Target Hand Analysis Method, Device, Electronic Device and Storage Medium

Abstract

The present application discloses a target hand analysis method, apparatus, electronic device, and storage medium. The method includes: obtaining a to-be-recognized image within a to-be-detected area; determining a first suspected target according to the to-be-recognized image and a preset hand contour; if the first suspected target belongs to a human body area, determining the first suspected target as a second suspected target; and if the gesture of the second suspected target conforms to a preset gesture, determining the second suspected target as a target hand. By judging whether the target conforms to the preset gesture, the embodiments of the present application can determine the target hand, thereby avoiding misrecognition caused by some objects being similar to the hand contour.

Description

Technical Field

[0001] This application relates to the technical field of gesture recognition, and in particular to a method, device, electronic device and storage medium for target hand analysis. Background Art

[0002] Gesture recognition technology has currently been widely applied to various household appliances, such as smart TVs, electronic picture frames, etc. A relatively common and reliable technology is to train a detector that can traverse and locate the hand area and label it on a frame of image by using the method of adaboost detector or deep learning training detector. The deep learning network can recognize most gestures, but it will make mistakes on some objects with similar contours. Summary of the Invention

[0003] Embodiments of this application provide a method, device, electronic device and storage medium for target hand analysis, which determine the target hand by judging whether the target conforms to a preset gesture, so as to avoid misrecognition caused by some objects having similar hand contours.

[0004] In a first aspect, an embodiment of this application provides a method for target hand analysis, including:

[0005] Obtain a to-be-recognized image in a to-be-detected area;

[0006] Determine a first suspected target according to the to-be-recognized image and a preset hand contour;

[0007] If the first suspected target belongs to a human body area, determine the first suspected target as a second suspected target;

[0008] If the gesture of the second suspected target conforms to a preset gesture, determine the second suspected target as the target hand.

[0009] In this embodiment, the target hand is determined by judging whether the target conforms to a preset gesture, so as to avoid misrecognition caused by some objects having similar hand contours. In addition, before gesture recognition, preliminary screening is performed through the determination of the preset hand contour and whether it belongs to the human body area, so as to narrow the recognition range and reduce the objects for gesture recognition in the later stage.

[0010] In some embodiments, the step of if the first suspected target belongs to a human body area, determine the first suspected target as a second suspected target includes:

[0011] Identify the human body image in the to-be-recognized image;

[0012] If the first suspected target belongs to the human body image, determine the first suspected target as a second suspected target.

[0013] In this embodiment, since the human body contour is relatively not easily confused with other objects, it is determined whether the first suspected target belongs to a human body image to exclude objects in the non-human body area of the first suspected target, so as to reduce the objects for which gesture recognition needs to be performed.

[0014] In some embodiments, the determining that the first suspected target is a second suspected target if the first suspected target belongs to a human body area includes:

[0015] Identifying the temperature value of the first suspected target;

[0016] If the temperature value belongs to a preset human body temperature range, it is determined that the first suspected target is a second suspected target.

[0017] In this embodiment, determining whether the first suspected target belongs to a human body area can further reduce the objects for which gesture recognition needs to be performed. Since the human body temperature has a specific range, combining the preset human body temperature range can further narrow down the range of the first suspected target.

[0018] In some embodiments, the determining that the second suspected target is a target hand if the gesture of the second suspected target conforms to a preset gesture includes:

[0019] Identifying the gesture information of the second suspected target;

[0020] If the gesture information conforms to a preset gesture, it is determined that the second suspected target is a target hand.

[0021] In this embodiment, a preset gesture is set to identify the target hand, which improves the recognition accuracy. In addition, multiple preset gestures can be set to avoid the low recognition rate of a single preset gesture due to external factors affecting the user experience.

[0022] In some embodiments, the obtaining of the image to be recognized in the area to be detected includes:

[0023] Prompting the gesture to be completed;

[0024] Obtaining the image to be recognized in the area to be detected based on the gesture to be completed;

[0025] The determining that the second suspected target is a target hand if the gesture information conforms to a preset gesture includes:

[0026] If the gesture information conforms to the gesture to be completed, it is determined that the second suspected target is a target hand.

[0027] In this embodiment, since the types of preset gestures can be freely set and some users may not be aware of the types of preset gestures, when obtaining the image to be recognized in the area to be detected, a prompt for the preset gesture, i.e., the gesture to be completed, is given to guide the user to make the preset gesture, thereby improving the accuracy of target hand recognition.

[0028] In some embodiments, the determining the first suspected target according to the image to be recognized and the preset hand contour includes:

[0029] Identifying the area in the image to be recognized that matches the preset hand contour through a deep learning network and determining it as the first suspected target.

[0030] In this embodiment, the contours of each object in the image to be recognized are processed by a deep learning network, and the deep learning network takes into account the relevance of the same object in multiple frames of images, so as to be able to recognize more accurately.

[0031] In some embodiments, after determining that the second suspected target is the target hand if the gesture of the second suspected target conforms to the preset gesture, it includes:

[0032] Tracking and recognizing the target gesture of the target hand;

[0033] Determining and executing the corresponding target instruction according to the correspondence between the gesture and the instruction and the target gesture.

[0034] In this embodiment, after determining the target hand, through the set correspondence between the gesture and the instruction, the target instruction can be quickly recognized to control the device to perform the corresponding operation, without actual contact and with a lower requirement for the operation distance, providing a better user experience.

[0035] In a second aspect, the present application provides a target hand analysis device, including:

[0036] An information acquisition module, configured to acquire an image to be recognized in an area to be detected;

[0037] A target analysis module, communicatively connected to the information acquisition module, configured to determine a first suspected target according to the image to be recognized and a preset hand contour; if the first suspected target belongs to the human body area, determine the first suspected target as a second suspected target;

[0038] A hand analysis module, communicatively connected to the target analysis module, configured to determine that the second suspected target is the target hand if the gesture of the second suspected target conforms to the preset gesture.

[0039] In this embodiment, by determining whether the target conforms to a preset gesture, the target hand is determined, avoiding misrecognition caused by some objects being similar to the hand contour. In addition, before gesture recognition, preliminary screening is performed through the determination of the preset hand contour and whether it belongs to the human body area, narrowing the recognition range to reduce the objects for gesture recognition in the later stage.

[0040] In some embodiments, the target analysis module is further configured to identify a human body image in the image to be recognized; if the first suspected target belongs to the human body image, the first suspected target is determined as the second suspected target.

[0041] In this embodiment, since the human body contour is relatively not easily confused with other objects, it is determined whether the first suspected target belongs to the human body image to exclude the objects in the non-human body area of the first suspected target, so as to reduce the objects that need to be gesture-recognized.

[0042] In some embodiments, the target analysis module is further configured to identify the temperature value of the first suspected target; if the temperature value belongs to the preset human body temperature range, the first suspected target is determined as the second suspected target.

[0043] In this embodiment, determining whether the first suspected target belongs to the human body area can further reduce the objects that need to be gesture-recognized, and the human body temperature has a specific range. Therefore, combining the preset human body temperature range can further narrow the range of the first suspected target.

[0044] In some embodiments, the hand analysis module is further configured to identify the gesture information of the second suspected target; if the gesture information conforms to the preset gesture, the second suspected target is determined as the target hand.

[0045] In this embodiment, a preset gesture is set to identify the target hand, improving the recognition accuracy. In addition, multiple preset gestures can be set to avoid the low recognition rate of a single preset gesture due to external factors affecting the user experience.

[0046] In some embodiments, the information acquisition module is further configured to prompt the gesture to be completed; acquire the image to be recognized in the area to be detected based on the gesture to be completed; the hand analysis module is further configured to determine the second suspected target as the target hand if the gesture information conforms to the gesture to be completed.

[0047] In this embodiment, since the type of the preset gesture can be freely set, some users may not be aware of the type of the preset gesture. Therefore, when acquiring the image to be recognized in the area to be detected, the preset gesture, that is, the gesture to be completed, is prompted to guide the user to make the preset gesture, improving the recognition accuracy of the target hand.

[0048] In some embodiments, the target analysis module is further configured to identify, through a deep learning network, a region in the image to be identified that matches a preset hand contour, and determine it as the first suspected target.

[0049] In this embodiment, the contours of each object in the image to be identified are analyzed through a deep learning network. At the same time, the deep learning network considers the relevance of the same object in multiple frames of images, so as to be able to identify more accurately.

[0050] In some embodiments, the hand analysis module is further configured to track and identify the target gesture of the target hand; determine and execute the corresponding target instruction according to the correspondence between the gesture and the instruction and the target gesture.

[0051] In this embodiment, after the target hand is determined, through the set correspondence between the gesture and the instruction, the target instruction can be quickly identified to control the device to perform corresponding operations, without actual contact, with lower requirements for the operation distance, and better user experience.

[0052] In a third aspect, the present application provides an electronic device, which includes a processor, a memory, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement the steps in any one of the above-mentioned target hand analysis methods.

[0053] In a fourth aspect, the present application provides a storage medium, in which a number of instructions are stored, and the instructions are used for a controller to execute to implement any one of the above-mentioned methods.

[0054] The target hand analysis method, device, electronic device, and storage medium provided by the embodiments of the present application determine the target hand by judging whether the target conforms to a preset gesture, avoiding misidentification caused by some objects being similar to the hand contour. In addition, before gesture recognition, preliminary screening is performed through the determination of the preset hand contour and whether it belongs to the human body area to narrow the recognition range and reduce the objects for gesture recognition in the later stage. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The following will make the technical solutions and other beneficial effects of the present application obvious by describing the specific embodiments of the present application in detail with reference to the drawings.

[0056] Figure 1 is a schematic diagram of the scenario of the target hand analysis method in the embodiments of the present application;

[0057] Figure 2 is a schematic flowchart of the target hand analysis method in the embodiments of the present application;

[0058] Figure 3It is a schematic flowchart of a target hand analysis method in another embodiment of the present application;

[0059] Figure 4 It is a schematic flowchart of a target hand analysis method in another embodiment of the present application;

[0060] Figure 5 It is a schematic structural diagram of a target hand analysis device in an embodiment of the present application;

[0061] Figure 6 It is a schematic structural diagram of an electronic device in an embodiment of the present application. Detailed implementation manners

[0062] Next, the technical solutions in the embodiments of the present application will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts shall fall within the protection scope of the present application.

[0063] In the description of the present application, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the described features. In the description of the present application, "a plurality of" means two or more, unless otherwise specifically defined.

[0064] In the description of the present application, the term "for example" is used to mean "serving as an example, illustration, or explanation". Any embodiment described as "for example" in the present application is not necessarily construed as being more preferred or having more advantages than other embodiments. In order to enable any person skilled in the art to implement and use the present invention, the following description is given. In the following description, details are set forth for the purpose of explanation. It should be understood that those skilled in the art can recognize that the present invention can be implemented without using these specific details. In other instances, well-known structures and processes are not described in detail to avoid unnecessary details from obscuring the description of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed in the present application.

[0065] In the embodiments of the present application, the target hand analysis method mainly involves computer vision technology (CV) in artificial intelligence (AI). Among them, artificial intelligence uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, and is a theory, method, technology, and application system that can perceive the environment, acquire knowledge, and use knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology in computer science that attempts to understand the essence of intelligence and produce an intelligent machine that can react in a way similar to human intelligence.

[0066] Computer vision is a science that studies how to enable machines to "see". Further, it refers to using cameras and computers to replace human eyes for machine vision such as target recognition, tracking, and measurement, and further performing graphic processing to make the computer-processed images more suitable for human eye observation or transmission to instrument detection. As a scientific discipline, computer vision studies related theories and technologies and attempts to build an artificial intelligence system that can obtain information from images or multi-dimensional data. Computer vision technology usually includes image processing, image recognition, image semantic understanding, image retrieval, optical character recognition (OCR), video processing, video semantic understanding, video content / behavior recognition, three-dimensional object reconstruction, 3D technology, virtual reality, augmented reality, simultaneous localization and mapping, etc. It also includes common biometric recognition technologies such as face recognition and fingerprint recognition.

[0067] In the embodiments of the present application, it should be noted that since the target hand analysis method provided in the present application is executed in an electronic device, the processing objects of each electronic device exist in the form of data or information. For example, time, which is actually time information. It can be understood that in subsequent embodiments, if dimensions, quantities, positions, etc. are mentioned, they are all corresponding data existences for the electronic device to process, and specific details are not elaborated here.

[0068] In the embodiments of the present application, it should also be noted that the target hand analysis method provided in the embodiments of the present application can be applied to, for example Figure 1In the target hand analysis system shown. Among them, the target hand analysis system includes a terminal 100 and a server 200. The terminal 100 can be a device that includes both receiving and transmitting hardware, that is, a device with receiving and transmitting hardware capable of performing two-way communication on a two-way communication link. Such a device can include: cellular or other communication devices, which have a single-line display or a multi-line display or a cellular or other communication device without a multi-line display. The terminal 100 can specifically be a desktop terminal or a mobile terminal. The terminal 100 can specifically also be one of a mobile phone, a tablet computer, a laptop computer, etc., or a camera installed at a monitoring site for information collection, storage, and transmission. The server 200 can be an independent server or a server network or server cluster composed of servers, including but not limited to a computer, a network host, a single network server, a set of multiple network servers, or a cloud server composed of multiple servers. Among them, the cloud server is composed of a large number of computers or network servers based on cloud computing.

[0069] Those skilled in the art can understand that Figure 1 the application environment shown in is only one application scenario of the solution of this application, and does not constitute a limitation on the application scenario of the solution of this application. Other application environments can also include more or fewer electronic devices than Figure 1 shown in, for example Figure 1 only 1 server 200 is shown in. It can be understood that the target hand analysis system can also include one or more other servers, which are not specifically limited here. In addition, as Figure 1 shown, the target hand analysis system can also include a memory for storing data, such as storing the to-be-recognized images in the to-be-detected area.

[0070] It should also be noted that Figure 1 the scene schematic diagram of the target hand analysis system shown is only an example. The target hand analysis system and scene described in the embodiments of the present invention are for more clearly explaining the technical solutions of the embodiments of the present invention, and do not constitute a limitation on the technical solutions provided by the embodiments of the present invention. Those skilled in the art know that with the evolution of the target hand analysis system and the emergence of new business scenarios, the technical solutions provided by the embodiments of the present invention are equally applicable to similar technical problems.

[0071] Please refer to Figure 2 , the embodiments of the present application provide a target hand analysis method, which is mainly exemplified by the method being applied to the server 200 in the above Figure 1 . The method includes steps S201 to S204, which are specifically as follows:

[0072] S201, Obtain the to-be-recognized images in the to-be-detected area.

[0073] Specifically, gesture recognition technology has currently been widely applied to various household appliances, such as smart TVs, video intercoms, etc. The area to be detected is the image acquisition area of the device that needs gesture recognition, or it can also be a specific part of the image acquisition area. This embodiment does not make specific limitations. In addition, the area to be detected can also be adjusted by adjusting the image acquisition device.

[0074] The to-be-recognized image in the area to be detected is collected through an image acquisition device such as a camera. Among them, since gesture recognition is a dynamic process, in fact, the collected video stream is analyzed. This embodiment takes one frame of the image as an example for illustration, and the recognition process of each frame of the image is the same. In addition, before this embodiment is used to recognize gestures, the target hand is determined. Therefore, it is also possible to select an image frame in the video stream that is more convenient for recognition as the to-be-recognized image, or to combine multiple frames of images as the to-be-recognized image for recognition.

[0075] The to-be-recognized image includes but is not limited to pictures, video frames in videos, etc., such as the picture directly captured by the surveillance video; the video includes but is not limited to short videos, long videos and other formats. The short video can be a video with a length less than 10 minutes, and the long video can be a video with a length greater than 10 minutes.

[0076] Specifically, before the server 200 executes the target hand analysis task, the user can send a task request to the server 200 through the terminal 100, and the task request carries the to-be-recognized image that needs to be recognized. After the server 200 receives the task request, it can perform detection and analysis based on the to-be-recognized image according to the target hand analysis method. Or, before the server 200 executes the target hand analysis task, the server 200 does not need to obtain the task request sent by the terminal 100. At this time, the terminal 100 is a camera with a shooting function. After the terminal 100 periodically or real-time collects and obtains the to-be-recognized image, it can be sent to the server 200 to execute the target hand analysis task. Furthermore, an image acquisition device can be installed on the terminal 100. The terminal 100 periodically or real-time collects videos or images, intercepts the to-be-recognized image and sends it to the server 200, so that the server 200 executes the target hand analysis task.

[0077] S202, determine a first suspected target according to the to-be-recognized image and the preset hand contour.

[0078] Specifically, since the hand includes multiple forms, such as clenched hands, stretched hands, fingers together, spread fingers, etc., multiple preset hand contours are included to correspond to different hand forms. The first suspected target is an object in the image to be identified that is determined to meet the preset hand contour feature area. Since there are many types of preset hand contours and the features of the preset hand contours are not unique, the image to be identified based on the preset hand contour may contain multiple first suspected targets, which require further identification and analysis.

[0079] In one embodiment, this step includes: S301, identifying the area in the image to be identified that matches the preset hand contour through a deep learning network, and determining it as the first suspected target.

[0080] Specifically, the area in the image to be identified that matches the preset hand contour is identified through a deep learning network, and is determined as the first suspected target. For example, an adaboost detector or a deep learning detector training method is used to train a detector that can traverse and locate and mark the area matching the preset hand contour on a frame of image. This method usually requires the use of LBP and HAAR. It should be noted that when the image to be identified is an image combination obtained by combining multiple frames of images, the deep learning network comprehensively analyzes the first suspected target in the multiple frames of images and correlates them with each other. Only when the area identified as the first suspected target in more than a preset number of images is determined as the first suspected target, otherwise it can be determined as a misidentification caused by a specific angle of a certain image.

[0081] S203: If the first suspected target belongs to the human body area, determine that the first suspected target is a second suspected target.

[0082] Specifically, the actual attribute of the first suspected target may be a human hand or other objects, so further determining whether the first suspected target belongs to the human body region can filter out most of the interference factors in the first suspected target. If the first suspected target belongs to the human body region, the first suspected target is determined to be the second suspected target. How to determine whether the first suspected target belongs to the human body region is not specifically limited in this embodiment.

[0083] In one embodiment, this step includes: S401, identifying a human body image in the image to be identified; S402, if the first suspected target belongs to the human body image, determining the first suspected target as a second suspected target.

[0084] Specifically, there may be multiple first suspected targets determined according to the image to be recognized and the preset hand contour, and their actual attributes may be human hands or other objects. Therefore, it is necessary to further identify whether the determined first suspected target is a human body to reduce the objects for which gesture recognition needs to be performed subsequently. When recognizing the image to be recognized, the human contour is relatively not easy to be confused with other objects. If an easily recognizable human image is collected in the area to be detected, that is, the human image in the image to be recognized is recognized, and it is further determined that the first suspected target belongs to the human image, then the first suspected target is determined to be the second suspected target, and the second suspected target is an object suspected of being a human hand, such as the left hand, the right hand, or parts with similar inner contours in the human image due to other factors, etc.

[0085] In one embodiment, this step includes: S501, identifying the temperature value of the first suspected target; S502, if the temperature value belongs to the preset human body temperature range, then determining that the first suspected target is the second suspected target.

[0086] Specifically, there may be multiple first suspected targets determined according to the image to be recognized and the preset hand contour, and their actual attributes may be human hands or other objects. Therefore, it is necessary to further identify whether the determined first suspected target is a human body to reduce the objects for which gesture recognition needs to be performed subsequently. Identify the temperature value of the first suspected target, compare it with the preset human body temperature range. If the temperature value belongs to the preset human body temperature range, then determine that the first suspected target is a human body part and determine that the first suspected target is the second suspected target.

[0087] In addition, when obtaining the image to be recognized in the area to be detected, the corresponding thermal infrared image can be obtained simultaneously to identify whether the first suspected target belongs to the human body area, that is, to determine whether the first suspected target is the second suspected target.

[0088] S204, if the gesture of the second suspected target conforms to the preset gesture, then determine that the second suspected target is the target hand.

[0089] Specifically, since a large amount of data needs to be processed for gesture recognition, if all suspected targets are recognized one by one, the processing speed will be too slow. After the screening in the foregoing steps, interference items are further eliminated, reducing the objects for which gesture recognition needs to be performed. Further, if it is recognized that the gesture of the second suspected target conforms to the preset gesture, then determine that the second suspected target is the target hand.

[0090] In one embodiment, as Figure 3 shown, this step includes: S601, identifying the gesture information of the second suspected target; S602, if the gesture information conforms to the preset gesture, then determine that the second suspected target is the target hand.

[0091] Specifically, gesture information of a second suspected target is recognized, which can be recognized through a deep network learning module or through similarity matching. The specific method is not specifically limited in this embodiment. If the gesture information of the second suspected target conforms to a preset gesture, the second suspected target is determined to be the target hand. The preset gesture is a specific gesture set in advance, such as palm stretching, palm clenching, etc. In addition, the preset gesture can be one or multiple. Setting multiple preset gestures can avoid the low recognition rate of a single preset gesture due to external factors and affect the user experience. It should be noted that the type and quantity of the preset gestures can be set differently based on needs, and are not specifically limited in this embodiment.

[0092] In one embodiment, as Figure 4 shown, step S201 of obtaining an image to be recognized within a region to be detected includes: S701, prompting a gesture to be completed; S702, obtaining an image to be recognized within the region to be detected based on the gesture to be completed; step S602 of determining the second suspected target to be the target hand if the gesture information conforms to the preset gesture includes: S703, determining the second suspected target to be the target hand if the gesture information conforms to the gesture to be completed.

[0093] Specifically, since the type of the preset gesture can be freely set and some users may not be aware of the type of the preset gesture, when obtaining the image to be recognized within the region to be detected, the preset gesture, that is, the gesture to be completed, can be prompted. Among them, it can be displayed through a display screen or prompted through a voice system to guide the user to make the preset gesture.

[0094] In addition, if there are multiple preset gestures, the user actually only needs to complete one of them. Setting multiple preset gestures is to facilitate the user to select a type that is easier for themselves to complete. In addition, if the user has no special requirements or instructions, the prompted gestures to be completed can be displayed according to the recognition success rate of each gesture, and the gesture to be completed with a higher recognition success rate is preferentially displayed. If no recognition is successful after exceeding the preset time, the next gesture to be completed is switched to prompt the user.

[0095] Obtain an image to be recognized within the region to be detected based on the gesture to be completed, then screen the suspected targets therein according to the steps of the above embodiment, and finally recognize the gesture information thereof to determine whether it is the target hand.

[0096] It should be noted that when there are multiple preset gestures, if there is no prompt for the unfinished gesture when obtaining the image to be recognized, it is necessary to compare the gesture information of the second suspected target with each of the multiple preset gestures one by one until a match is successful or all comparisons are completed. If there is a prompt for the unfinished gesture when obtaining the image to be recognized, it is only necessary to compare and recognize the gesture information of the second suspected target with the unfinished gesture, and the recognition process is faster.

[0097] In one embodiment, in step S204, if the gesture of the second suspected target conforms to the preset gesture, after determining that the second suspected target is the target hand, it includes: S801, tracking and recognizing the target gesture of the target hand; S802, determining and executing the corresponding target instruction according to the correspondence between the gesture and the instruction and the target gesture.

[0098] Specifically, after determining the target hand, taking the target hand as the recognition object, tracking and recognizing the target gesture of the target hand in real time, and then determining the target instruction corresponding to the target gesture in combination with the correspondence between the gesture and the instruction. Among them, the correspondence between the gesture and the instruction can be set in advance, and the gesture and the instruction are in one-to-one correspondence. For example, for a smart TV, stretching and sliding the palm corresponds to adjusting the sound volume, etc. The specific corresponding method is not specifically limited in this embodiment.

[0099] In this embodiment, by determining whether the target conforms to the preset gesture to determine the target hand, it is avoided that misrecognition is caused because some objects are similar to the hand contour. In addition, before gesture recognition, preliminary screening is carried out through the determination of the preset hand contour and whether it belongs to the human body area, narrowing the recognition range to reduce the objects for gesture recognition in the later stage.

[0100] In order to better implement the target hand analysis method in the embodiments of the present application, on the basis of the target hand analysis method, an embodiment of the present application also provides a target hand analysis device, as Figure 5 shown, the target hand analysis device 900 includes:

[0101] An information acquisition module 910, configured to acquire an image to be recognized in a region to be detected;

[0102] A target analysis module 920, communicatively connected to the information acquisition module 910, configured to determine a first suspected target according to the image to be recognized and a preset hand contour; if the first suspected target belongs to the human body area, determine the first suspected target as the second suspected target;

[0103] A hand analysis module 930, communicatively connected to the target analysis module 920, configured to determine that the second suspected target is the target hand if the gesture of the second suspected target conforms to the preset gesture.

[0104] In some embodiments of the present application, the target analysis module 920 is further configured to identify a human body image in the image to be identified; if the first suspected target belongs to the human body image, determine the first suspected target as the second suspected target.

[0105] In some embodiments of the present application, the target analysis module 920 is further configured to identify the temperature value of the first suspected target; if the temperature value belongs to a preset human body temperature range, determine the first suspected target as the second suspected target.

[0106] In some embodiments of the present application, the hand analysis module 930 is further configured to identify the gesture information of the second suspected target; if the gesture information conforms to a preset gesture, determine the second suspected target as the target hand.

[0107] In some embodiments of the present application, the information acquisition module 910 is further configured to prompt a gesture to be completed; acquire an image to be identified in the area to be detected based on the gesture to be completed; the hand analysis module 930 is further configured to determine the second suspected target as the target hand if the gesture information conforms to the gesture to be completed.

[0108] In some embodiments of the present application, the target analysis module 920 is further configured to identify, through a deep learning network, an area in the image to be identified that matches a preset hand contour, and determine it as the first suspected target.

[0109] In some embodiments of the present application, the hand analysis module 930 is further configured to track and identify the target gesture of the target hand; determine and execute a corresponding target instruction according to the correspondence between the gesture and the instruction and the target gesture.

[0110] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For parts not detailed in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

[0111] In some embodiments of the present application, the target hand analysis device 900 may be implemented in the form of a computer program, and the computer program can run on an electronic device as shown in Figure 6 The memory of the electronic device may store each program module that constitutes the target hand analysis device 900. For example, Figure 5 the information acquisition module 910, the target analysis module 920, and the hand analysis module 930 shown. The computer program constituted by each program module enables the processor to execute the steps in the target hand analysis method of each embodiment of the present application described in this specification.

[0112] For example, Figure 6 the electronic device shown can be connected through, for example, Figure 5The information acquisition module 910 in the target hand analysis device 900 shown executes step S201. The electronic device can execute step S202 through the target analysis module 920. The electronic device can execute step S203 through the hand analysis module 930. The electronic device includes a processor, a memory, and a network interface connected through a system bus. Among them, the processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the electronic device is used to communicate with external electronic devices through a network connection. When the computer program is executed by the processor, it realizes a target hand analysis method.

[0113] Those skilled in the art can understand that Figure 6 the structure shown in is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the electronic device to which the solution of this application is applied. The specific electronic device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0114] In some embodiments of this application, an electronic device is provided, including one or more processors; a memory; and one or more applications, where the one or more applications are stored in the memory and configured to be executed by the processor to perform the steps of the above-mentioned target hand analysis method. Here, the steps of the target hand analysis method may be the steps in the target hand analysis method of the above-mentioned various embodiments.

[0115] In some embodiments of this application, a computer-readable storage medium is provided, storing a computer program, and the computer program is loaded by the processor, so that the processor executes the steps of the above-mentioned target hand analysis method. Here, the steps of the target hand analysis method may be the steps in the target hand analysis method of the above-mentioned various embodiments.

[0116] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Any reference to a memory, storage, database, or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0117] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered to be within the scope described in this specification.

[0118] The above has introduced in detail a target hand analysis method, device, electronic device, and storage medium provided by the embodiments of the present application. Specific examples are used in this article to elaborate on the principles and implementation manners of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A target hand analysis method, characterized in that Including: Obtain a to-be-recognized image within a to-be-detected area; Determine a first suspected target according to the to-be-recognized image and a preset hand contour; The first suspected target is an object in the to-be-recognized image that is determined to conform to the area with the preset hand contour features; If the first suspected target belongs to a human body area, determine the first suspected target as a second suspected target; the second suspected target is an object suspected to be a human hand; If the gesture of the second suspected target conforms to a preset gesture, determine the second suspected target as a target hand; The step of if the first suspected target belongs to a human body area, determining the first suspected target as a second suspected target, includes: Identify the temperature value of the first suspected target; If the temperature value belongs to a preset human body temperature range, determine the first suspected target as a second suspected target.

2. The target hand analysis method according to claim 1, wherein The step of if the first suspected target belongs to a human body area, determining the first suspected target as a second suspected target, includes: Identify the human body image in the to-be-recognized image; If the first suspected target belongs to the human body image, determine the first suspected target as a second suspected target.

3. The target hand analysis method according to claim 1, wherein The step of if the gesture of the second suspected target conforms to a preset gesture, determining the second suspected target as a target hand, includes: Identify the gesture information of the second suspected target; If the gesture information conforms to a preset gesture, determine the second suspected target as a target hand.

4. The target hand analysis method according to claim 3, wherein The step of obtaining a to-be-recognized image within a to-be-detected area includes: Prompt a to-be-completed gesture; Obtain a to-be-recognized image within the to-be-detected area based on the to-be-completed gesture; The step of if the gesture information conforms to a preset gesture, determining the second suspected target as a target hand, includes: If the gesture information conforms to the to-be-completed gesture, determine the second suspected target as a target hand.

5. The target hand analysis method according to claim 1, wherein The step of determining a first suspected target according to the to-be-recognized image and a preset hand contour includes: Identify, through a deep learning network, the area in the to-be-recognized image that matches the preset hand contour, and determine it as the first suspected target.

6. The target hand analysis method according to claim 1, wherein, After the step of if the gesture of the second suspected target conforms to a preset gesture, determining the second suspected target as a target hand, includes: Track and identify the target gesture of the target hand; Determine and execute a corresponding target instruction according to the correspondence between the gesture and the instruction and the target gesture.

7. A target hand analysis device, characterized in that, Including: An information acquisition module, configured to obtain a to-be-recognized image within a to-be-detected area; A target analysis module, communicatively connected to the information acquisition module, configured to determine a first suspected target according to the to-be-recognized image and a preset hand contour; the first suspected target is an object in the to-be-recognized image that is determined to conform to the area with the preset hand contour features; if the first suspected target belongs to a human body area, determine the first suspected target as a second suspected target; the second suspected target is an object suspected to be a human hand; A hand analysis module, communicatively connected to the target analysis module, configured to if the gesture of the second suspected target conforms to a preset gesture, determine the second suspected target as a target hand; The target analysis module is further configured to identify the temperature value of the first suspected target; if the temperature value belongs to a preset human body temperature range, determine that the first suspected target is a second suspected target.

8. An electronic device, characterized in that, The electronic device includes a processor, a memory, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement the steps in the target hand analysis method according to any one of claims 1 to 6.

9. A storage medium storing a plurality of instructions, characterized in that, The instructions are used to be executed by a controller to implement the method according to any one of claims 1 to 6.

Images