filtered target hand region image.

6. A hand area recognition device, characterized in that, include: The hand image acquisition module is used to acquire the hand image to be identified obtained by the image acquisition device in the cockpit; The image of the hand to be identified includes the hand region of the human body to be identified; A hand region processing module is used to perform hand region segmentation processing on the hand image to be identified, to obtain a hand region segmentation image; and, The key point determination module is used to extract key points of the hand region from the hand image to be identified, and obtain the key points of the hand region. The sliding radius determination module is used to determine the radius of the sliding area; An image filtering module is used to perform image filtering on the segmented image of the hand region based on the key points of the hand region and the radius of the sliding region to obtain an image of the target hand region. The hand image to be identified also includes the human body region to be identified; the sliding radius determination module includes: The human body height and width determination unit is used to perform region detection on the human body region of the hand image to be identified, and determine the human body height and width. The sliding radius determining unit is used to determine the radius of the sliding area based on the height and width of the human body.

7. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the hand region recognition method according to any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the hand region recognition method according to any one of claims 1-5.

Images