Image processing method and device applied in vehicle cabin, equipment and storage medium

Abstract

The embodiment of the invention discloses an image processing method and device applied in a vehicle cabin, equipment and a storage medium, and relates to the field of image recognition processing. According to the specific implementation scheme, at least one to-be-processed image in a vehicle cabin is obtained, the at least one to-be-processed image comprises an image under a target collection view angle, and the target collection view angle corresponds to a view angle corresponding to a mirror surface material laid at the bottom of the vehicle cabin; inputting the at least one to-be-processed image into a gesture detection model so as to output a gesture category and confidence corresponding to a target part when the to-be-processed image comprises the target part; determining a target gesture corresponding to the at least one to-be-processed image based on the gesture category and the confidence coefficient corresponding to the at least one to-be-processed image; and sending a control instruction to a target component corresponding to the target gesture, so that the target component executes the target event based on the control instruction. According to the embodiment of the invention, gestures can be effectively recognized, so that the effect of executing corresponding events is achieved.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of image recognition processing, and in particular to an image processing method, device, equipment and storage medium applied in a vehicle cabin. Background Art

[0002] With the popularization of intelligence, intelligence is reflected in all aspects of life. For example, in the automotive field, intelligence is mainly reflected in human-computer interaction. For example, components in the vehicle can be controlled based on voice or gestures.

[0003] At present, the recognition of user gestures in vehicles mainly involves the recognition and processing of the collected images to be processed by the camera device deployed at the target position to determine the user's gesture in the image to be processed. The above-mentioned method for gesture recognition has the following problems: the camera device is deployed at a fixed position, and can only recognize the user's gestures at a specific position, which has the problem of limited gesture recognition range. Furthermore, it is easily affected by ambient light during the gesture recognition process, resulting in low recognition accuracy. Finally, in the recognition process, the user's unconscious gestures may be recognized as specific gestures, resulting in low recognition effectiveness. Summary of the invention

[0004] Embodiments of the present invention provide an image processing method, apparatus, device and storage medium applied in a vehicle cabin, so as to realize effective and accurate recognition of gestures of a user in the vehicle cabin, thereby controlling corresponding components to execute target instructions.

[0005] In a first aspect, an embodiment of the present invention provides an image processing method applied in a vehicle cabin, comprising:

[0006] Acquire at least one image to be processed of the interior of the vehicle cabin, wherein the at least one image to be processed includes an image under a target acquisition viewing angle, and the target acquisition viewing angle corresponds to a viewing angle corresponding to a mirror material laid on the bottom of the vehicle cabin;

[0007] Inputting at least one image to be processed into a gesture detection model to output a gesture category and a confidence level corresponding to the target part when the image to be processed includes the target part;

[0008] Determining a target gesture corresponding to at least one image to be processed based on a gesture category and a confidence level corresponding to at least one image to be processed;

[0009] A control instruction is sent to a target component corresponding to the target gesture, so that the target component executes a target event based on the control instruction.

[0010] Furthermore, before acquiring at least one image to be processed of the interior of the vehicle cabin, the method further includes:

[0011] Based on a camera device with data acquisition and transmission function deployed at the bottom of the cabin, a first original image of the interior of the cabin under a target acquisition angle of view is acquired; wherein the camera device is associated with a mirror material laid at the bottom of the cabin; and / or,

[0012] collecting a second original image of the vehicle cabin at a first viewing angle based on a camera device deployed on the top of the vehicle cabin;

[0013] Preprocessing the first original image and the second original image respectively to obtain images to be processed;

[0014] The preprocessing includes grayscale processing of the original image and then filtering processing.

[0015] Furthermore, the at least one image to be processed includes two images to be processed generated at the same time, and the gesture detection model includes a part detection sub-model and a gesture detection sub-model;

[0016] The step of inputting the at least one image to be processed into a gesture detection model to output a gesture category and a confidence level corresponding to the target part when the image to be processed includes the target part comprises:

[0017] Processing the two images to be processed based on the part detection sub-model to output a part detection frame corresponding to the target part in each image to be processed;

[0018] The images within the part detection frames corresponding to the two images to be processed are input into the gesture detection sub-model, so that the gesture detection sub-model outputs the gesture category and confidence corresponding to the target part in each image to be processed.

[0019] Furthermore, the gesture detection model further includes a part matching sub-model connected to the part detection sub-model, and the output result of the gesture detection model further includes a part matching result. The method further includes:

[0020] The images within the part detection frames corresponding to the two images to be processed are input into the part matching sub-model, and the part matching results of the two images to be processed are output.

[0021] Furthermore, the at least one image to be processed is two images to be processed acquired at the same time, and determining the target gesture corresponding to the at least one image to be processed based on the gesture category and confidence corresponding to the at least one image to be processed includes:

[0022] When the part matching results of the two images to be processed are preset results and the gesture categories are the same, weighting the confidences of the two images to be processed to obtain a target confidence;

[0023] When the target confidence is higher than a preset confidence threshold, a target gesture corresponding to the gesture category is determined.

[0024] Furthermore, the method further comprises:

[0025] When the part matching results of the two images to be processed are different from the preset results, the confidence of the image to be processed under the target acquisition viewing angle is used as the target confidence;

[0026] When the target confidence is higher than a preset confidence threshold, a target gesture corresponding to the gesture category is determined.

[0027] Further, the at least one image to be processed includes an image to be processed, and the at least one image to be processed is input into a gesture detection model to output a gesture category and a confidence level corresponding to the target part when the image to be processed includes the target part, including:

[0028] Inputting the image to be processed into the gesture detection model to identify a part detection frame corresponding to a target part in the image to be processed based on a part detection sub-model in the gesture detection model;

[0029] Based on the gesture detection sub-model in the gesture detection model, image recognition processing is performed in the part detection frame to obtain a gesture category and a confidence level corresponding to the target part;

[0030] Accordingly, determining the target gesture corresponding to the at least one image to be processed based on the gesture category and confidence level corresponding to the at least one image to be processed includes:

[0031] If the confidence level is higher than a preset confidence threshold, a target gesture corresponding to the gesture category is determined.

[0032] In a second aspect, an embodiment of the present invention provides an image processing device for use in a vehicle cabin, comprising:

[0033] An image acquisition module, used to acquire at least one image to be processed of the interior of the vehicle cabin, wherein the at least one image to be processed includes an image under a target acquisition viewing angle, and the target acquisition viewing angle corresponds to a viewing angle corresponding to a mirror material laid on the bottom of the vehicle cabin;

[0034] A gesture detection module, used to input at least one image to be processed into a gesture detection model, so as to output a gesture category and a confidence level corresponding to the target part when the image to be processed includes the target part;

[0035] A gesture determination module, used to determine a target gesture corresponding to at least one image to be processed based on the gesture category and confidence level corresponding to at least one image to be processed;

[0036] The instruction sending module is used to send a control instruction to a target component corresponding to a target gesture, so that the target component executes a target event based on the control instruction.

[0037] In a third aspect, an embodiment of the present invention provides an electronic device, including:

[0038] at least one processor;

[0039] and a memory in communication with the at least one processor;

[0040] The memory stores instructions that can be executed by at least one processor, and the instructions are executed by at least one processor so that the at least one processor can execute the image processing method applied to the vehicle cabin provided by any embodiment of the present invention.

[0041] In a fourth aspect, an embodiment of the present invention further provides a non-transitory computer-readable storage medium storing computer instructions, where the computer instructions are used to enable a computer to execute the image processing method applied to a vehicle cabin provided by any embodiment of the present invention.

[0042] The technical solution provided by the embodiment of the present invention can collect the image to be processed under the target perspective based on the mirror material laid on the bottom of the cabin, and can collect the image to be processed in the cabin based on the camera device deployed in the cabin. The above-collected image to be processed can be input into the gesture detection model, so that when the target part is included in the image to be processed, the gesture category and confidence corresponding to the target part can be output. According to the gesture category and confidence of the image to be processed, the target gesture corresponding to the image to be processed can be determined. After determining the target gesture, a control instruction can be sent to the target part corresponding to the target gesture, so that the target component executes the target event based on the control. Since the processed image to be processed includes the image to be processed corresponding to the mirror material laid on the bottom of the cabin, and this image is mainly collected from the bottom, generally speaking, the collection range is relatively wide. When processing based on the above-mentioned image to be processed, the accuracy and effectiveness of the target part recognition can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 is a flow chart of an image processing method applied to a vehicle cabin provided by an embodiment of the present invention;

[0045] Figure 2 is a flow chart of an image processing method applied to a vehicle cabin provided by an embodiment of the present invention;

[0046] Figure 3 is a flow chart of an image processing method applied to a vehicle cabin provided by an embodiment of the present invention;

[0047] Figure 4 is a flow chart of an image processing method applied to a vehicle cabin provided by an embodiment of the present invention;

[0048] Figure 5 is a flow chart of an image processing method applied to a vehicle cabin provided by an embodiment of the present invention;

[0049] Figure 6 It is a picture taken by a camera installed at the front reading light in the cabin provided by an embodiment of the present invention;

[0050] Figure 7 is a schematic structural diagram of an image processing device applied in a vehicle cabin provided by an embodiment of the present invention;

[0051] Figure 8 4 is a block diagram of an image processing electronic device applied in a vehicle cabin provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0053] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0054] Before introducing the technical solution provided by the embodiment of the present invention, an application scenario may be exemplified. The solution provided by the embodiment of the present invention may be applied to a scenario of performing gesture recognition on all users in any closed or semi-closed area.

[0055] Figure 1 It is a flowchart of an image processing method applied in a vehicle cabin provided by an embodiment of the present invention. This embodiment is applicable to any scenario where gesture information needs to be recognized. The method can be executed by an image processing device applied in a vehicle cabin. The device can be implemented in hardware / software. The device can be integrated in an electronic device. Optionally, the electronic device can be a mobile terminal or a PC, etc.

[0056] like Figure 1 As shown, an image processing method applied in a vehicle cabin disclosed in this embodiment may include:

[0057] S110, obtaining at least one image to be processed of the interior of the vehicle cabin, wherein the at least one image to be processed includes an image under a target acquisition perspective, and the target acquisition perspective corresponds to a perspective corresponding to a mirror material laid on the bottom of the vehicle cabin.

[0058] The image to be processed may be directly acquired image data without any processing, or may be an image after preprocessing. For example, the image to be processed may be an initial image captured by an image acquisition device in a vehicle, or may be an image obtained after noise reduction processing of an image captured by an image acquisition device. The target acquisition viewing angle may be a preset viewing angle that must be captured, for example, the acquisition viewing angle corresponding to the mirror material laid on the bottom of the cabin is used as the target acquisition viewing angle. The mirror material may be a material that forms a clear image through refraction, reflection or mirroring. Products made of mirror material may be laid on the bottom of the cabin to capture images to be processed of all users in the cabin.

[0059] Specifically, the front camera at the front reading light in the cabin can be used to collect the gesture image of the user in the cabin and use it as the image to be processed. At the same time, the mirror material laid on the bottom of the cabin can be used to provide the image to be processed under the target viewing angle.

[0060] It should be noted that the mirror material laid on the bottom of the cabin may include a transmission module, so that after acquiring the image to be processed under the target viewing angle, the image to be processed can be transmitted to the processing module of the vehicle to continue the subsequent recognition processing. Exemplarily, an image is captured by the camera device corresponding to the mirror material laid on the bottom of the cabin, and this image is used as the image to be processed. The image to be processed may also include an image captured by the front camera at the same time.

[0061] S120: Input at least one image to be processed into a gesture detection model, so as to output a gesture category and a confidence level corresponding to the target part when the image to be processed includes the target part.

[0062] Among them, the gesture detection model can be a neural network model currently authorized for use, which can identify whether the image to be processed includes the corresponding gesture, the gesture category and the confidence corresponding to the gesture category. The target part can be a pre-set part to be detected. Optionally, the target part is a hand part. The gesture category can be the category identifier corresponding to various hand gestures during human-computer interaction, for example, the gesture category of the OK gesture is 1, the gesture category of the thumbs up is 2, and the gesture category of the high-five is 3. The confidence level can be a quantitative representation of the reliability of a certain prediction result. Optionally, the higher the confidence level, the more accurate the gesture category output by the gesture detection model is. Conversely, the greater the error between the gesture category output by the gesture detection model and the actual one.

[0063] Specifically, the image to be processed can be input into the gesture detection model. When the gesture detection model detects the existence of the target part in the image, the gesture category corresponding to the target part and the reliability corresponding to the gesture category, i.e., the confidence level, can be output. If the gesture detection model does not have the target part, the corresponding result may not be output.

[0064] For example, taking the target part as the hand, the collected image to be processed can be input into the gesture detection model. If the gesture detection model detects that the image to be processed includes the hand, it can output the gesture category of the hand and the confidence level corresponding to the gesture category.

[0065] S130: Determine a target gesture corresponding to at least one image to be processed based on the gesture category and confidence level corresponding to at least one image to be processed.

[0066] The target gesture is the final gesture determined after analyzing and processing the gesture category and confidence in the image to be processed.

[0067] Exemplarily, the gesture category of the image to be processed is a high-five gesture category, and the confidence level is 0.8, then it is determined that the target gesture is a high-five gesture.

[0068] S140: Send a control instruction to a target component corresponding to the target gesture, so that the target component executes a target event based on the control instruction.

[0069] The target component may be a component or module responsible for implementing a certain function, for example, the target component may be a button for controlling music. The control instruction may be a response command to an input, for example, the control instruction may be a command for playing music. The target event may be a specific event that is expected to occur and is initiated by the user, for example, the target event may be playing music.

[0070] Specifically, a mapping relationship between different gesture categories and control components can be created in advance, so that after the target gesture is determined, the target component corresponding to the target gesture can be determined based on the mapping relationship, and a corresponding control instruction can be sent to the target component to make the target part work.

[0071] Exemplarily, the target gesture is a high-five gesture, and the corresponding target component is a music component. After the high-five gesture is detected, a control instruction can be sent to the music component to start the music component to work.

[0072] The technical solution provided by the embodiment of the present invention can collect the image to be processed under the target perspective based on the mirror material laid on the bottom of the cabin, and can collect the image to be processed in the cabin based on the camera device deployed in the cabin. The above-collected image to be processed can be input into the gesture detection model, so that when the target part is included in the image to be processed, the gesture category and confidence corresponding to the target part can be output. According to the gesture category and confidence of the image to be processed, the target gesture corresponding to the image to be processed can be determined. After determining the target gesture, a control instruction can be sent to the target part corresponding to the target gesture, so that the target component executes the target event based on the control. Since the processed image to be processed includes the image to be processed corresponding to the mirror material laid on the bottom of the cabin, and this image is mainly collected from the bottom, generally speaking, the collection range is relatively wide. When processing based on the above-mentioned image to be processed, the accuracy and effectiveness of the target part recognition can be improved.

[0073] Figure 2It is a flow chart of the image processing method applied to the vehicle cabin provided by an embodiment of the present invention. On the basis of the aforementioned embodiment, how to obtain the image to be processed can be further refined. Its specific implementation method can refer to the detailed description of this embodiment, wherein the technical terms that are the same as or corresponding to the aforementioned embodiment are not repeated in this embodiment.

[0074] like Figure 2 As shown, an image processing method applied in a vehicle cabin disclosed in this embodiment may include:

[0075] Step 1: A first original image of the interior of the cabin under a target acquisition perspective is acquired based on a camera device with data acquisition and transmission functions deployed at the bottom of the cabin.

[0076] Specifically, the camera device at the bottom of the cabin can not only capture images, but also has a data transmission function. The image captured by the camera device can be used as the first original image. It should be noted that the camera device is associated with the mirror material laid on the bottom of the cabin, that is, the mirror material laid on the bottom of the cabin can be used as a camera device.

[0077] Exemplarily, an image is captured by a camera device at the bottom of the cabin, and the target viewing angle corresponding to the image is an upward viewing angle.

[0078] Step 2: Capture a second original image of the vehicle cabin at the first viewing angle based on a camera device deployed on the top of the vehicle cabin.

[0079] The first viewing angle may be an angle corresponding to the camera device on the top of the cabin, for example, the first viewing angle may be a frontal viewing angle. The image captured by the camera device on the top of the cabin at the first viewing angle is used as the second original image.

[0080] Exemplarily, an image is captured using a camera device on top of the cabin, and the viewing angle corresponding to the image is a frontal viewing angle.

[0081] Step three, preprocessing the first original image and the second original image respectively to obtain the to-be-processed image.

[0082] In this embodiment, the first original image and the second original image may be preprocessed respectively to obtain images to be processed.

[0083] Exemplarily, image preprocessing is performed on the image captured by the camera device at the bottom of the cabin and the image captured by the camera device at the top of the cabin to obtain an image to be processed.

[0084] In this embodiment, the pre-processing of the first original image and the second original image may be: grayscale processing of the first original image and the second original image, followed by denoising.

[0085] Among them, the first original image and the second original image are both color images, that is, the pixel points are images including RGB values. Grayscale processing can be understood as the processing of converting a color image into a grayscale image. Optionally, the grayscale processing method includes but is not limited to a simple pixel averaging method, a weighted averaging method, and histogram equalization. In this embodiment, the grayscale processing algorithm is not limited as long as the grayscale processing effect can be achieved. In this embodiment, grayscale processing can reduce the amount of data for subsequent processing, thereby saving computing resources. Filtering processing can be to enhance or suppress certain components in an image by some method. Optionally, the filtering processing method includes but is not limited to filtering algorithms such as mean filtering, median filtering, and Gaussian blur. Filtering processing can remove random noise in the image to achieve the effect of improving image quality.

[0086] The technical solution provided by the embodiment of the present invention can pre-process the collected first original image and the second original image to obtain the image to be processed, so that the subsequent gesture detection model can analyze and process the image to be processed, thereby improving the data processing efficiency and reducing the data processing amount.

[0087] Figure 3 It is a flow chart of the image processing method applied to the vehicle cabin provided by an embodiment of the present invention. If at least one image contains two images to be processed generated at the same time, it can be processed based on the technical solution of this embodiment. The specific implementation method can refer to the detailed description of this embodiment, among which the technical terms that are the same as or corresponding to the above embodiments are not repeated in this embodiment.

[0088] like Figure 3 As shown, an image processing method applied in a vehicle cabin disclosed in this embodiment may include:

[0089] S310: Acquire two images to be processed generated at the same time inside the vehicle cabin.

[0090] That is to say, the number of images to be processed generated at the same time may be one or two. When the images to be processed generated at the same time include two images, two images to be processed generated at the same time inside the vehicle cabin may be obtained. Among them, one of the two images to be processed is an image preprocessed from the first original image, and the other is an image preprocessed from the second original image.

[0091] S320, processing the two images to be processed based on the part detection sub-model to output a part detection frame corresponding to the target part in each image to be processed.

[0092] The part detection submodel may be a part of the gesture detection model. For example, the part detection submodel is used to determine whether the image to be processed includes the target part. If the target part is included, the target part may be selected in the image to be processed to obtain the image to be processed including the part detection frame. In other words, the part detection frame is a detection frame that selects the target part in the image to be processed.

[0093] Specifically, the part detection submodel in the gesture detection model identifies the target part in the image to be processed and selects the target part to obtain a part detection frame including the target part in the image to be processed. Further, the gesture detection submodel identifies the image content in the part detection frame to determine the gesture category and the corresponding confidence.

[0094] Exemplarily, the hand detection sub-model is used to detect two images captured by a camera device inside the vehicle cabin. When hand parts exist in the images, the hand parts can be framed out in the image to be processed to obtain a part detection frame including the hand parts.

[0095] S330, input the images within the part detection frames corresponding to the two images to be processed into the part matching sub-model, and output the part matching results of the two images to be processed.

[0096] In this embodiment, the part matching sub-model may be a yolov8 model. The part matching sub-model may analyze and process the images within the part detection frames in the two images to be processed to determine whether they are the hand parts corresponding to the same user. The part matching results include the hand parts corresponding to the same user in the two images to be processed, or the hand parts in the two images to be processed are not the parts corresponding to the same user.

[0097] Specifically, the image contents within the part detection boxes in the two images to be processed can be input into the part matching model, and the part matching model can output the part matching results of whether the target parts in the two images to be processed belong to the same user. Optionally, the output result of the part matching sub-model is 0, indicating that the target parts in the two images to be processed do not belong to the same user. Conversely, if the output result of the part matching sub-model is 1, it indicates that the target parts in the two images to be processed belong to the same user.

[0098] S341: When the part matching results of the two images to be processed are preset results and the gesture categories are the same, weighted processing is performed on the confidences of the two images to be processed to obtain a target confidence.

[0099] The preset result may be that the part matching result belongs to the same user. The gesture detection submodel may output the gesture category and confidence level corresponding to the target part in each image to be processed. The target confidence level is the confidence level finally determined.

[0100] Specifically, if the part matching result of the two images to be processed is a preset result, the gesture categories of the two images to be processed can be obtained. If the gesture categories are the same, the confidence of the brightness image to be processed can be weighted to obtain the final target confidence. If the target confidence is higher than the preset confidence threshold, the gesture corresponding to the gesture category can be used as the target gesture.

[0101] S342: When the part matching result of the two images to be processed is different from the preset result, the confidence of the image to be processed under the target acquisition viewing angle is used as the target confidence.

[0102] It can be understood that if the part matching results of the two images to be processed are inconsistent with the preset results, that is, the users corresponding to the target parts in the two images to be processed are not the same user, then the confidence of the image to be processed under the target acquisition perspective can be used as the target confidence.

[0103] S350: When the target confidence is higher than a preset confidence threshold, determine a target gesture corresponding to the gesture category.

[0104] The preset confidence threshold may be a value preset according to actual conditions or empirical data, and the preset confidence threshold is used to indicate whether the gesture corresponding to the gesture category can be used as a target gesture.

[0105] Specifically, if the target confidence is higher than a preset confidence threshold, it means that the determined gesture category is accurate, and the gesture corresponding to the gesture category can be used as the target gesture.

[0106] Exemplarily, when the confidence threshold is preset to 0.8, whether it is the predicted probability obtained by weighted calculation or the predicted probability corresponding to the image captured by the camera device at the bottom of the cabin, when the predicted probability is higher than 0.8, the gesture corresponding to the gesture category output by the gesture detection model can be used as the target gesture.

[0107] The technical solution provided by the embodiment of the present invention can collect the image to be processed under the target perspective based on the mirror material laid on the bottom of the cabin, and can collect the image to be processed in the cabin based on the camera device deployed in the cabin. The above-collected image to be processed can be input into the gesture detection model, so that when the target part is included in the image to be processed, the gesture category and confidence corresponding to the target part can be output. According to the gesture category and confidence of the image to be processed, the target gesture corresponding to the image to be processed can be determined. After determining the target gesture, a control instruction can be sent to the target part corresponding to the target gesture, so that the target component executes the target event based on the control. Since the processed image to be processed includes the image to be processed corresponding to the mirror material laid on the bottom of the cabin, and this image is mainly collected from the bottom, generally speaking, the collection range is relatively wide. When processing based on the above-mentioned image to be processed, the accuracy and effectiveness of the target part recognition can be improved.

[0108] Embodiment 4

[0109] Figure 4 This is a flowchart of an image processing method applied to a vehicle cabin provided by Embodiment 4 of the present invention. If at least one image includes an image to be processed, the image is processed based on the technical solution of this embodiment. For specific implementation methods, please refer to the detailed description of this embodiment, wherein the technical terms that are the same as or corresponding to the above embodiments are not repeated in this embodiment.

[0110] like Figure 4 As shown, an image processing method applied in a vehicle cabin disclosed in this embodiment may include:

[0111] S410: Acquire an image to be processed under a target viewing angle.

[0112] It can be understood that if the images to be processed obtained at the same time only include the images to be processed corresponding to the first original image, then they may not be processed. If the images to be processed obtained at the same time only include the images to be processed corresponding to the second original image, then these images to be processed may be analyzed and processed.

[0113] S420: Input the image to be processed into the gesture detection model to identify a part detection frame corresponding to the target part in the image to be processed based on the part detection sub-model in the gesture detection model.

[0114] Specifically, the image to be processed is input into the gesture detection model, and the part detection sub-model in the gesture detection model will identify the part detection frame in the image.

[0115] It should be noted that since there is only one image to be processed, there is no need to use the part matching sub-model for matching. It is only necessary to input the image within the part detection frame into the gesture recognition model, or to input the image to be processed including the part detection frame into the gesture detection sub-model to output the gesture category and confidence of the image to be processed.

[0116] S430 , based on the gesture detection sub-model in the gesture detection model, perform image recognition processing on the part detection frame to obtain a gesture category and a confidence level corresponding to the target part.

[0117] It can be understood that the image recognition processing in the part detection frame is performed based on the gesture detection sub-model to obtain the gesture category and confidence corresponding to the target part. The confidence at this time is the target confidence mentioned above.

[0118] S440: If the confidence level is higher than a preset confidence threshold, determine a target gesture corresponding to the gesture category.

[0119] Specifically, if the confidence level output by the gesture detection sub-model is higher than a preset confidence threshold, the target gesture corresponding to the gesture category output by the gesture detection sub-model may be determined.

[0120] The technical solution provided by the embodiment of the present invention can collect the image to be processed under the target perspective based on the mirror material laid on the bottom of the cabin, and can collect the image to be processed in the cabin based on the camera device deployed in the cabin. The above-collected image to be processed can be input into the gesture detection model, so that when the target part is included in the image to be processed, the gesture category and confidence corresponding to the target part can be output. According to the gesture category and confidence of the image to be processed, the target gesture corresponding to the image to be processed can be determined. After determining the target gesture, a control instruction can be sent to the target part corresponding to the target gesture, so that the target component executes the target event based on the control. Since the processed image to be processed includes the image to be processed corresponding to the mirror material laid on the bottom of the cabin, and this image is mainly collected from the bottom, generally speaking, the collection range is relatively wide. When processing based on the above-mentioned image to be processed, the accuracy and effectiveness of the target part recognition can be improved.

[0121] As an alternative embodiment of the above embodiment, Figure 5 The figure is a flow chart of an image processing method applied in a vehicle cabin provided by an embodiment of the present invention. Figure 6 It is the picture content captured by the camera installed at the front reading light in the cabin provided by the embodiment of the present invention. Based on the above picture content, it can be known that the hand parts of all users in the vehicle can be included. For some users' unconscious actions, false detection and false alarm phenomena are often caused. Figure 6For the yellow circle shown in the figure, there may be false alarms. In order to solve this problem, the hand detection area can be set, that is, the hand will only be detected when it is placed in the specified detection area in the car, as shown in the red box in the figure. The false alarm problem can be solved to a certain extent, but because the visual camera cannot obtain distance information, false alarms will still occur when the occupant sits behind the frame. In addition, when the co-pilot and the back-seat user make gestures at the same time, the gestures in the back row will not be seen due to visual occlusion, and the accuracy of gesture recognition in the front row will also be affected.

[0122] Based on the above factors, hand gestures can be detected by combining mirror reflection images and front camera images to reduce the effect of false detection of unconscious gestures.

[0123] The solution provided in the embodiment of the present invention proposes a method for gesture recognition using mirror image data, which not only solves the problem of false alarms of unconscious gestures (when the user is sitting, there is no hand information in the mirror below, only in the front camera image. In this case, the present invention will filter out the gesture), but also solves the problem of simultaneous recognition of front and rear gestures (when there are gestures in the front and rear at the same time, there is no occlusion problem in the mirror image and they can be clearly recognized).

[0124] Next, combine Figure 5 To understand the solution provided by the embodiment of the present invention:

[0125] First of all, the hardware configuration uses a front camera and a cabin mirror material.

[0126] The front camera is installed at the front reading light in the car to capture the gestures of the passengers. The bottom of the vehicle cabin is made of mirror material to provide a perspective for reflecting the images of the passengers.

[0127] Secondly, image acquisition and preprocessing are performed.

[0128] In order to reduce false detection and distinguish gestures of multiple users, image acquisition is divided into two situations: mirror image acquisition and front camera image acquisition. The mirror reflection image contains the gesture information of the occupant and serves as the main gesture detection source. The image collected from the front row assists in obtaining the image information of the occupant's gesture to increase the confidence of gesture recognition. Image preprocessing is divided into grayscale processing and denoising processing. Grayscale processing converts color images into grayscale images to simplify subsequent image processing. Denoising processing uses filtering algorithms such as Gaussian blur to remove noise in the image and enhance image quality.

[0129] Next, hand matching is performed.

[0130] When there are two images to be processed at the same time, the mirror reflection image and the front camera image can be detected based on the part detection sub-model such as YOLOV8 to see whether the hand frame (part detection frame) is included. If there are part detection frames in both images to be processed, they are matched, that is, if there are hand parts in both, determine whether the hand parts are in the same position, that is, whether they are the hand parts of the same user.

[0131] Then, gesture detection is performed.

[0132] Use pre-trained deep learning models (such as YOLO, Faster R-CNN) (gesture detection sub-model) to perform gesture detection on mirror reflection images and front camera images respectively. The specific steps are: input the pre-processed image into the gesture detection model to obtain the detection result; extract the gesture category and confidence information in the detection result.

[0133] Finally, the gesture confidence is weighted to improve recognition accuracy.

[0134] It should be noted that if the gesture recognition result only contains the result of the front-row camera, but not the result corresponding to the mirror reflection image, it can be ignored and considered as an unconscious false detection; if only the gesture is detected in the mirror reflection image, the confidence is directly output. In this case, it may be that the front and back rows make gestures at the same time, and the front row blocks the hands of the back row; if the corresponding gesture is detected in both images, the detection accuracy is improved by confidence weighting. That is, the confidence of the gesture in the mirror reflection image and the front camera image is calculated, and the average is taken as the final confidence. If the confidence of the detected gesture exceeds the set threshold, such as 0.8, the gesture detection result is output.

[0135] When gesture recognition is output, the car computer functions are linked.

[0136] In addition, it is worth noting that in the existing method, the hand is facing the front camera when making gestures, while in the present invention, the hand is required to face the mirror below, so that the mirror can clearly capture the real-time picture.

[0137] The technical solution of this embodiment is a method combining software and hardware, including both hardware innovation and software algorithm innovation. At present, the existing system is prone to misidentifying the unconscious gestures of passengers as specific gestures, resulting in false detection problems. The gesture recognition method using the mirror material of the cabin bottom, that is, a technical method for obtaining reflected images based on the mirror material of the cabin bottom for gesture recognition, reduces false detection of unconscious gestures by using mirror images as the main and camera images as the auxiliary.

[0138] It is difficult for a single camera system to fully capture the gestures of the front and rear passengers, resulting in blind spots and limited viewing angles. Multi-view gesture detection, that is, a combination of the cabin mirror material and the front camera, is used to accurately distinguish the gestures of the front and rear passengers using the multi-view information provided by the cabin mirror material. The mirror material provides an additional viewing angle, overcoming the viewing angle limitation of the single camera system and expanding the viewing angle.

[0139] The existing gesture recognition system has low accuracy in complex in-vehicle environments and is easily affected by occlusion and light changes, resulting in recognition accuracy problems. The accuracy and robustness of gesture recognition are improved through gesture confidence weighted calculation, that is, the confidence weighted calculation of gesture recognition results is performed, and the detection results from mirror reflection and front camera are combined to improve recognition accuracy.

[0140] That is to say, the technical solution provided by the embodiment of the present invention determines whether the target gesture exists by mainly using the mirror image and supplemented by the image collected by the camera, which has the effect of falsely detecting unconscious gestures. Furthermore, the image of multi-perspective information provided by the mirror material of the cabin bottom can accurately distinguish the gestures of the front and rear passengers, achieving the effect of distinguishing gestures of multiple passengers. Furthermore, in the case of including both the mirror image and the front row image, and for the same position or user, the confidence can be weighted to improve the accuracy and robustness of gesture recognition. Of course, since the mirror material provides an additional perspective, it overcomes the perspective limitation of the single camera system and improves the accuracy of data processing.

[0141] Figure 7 : is a schematic diagram of the structure of an image processing device applied to a vehicle cabin provided by an embodiment of the present invention. This embodiment is applicable to any vehicle cabin with a gesture control function. The specific structure of the device is as follows:

[0142] The image acquisition module 710 is used to acquire at least one image to be processed of the interior of the vehicle cabin, wherein at least one image to be processed includes an image under a target acquisition perspective, and the target acquisition perspective corresponds to the perspective corresponding to the mirror material laid on the bottom of the vehicle cabin; the gesture detection module 720 is used to input at least one image to be processed into a gesture detection model, so as to output a gesture category and confidence corresponding to the target part when the target part is included in the image to be processed; the gesture determination module 730 is used to determine the target gesture corresponding to the at least one image to be processed based on the gesture category and confidence corresponding to the at least one image to be processed; the instruction sending module 740 is used to send a control instruction to the target component corresponding to the target gesture, so that the target component executes the target event based on the control instruction.

[0143] The technical solution provided by the embodiment of the present invention can collect the image to be processed under the target perspective based on the mirror material laid on the bottom of the cabin, and can collect the image to be processed in the cabin based on the camera device deployed in the cabin. The above-collected image to be processed can be input into the gesture detection model, so that when the target part is included in the image to be processed, the gesture category and confidence corresponding to the target part can be output. According to the gesture category and confidence of the image to be processed, the target gesture corresponding to the image to be processed can be determined. After determining the target gesture, a control instruction can be sent to the target part corresponding to the target gesture, so that the target component executes the target event based on the control. Since the processed image to be processed includes the image to be processed corresponding to the mirror material laid on the bottom of the cabin, and this image is mainly collected from the bottom, generally speaking, the collection range is relatively wide. When processing based on the above-mentioned image to be processed, the accuracy and effectiveness of the target part recognition can be improved.

[0144] On the basis of the above technical solution, the device also includes: a first image acquisition module, which is used to acquire a first original image of the interior of the cabin under a target acquisition angle of view based on a camera device with data acquisition and transmission functions deployed at the bottom of the cabin; wherein the camera device is associated with a mirror material laid at the bottom of the cabin; and / or, a second image acquisition module, which is used to acquire a second original image of the cabin under a first angle of view based on a camera device deployed at the top of the cabin; a module for determining an image to be processed, which is used to pre-process the first original image and the second original image respectively to obtain an image to be processed to be acquired; wherein the pre-processing includes grayscale processing of the original image and then filtering processing.

[0145] On the basis of the above technical solutions, at least one image to be processed includes two images to be processed generated at the same time, and the gesture detection model includes a part detection sub-model and a gesture detection sub-model; the gesture detection module includes: a part detection frame determination unit, which is used to process at least two images to be processed based on the part detection sub-model to output a part detection frame corresponding to the target part in each image to be processed; a confidence determination unit, which is used to input the images in the part detection frames corresponding to the two images to be processed into the gesture detection sub-model, so that the gesture detection sub-model outputs the gesture category and confidence corresponding to the target part in each image to be processed.

[0146] On the basis of the above technical solutions, the gesture detection model also includes a part matching sub-model connected to the part detection sub-model, and the output result of the gesture detection model also includes a part matching result. The device also includes a matching result output module, which is used to: input the images in the part detection boxes corresponding to the two images to be processed into the part matching sub-model, and output the part matching results of the two images to be processed.

[0147] On the basis of the above technical solutions, the at least one image to be processed is two images to be processed acquired at the same time, and the gesture detection module includes:

[0148] A first confidence determination unit, configured to perform weighted processing on the confidences of the two images to be processed to obtain a target confidence when the part matching results of the two images to be processed are preset results and the gesture categories are the same;

[0149] The target gesture determining unit is used to determine the target gesture corresponding to the gesture category when the target confidence is higher than a preset confidence threshold.

[0150] On the basis of the above technical solutions, the device further includes:

[0151] A second confidence determination unit, configured to use the confidence of the image to be processed under the target acquisition viewing angle as the target confidence when the part matching result of the two images to be processed is different from the preset result;

[0152] The target gesture determining unit is used to determine the target gesture corresponding to the gesture category when the target confidence is higher than a preset confidence threshold.

[0153] On the basis of the above technical solutions, the at least one image to be processed includes an image to be processed under the target viewing angle, and the gesture detection module further includes:

[0154] A part detection frame output unit, used for inputting the image to be processed into the gesture detection model, so as to identify a part detection frame corresponding to the target part in the image to be processed based on the part detection sub-model in the gesture detection model;

[0155] A third confidence determination unit, configured to process image recognition in the part detection frame based on a gesture detection sub-model in the gesture detection model to obtain a gesture category and a confidence corresponding to the target part;

[0156] The gesture determination module is further configured to determine a target gesture corresponding to the gesture category if the confidence level is higher than a preset confidence threshold.

[0157] The image processing device for use in a vehicle cabin provided in an embodiment of the present invention can execute the image processing method for use in a vehicle cabin provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method. For the contents not described in detail in this embodiment, reference can be made to the description in any method embodiment of the present invention.

[0158] Figure 8is a block diagram of an image processing electronic device for use in a vehicle cabin provided by an embodiment of the present invention. 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 may represent any vehicle-mounted device, and may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0159] like Figure 8 As shown, the electronic device includes: one or more processors 801, a memory 802, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are connected to each other using different buses and can be installed on a common mainboard or installed in other ways as needed. The processor can process instructions executed in the electronic device, including instructions stored in or on the memory to display graphical information of a graphical user interface (GUI) on an external input / output device (such as a display device coupled to the interface). In other embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple electronic devices can be connected, and each device provides some necessary operations, for example, as a server array, a group of blade servers, or a multi-processor system. Figure 8 A processor 801 is taken as an example.

[0160] The memory 802 is a non-transitory computer-readable storage medium provided by the present invention. The memory stores instructions executable by at least one processor, so that at least one processor executes the image processing method for vehicle cabin provided by the embodiment of the present invention. The non-transitory computer-readable storage medium of the present invention stores computer instructions, which are used to cause a computer to execute the image processing method for vehicle cabin provided by the embodiment of the present invention.

[0161] The memory 802 is a non-transient computer-readable storage medium that can be used to store non-transient software programs, non-transient computer executable programs and modules, such as program instructions / modules corresponding to the image processing method applied in the vehicle cabin in the embodiment of the present invention, for example, Figure 7The image acquisition module 710, gesture detection module 720, gesture determination module 730 and instruction sending module 740 are shown. The processor 801 executes various functional applications and data processing of the server by running the non-transient software programs, instructions and modules stored in the memory 802, that is, the image processing method applied to the vehicle cabin in the above method embodiment is realized.

[0162] The memory 802 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required by at least one function; the data storage area may store data created according to the use of an electronic device for implementing the image processing method applied in the vehicle cabin in the embodiment of the present invention, etc. In addition, the memory 802 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory 802 may optionally include a memory remotely arranged relative to the processor 801, and these remote memories may be connected to the electronic device for implementing the image processing method applied in the vehicle cabin in the embodiment of the present invention via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0163] The electronic device for implementing the image processing method applied to the vehicle cabin in the embodiment of the present invention may further include: an input device 803 and an output device 804. The processor 801, the memory 802, the input device 803 and the output device 804 may be connected via a bus or other means. Figure 8 The example of connecting through bus is taken in the following.

[0164] The input device 803 can receive input digital or character information, and generate key signal input related to user settings and function control of the electronic device used to implement the image processing method applied to the vehicle cabin in the embodiment of the present invention, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick and other input devices. The output device 804 may include a display device, an auxiliary lighting device and a tactile feedback device, wherein the auxiliary lighting device is such as a light emitting diode (LED); the tactile feedback device is such as a vibration motor. The display device may include, but is not limited to, a liquid crystal display (LCD), an LED display and a plasma display. In some embodiments, the display device may be a touch screen.

[0165] Various implementations of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, application specific integrated circuits (ASICs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0166] These computer programs, also referred to as programs, software, software applications, or codes, include machine instructions for programmable processors, and may be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. As used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, device, and / or means for providing machine instructions and / or data to a programmable processor, such as a disk, an optical disk, a memory, a programmable logic device (PLD), including a machine-readable medium that receives machine instructions as machine-readable signals. The term "machine-readable signal" refers to any signal for providing machine instructions and / or data to a programmable processor.

[0167] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to a user, such as a cathode ray tube (CRT) or an LCD monitor; and a keyboard and a pointing device, such as a mouse or a trackball, through which a user can provide input to the computer. Other types of devices can also be used to provide interaction with a user; for example, the feedback provided to the user can be any form of sensory feedback, such as visual feedback, auditory feedback, or tactile feedback; and input from the user can be received in any form, including acoustic input, voice input, or tactile input.

[0168] The systems and techniques described herein may be implemented in a computing system that includes back-end components, e.g., as a data server, or in a computing system that includes middleware components, e.g., an application server, or in a computing system that includes front-end components, e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein, or in a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication, such as a communication network. Examples of communication networks include: a Local Area Network (LAN), a Wide Area Network (WAN), and the Internet.

[0169] A computer system may include clients and servers. Clients and servers are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship to each other.

[0170] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution disclosed in the present invention can be achieved, and this document does not limit this.

[0171] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. An image processing method applied in a vehicle cabin, characterized in that: include: Acquire at least one image to be processed of the interior of the vehicle cabin, wherein the at least one image to be processed includes an image under a target acquisition viewing angle, and the target acquisition viewing angle corresponds to a viewing angle corresponding to a mirror material laid on the bottom of the vehicle cabin; Inputting the at least one image to be processed into a gesture detection model, so as to output a gesture category and a confidence level corresponding to the target part when the image to be processed includes the target part; Determining a target gesture corresponding to the at least one image to be processed based on the gesture category and confidence level corresponding to the at least one image to be processed; A control instruction is sent to a target component corresponding to the target gesture, so that the target component executes a target event based on the control instruction.

2. The method according to claim 1, characterized in that Before acquiring at least one image to be processed of the interior of the vehicle cabin, the method further includes: Based on a camera device with data acquisition and transmission function deployed at the bottom of the cabin, a first original image of the interior of the cabin under a target acquisition angle of view is acquired; wherein the camera device is associated with a mirror material laid at the bottom of the cabin; and / or, collecting a second original image of the vehicle cabin at a first viewing angle based on a camera device deployed on the top of the vehicle cabin; Preprocessing the first original image and the second original image respectively to obtain images to be processed; The preprocessing includes grayscale processing of the original image and then filtering processing.

3. The method according to claim 1, characterized in that The at least one image to be processed includes two images to be processed generated at the same time, and the gesture detection model includes a part detection sub-model and a gesture detection sub-model; The step of inputting the at least one image to be processed into a gesture detection model to output a gesture category and a confidence level corresponding to the target part when the image to be processed includes the target part comprises: Processing the two images to be processed based on the part detection sub-model to output a part detection frame corresponding to the target part in each image to be processed; The images within the part detection frames corresponding to the two images to be processed are input into the gesture detection sub-model, so that the gesture detection sub-model outputs the gesture category and confidence corresponding to the target part in each image to be processed.

4. The method according to claim 3, characterized in that The gesture detection model further includes a part matching sub-model connected to the part detection sub-model, and the output result of the gesture detection model further includes a part matching result. The method further includes: The images within the part detection frames corresponding to the two images to be processed are input into the part matching sub-model, and the part matching results of the two images to be processed are output.

5. The method according to claim 4, characterized in that The at least one image to be processed is two images to be processed acquired at the same time, and determining the target gesture corresponding to the at least one image to be processed based on the gesture category and confidence corresponding to the at least one image to be processed includes: When the part matching results of the two images to be processed are preset results and the gesture categories are the same, weighting the confidences of the two images to be processed to obtain a target confidence; When the target confidence is higher than a preset confidence threshold, a target gesture corresponding to the gesture category is determined.

6. The method according to claim 5, characterized in that Also includes: When the part matching results of the two images to be processed are different from the preset results, the confidence of the image to be processed under the target acquisition viewing angle is used as the target confidence; When the target confidence is higher than a preset confidence threshold, a target gesture corresponding to the gesture category is determined.

7. The method according to claim 1, characterized in that The at least one image to be processed includes an image to be processed under a target perspective, and the at least one image to be processed is input into a gesture detection model to output a gesture category and a confidence level corresponding to the target part when the image to be processed includes the target part, including: Inputting the image to be processed into the gesture detection model to identify a part detection frame corresponding to a target part in the image to be processed based on a part detection sub-model in the gesture detection model; Based on the gesture detection sub-model in the gesture detection model, image recognition processing is performed in the part detection frame to obtain a gesture category and a confidence level corresponding to the target part; Accordingly, determining the target gesture corresponding to the at least one image to be processed based on the gesture category and confidence level corresponding to the at least one image to be processed includes: If the confidence level is higher than a preset confidence threshold, a target gesture corresponding to the gesture category is determined.

8. A device for image processing in a vehicle cabin, characterized in that: include: An image acquisition module, used to acquire at least one image to be processed of the interior of the vehicle cabin, wherein the at least one image to be processed includes an image under a target acquisition viewing angle, and the target acquisition viewing angle corresponds to a viewing angle corresponding to a mirror material laid on the bottom of the vehicle cabin; A gesture detection module, configured to input the at least one image to be processed into a gesture detection model, so as to output a gesture category and a confidence level corresponding to the target part when the image to be processed includes the target part; A gesture determination module, configured to determine a target gesture corresponding to the at least one image to be processed based on the gesture category and confidence level corresponding to the at least one image to be processed; The instruction sending module is used to send a control instruction to a target component corresponding to the target gesture, so that the target component executes a target event based on the control instruction.

9. An electronic device, characterized in that: include: at least one processor; and a memory in communication with the at least one process; The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method for image processing applied in a vehicle cabin as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to cause the computer to execute the method for image processing applied in a vehicle cabin according to any one of claims 1 to 7.

Images