An image processing method, apparatus, electronic device, and storage medium

Abstract

The application discloses an image processing method and device, electronic equipment and storage medium. It relates to the field of artificial intelligence, and the method comprises the following steps: acquiring brain electrical signals of a target object receiving image stimulation collected by a brain electrical device, wherein the brain electrical device is worn on the target object watching the image, and the image is displayed on a display device; performing a cutting operation on the brain electrical signals according to the display time of the image to obtain a plurality of target signal segments; performing waveform processing on the plurality of target signal segments respectively to obtain recognition results corresponding to the plurality of target signal segments respectively, wherein the recognition results are recognition results of the image; and determining a target recognition result of the image based on the recognition results corresponding to the plurality of target signal segments. Through the application, the problem that the accuracy of the picture classification method in the related art is low and the efficiency of the picture classification by artificial is low is solved.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence, and more specifically, to an image processing method, apparatus, electronic device, and storage medium. Background Technology

[0002] The 21st century is an era of information explosion, with life filled with various types of information, such as visual, auditory, and tactile information. According to incomplete statistics, visual information already accounts for 60% of total information. Images, as a crucial carrier of visual information, are often cluttered with useless data. Therefore, how to sift through this redundant information to extract valuable data for military and civilian use is a pressing issue. Currently, image filtering and recognition can be achieved through various methods, including computer vision and manual screening. Computer vision offers fast processing speeds but suffers from low accuracy when dealing with noisy images. Manual screening offers higher accuracy but is generally less efficient. However, the visual information processing capabilities of humans enable them to discriminate images in a very short time.

[0003] There is currently no effective solution to the problem that computer-based image classification methods have low accuracy and that accuracy also decreases over time with manual judgment. Summary of the Invention

[0004] The main objective of this application is to provide an image processing method, apparatus, electronic device, and storage medium to solve the problems of low accuracy and low efficiency in image classification methods that rely solely on computers or manual methods in the related art.

[0005] To achieve the above objectives, according to one aspect of this application, an image processing method is provided. The method includes: acquiring electroencephalogram (EEG) signals of a target object receiving image stimulation via an EEG device, wherein the EEG device is worn on the target object viewing the image, and the image is displayed on a display device; performing a segmentation operation on the EEG signals according to the display time of the image to obtain multiple target signal segments; performing waveform processing on the multiple target signal segments respectively to obtain recognition results corresponding to the multiple target signal segments, wherein the recognition results are recognition results for the image; and determining a target recognition result for the image based on the recognition results corresponding to the multiple target signal segments.

[0006] Optionally, before acquiring the EEG signals of the target object receiving image stimulation through the EEG device, the method further includes: sending the image and a display instruction to display the image to multiple display devices, wherein the multiple display devices correspond one-to-one with multiple EEG devices; and sending a signal synchronization tag to the corresponding EEG device after the display device displays the image, wherein the signal synchronization tag includes the display time of the display device displaying the image.

[0007] Optionally, acquiring the EEG signals of the target object receiving image stimulation through the EEG device includes: receiving event potential signals acquired by multiple EEG devices, wherein the event potential signals are potential signals of the event of the target object recognizing the image acquired by the EEG device; and receiving a signal synchronization tag sent by the display device to the EEG device.

[0008] Optionally, based on the display time of the image, a segmentation operation is performed on the EEG signal to obtain multiple target signal segments, including: determining the triggering time period of the event potential triggered by the recognition result corresponding to the event; determining the display time of the image based on the signal synchronization tag; and segmenting the event potential signal according to the display time of the image and the triggering time period to obtain the target signal segments.

[0009] Optionally, performing waveform processing on the plurality of target signal segments to obtain recognition results corresponding to the plurality of target signal segments includes: processing the waveforms of the plurality of target signal segments using a classification algorithm to obtain classification results corresponding to the plurality of target signal segments; and using the classification results as recognition results corresponding to the plurality of target signal segments.

[0010] Optionally, before performing waveform processing on the plurality of target signal segments to obtain recognition results corresponding to the plurality of target signal segments, the method further includes: training the classification algorithm using training images, wherein the events recognized by the training images and the events recognized by the images are of the same type; before performing a truncating operation on the EEG signal according to the display time of the image to obtain the plurality of target signal segments, the method further includes: filtering and denoising the event potential signal.

[0011] Optionally, determining the target recognition result of the image based on the recognition results corresponding to the plurality of target signal segments includes: determining the weights of the plurality of EEG devices; and determining the recognition result of the image based on the weights of the plurality of EEG devices and the recognition results of the corresponding EEG signals.

[0012] To achieve the above objectives, according to another aspect of this application, an image processing apparatus is provided. The apparatus includes: an acquisition module for acquiring electroencephalogram (EEG) signals of a target object receiving image stimulation via an EEG device, wherein the EEG device is worn on the target object viewing the image, and the image is displayed on a display device; a cropping module for performing a cropping operation on the EEG signals according to the display time of the image to obtain multiple target signal segments; a processing module for performing waveform processing on the multiple target signal segments respectively to obtain recognition results corresponding to the multiple target signal segments, wherein the recognition results are recognition results for the image; and a determination module for determining the target recognition result of the image based on the recognition results corresponding to the multiple target signal segments.

[0013] According to another aspect of this application, a computer-readable storage medium is also provided, the storage medium including a stored program, wherein the program performs the image processing method described in any one of the foregoing.

[0014] According to another aspect of this application, an electronic device is also provided, including one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the image processing method described in any one of the foregoing.

[0015] This application employs the following steps: acquiring brainwave signals from a target object receiving image stimulation via an EEG device, wherein the EEG device is worn on the target object viewing the image, and the image is displayed on a display device; performing a segmentation operation on the EEG signals based on the image display time to obtain multiple target signal segments; performing waveform processing on each of the multiple target signal segments to obtain recognition results corresponding to each of the multiple target signal segments, wherein the recognition results are the image recognition results; determining the target recognition result of the image based on the recognition results corresponding to the multiple target signal segments. By combining EEG device and manual recognition, the aim of quickly and accurately classifying target and non-target images is achieved, realizing the technical effect of improving image recognition accuracy and efficiency. This solves the problems of low accuracy in image classification methods relying solely on computers and low efficiency in image classification methods relying solely on manual methods in related technologies. Attached Figure Description

[0016] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0017] Figure 1 This is a flowchart of an image processing method provided according to an embodiment of this application;

[0018] Figure 2 This is a schematic diagram of the framework of an image recognition system based on a group brain-computer interface according to an embodiment of this application;

[0019] Figure 3 This is a schematic diagram of the audiovisual stimulus synchronization and signal acquisition synchronization principle provided in the embodiments of this application;

[0020] Figure 4 This is a flowchart of the system control terminal provided according to an embodiment of this application;

[0021] Figure 5 This is a flowchart of the visual stimulation end according to an embodiment of this application;

[0022] Figure 6 This is a distribution map of 64-channel EEG acquisition points provided according to an embodiment of this application;

[0023] Figure 7 This is a flowchart of the signal acquisition terminal provided according to an embodiment of this application;

[0024] Figure 8 This is a schematic diagram of EEG signals and synchronization tag signals provided according to embodiments of this application;

[0025] Figure 9 This is a flowchart of the data processing terminal provided according to an embodiment of this application;

[0026] Figure 10 This is a flowchart of collaborative processing of EEG signals from multiple people according to an embodiment of this application;

[0027] Figure 11 This is a schematic diagram of an image processing apparatus provided according to an embodiment of this application;

[0028] Figure 12 This is a schematic diagram of an electronic device provided according to an embodiment of this application. Detailed Implementation

[0029] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

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

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

[0032] It should be noted that all information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data used for analysis, etc.) involved in this disclosure are information and data authorized by the user or fully authorized by all parties. For example, this system has an interface with relevant users or organizations. Before obtaining relevant information, it is necessary to send an acquisition request to the aforementioned user or organization through the interface, and obtain the relevant information after receiving consent information from the aforementioned user or organization.

[0033] The present invention will now be described in conjunction with preferred implementation steps.

[0034] To explain, Figure 1 This is a flowchart of an image processing method provided according to an embodiment of this application, such as... Figure 1 As shown, the method includes the following steps:

[0035] Step S101: Acquire the brain signals of the target object receiving image stimulation by the brain electrical device, wherein the brain electrical device is worn on the target object viewing the image and the image is displayed on the display device.

[0036] Step S102: Based on the display time of the image, perform a segmentation operation on the EEG signal to obtain multiple target signal segments;

[0037] Step S103: Perform waveform processing on multiple target signal segments respectively to obtain recognition results corresponding to the multiple target signal segments respectively, wherein the recognition results are the recognition results of the image;

[0038] Step S104: Determine the target recognition result of the image based on the recognition results corresponding to multiple target signal segments.

[0039] Using the above steps, EEG signals from a target object receiving image stimulation are acquired via an EEG device worn on the target object viewing the image, which is then displayed on a display device. Based on the image display time, the EEG signals are truncated to obtain multiple target signal segments. Waveform processing is then performed on each of these target signal segments to obtain corresponding recognition results, which are the image recognition results. Based on the recognition results corresponding to the multiple target signal segments, the target recognition result of the image is determined. By combining EEG device and manual recognition, the goal of quickly and accurately classifying target and non-target images is achieved, improving image recognition accuracy and efficiency. This solves the problems of low accuracy with computer-based image classification and low efficiency with manual image classification in related technologies.

[0040] The target object can view images through the aforementioned display device. These images serve as visual stimuli. When the target object views these images, based on pre-known recognition information, its brain generates corresponding electroencephalogram (EEG) signals, triggering an event potential in the brain corresponding to the recognition event. In other words, the target object can generate a corresponding event potential signal in response to the event of recognizing the image. For example, when identifying an image containing a car from 10 images (which may or may not contain cars), if the target object, with the recognition information for recognizing cars, views an image containing a car, its brain will generate a certain potential change at the corresponding location, triggering the corresponding event potential. By monitoring whether this event potential occurs, it can be determined whether the target object, when viewing a particular image, determines that the image contains a car.

[0041] The aforementioned images can be displayed to multiple target objects through multiple display devices, or to multiple target objects through a single display device. Since each display device records the display time of an image—that is, the time when the image stimulates the target object—this event allows the identification of a potential event band within the target object's EEG signal. By checking whether an event potential is generated within that band, the target object's recognition result of the image can be determined. Therefore, in this embodiment, the correspondence between display devices and target objects can be such that one display device is simultaneously viewed by multiple target objects, the same display device is viewed sequentially by multiple target objects, multiple display devices are simultaneously viewed by multiple target objects, one target object views one display device, or multiple display devices are viewed by multiple target objects at different times. This reduces the limitations of target objects viewing display devices, eliminating the need for viewing on a specific display device at a specific time to be effective, greatly reducing the spatial and temporal limitations of image recognition.

[0042] When different target objects view a display device, multiple images on the display device can be displayed sequentially after being shuffled. Since the display time of each image is recorded, there is no need to set the display order of multiple images. Moreover, shuffling the order can avoid the brain's mental inertia caused by a fixed order, which would reduce the accuracy of recognition.

[0043] When displaying the aforementioned multiple images, the interval between their display must not exceed the target subject's brain's reaction time to the recognition event. For example, if the brain's reaction time to image stimuli is 2 seconds, then the time interval between the display of the multiple images must not exceed this reaction time. This prevents the images from switching too quickly, making it difficult for the target subject's brain to react and recognize them. The aforementioned display frequency setting can meet the target subject's reaction time requirements and ensure the accuracy of the target subject's image recognition.

[0044] The aforementioned EEG device is worn on the target object viewing the image. Multiple display devices can correspond one-to-one with multiple target objects, and multiple target objects can also correspond one-to-one with multiple EEG devices. The EEG device can collect the EEG signals generated when the target object receives image stimulation; that is, the target object can generate event potential signals to recognize the image when viewing the image. The EEG device can synchronize the EEG signals with the corresponding images based on the display time of each image display device. The event potential signals collected by the EEG device can be preprocessed with noise reduction and filtering to ensure the clarity of the collected EEG signals, facilitating the analysis and processing of the EEG signals. By combining the aforementioned EEG device with the target object viewing the image, the traditional, less efficient manual image recognition method can be eliminated, and the traditional brain-computer interface technology with its low accuracy can be avoided.

[0045] The aforementioned target signal segment can refer to the event potential signal segment of the image recognition event generated by the visual stimulation of the target object. Based on the display time of the image, the EEG signal is truncated to obtain multiple target signal segments. The waveforms of multiple target signal segments can be processed, calculated, and image discriminant analysis can be performed to obtain the target recognition result of the image. The recognition result of the image can be calculated based on the weights of multiple EEG devices for multiple target objects and the corresponding target recognition result, which can ensure that the obtained image recognition result is accurate and reliable.

[0046] By combining EEG devices with human recognition, the goal of quickly and accurately classifying target and non-target images is achieved, thereby improving the accuracy and efficiency of image recognition. This solves the problems of low accuracy when using computers alone for image classification and low efficiency when using humans alone for image classification.

[0047] Optionally, before acquiring the EEG signals of the target object receiving image stimulation through the EEG device, the method further includes: sending images and display instructions for displaying the images to multiple display devices, wherein the multiple display devices correspond one-to-one with multiple EEG devices; and sending a signal synchronization tag to the corresponding EEG device after the display device displays the image, wherein the signal synchronization tag includes the display time of the image displayed by the display device.

[0048] Before collecting EEG data using EEG devices, images and display instructions can be sent to multiple display devices via a control terminal. The control terminal can control the overall operation of the system, and the display devices can receive the image display instructions sent by the control terminal. The display instructions can include the method of displaying the images on the display devices, which can be to display the images at a certain frequency or in a certain order. The display devices can use the images sent by the control terminal as a visual stimulus. Multiple display devices can correspond one-to-one with multiple EEG devices. The display devices can also send the corresponding display time information of the displayed images as a system synchronization tag to the corresponding EEG devices, which can ensure that there is a one-to-one correspondence between the visual stimulus generated by the display devices and the signal generated by the visual stimulus received by the EEG devices.

[0049] By using multiple display devices and corresponding EEG devices, and employing signal tags for signal synchronization, the target object and the EEG devices are combined, thereby achieving the technical effect of improving image recognition speed.

[0050] Optionally, acquiring the EEG signals of the target object receiving image stimulation through the EEG device includes: receiving event potential signals acquired by multiple EEG devices, wherein the event potential signals are the potential signals of events in which the EEG device acquires the target object's image recognition; and receiving a signal synchronization tag sent by the display device to the EEG device.

[0051] The aforementioned EEG devices can acquire image information displayed by a display device and generate EEG signals. The system control terminal can receive event potential signals of the target object in response to the recognition image acquired by multiple EEG devices. These event potential signals of the recognition image can be EEG signals. Using multiple EEG devices to acquire EEG signals of recognition images of multiple target objects can ensure the accuracy of the recognition results.

[0052] The target object can be an object that can recognize images and generate corresponding EEG signal responses. The image to be recognized can be a certain type of image or an image with certain features. The EEG device can collect the event potential signal generated when the target object recognizes the image. The EEG device can synchronize the EEG signal collected by the EEG device with the corresponding recognition image according to the synchronization tag sent when the display device displays the image, so as to ensure that the visual stimulus generated by the image corresponds one-to-one with the corresponding EEG signal.

[0053] The aforementioned image recognition event is determined based on the nature of the image being recognized. For example, the aforementioned time can be a low-probability target event potential. Specifically, it can be that 100 images are recognized, and 2 images that meet the target conditions are selected. When viewing each image, the event of an image that meets the target conditions is a low-probability target event with a probability less than a preset probability. The brain's response to this low-probability target event is at a specific point in a specific location in the brain. By detecting whether the target signal segment of the target object generates a specific potential at a specific location, it is determined whether the image recognized by the target object triggers the event of an image that meets the target conditions, thereby determining whether the image recognized by the target object is an image that meets the target conditions.

[0054] By collecting and recognizing event potential signals from image events using multiple EEG devices, the goal of image recognition using EEG devices was achieved, thus realizing the technical effect of improving image recognition efficiency.

[0055] Optionally, based on the display time of the image, the EEG signal is truncated to obtain multiple target signal segments, including: determining the triggering time period of the event potential triggered by the recognition result corresponding to the event; determining the image display time based on the signal synchronization tag; and truncating the event potential signal according to the image display time and the triggering time period to obtain the target signal segments.

[0056] Based on the display time of the image, the corresponding EEG signal generated by the displayed image can be extracted. The display time of the image can be determined based on the signal synchronization tag. Based on the display time of the image and the display frequency of the image, the signal segment of the event potential signal can be extracted to obtain the target signal segment triggered by the recognition image.

[0057] By capturing the EEG signal during the image display time, the goal of obtaining image recognition signal segments is achieved, thereby increasing the speed of signal processing for recognition images.

[0058] Optionally, waveform processing is performed on multiple target signal segments to obtain recognition results corresponding to the multiple target signal segments, including: processing the waveforms of multiple target signal segments using a classification algorithm to obtain classification results corresponding to the multiple target signals; and using the classification results as recognition results corresponding to the multiple target signal segments.

[0059] The target signal segments determined based on the synchronization tags can be analyzed and processed to obtain the image recognition results of the target object. Multiple target signal segments generated by image stimuli can be processed by a classification algorithm to obtain recognition results corresponding to the multiple target signal segments. The aforementioned algorithms can be used to extract features from EEG signals for EEG signal research, or to discriminate signals based on the extracted features to obtain the classification results corresponding to the target signals. The classification results can be used as the recognition results corresponding to the target signal segments.

[0060] Specifically, the EEG signals of the target subjects within the target group are preprocessed. Features are then extracted from the preprocessed signals. Based on research on low-probability target event potentials induced by the odd-ball paradigm, the extracted features primarily focus on the period within 400 ms after event stimulation. Feature extraction can utilize the corresponding EEG set from 0-400 ms after the preprocessed image presentation as features. The signals can then be discriminated based on these features to obtain the discrimination results. Linear Discriminant Analysis (LDA) and Logistic Regression (LR) can be used for classification.

[0061] By processing the waveform of the target signal segment, the goal of obtaining the recognition result of the target signal segment is achieved, thereby improving the technical effect of improving the accuracy of the recognition result.

[0062] Optionally, before performing waveform processing on multiple target signal segments to obtain recognition results corresponding to the multiple target signal segments, the method further includes: training the classification algorithm using training images, wherein the events recognized by the training images and the events recognized by the images are of the same type; before performing a truncating operation on the EEG signal according to the display time of the image to obtain multiple target signal segments, the method further includes: filtering and denoising the event potential signal.

[0063] When running an image recognition system, an experiment can be conducted to collect sets of EEG signals induced by target images and non-target images of multiple target objects. The EEG signal sets obtained from the experiment can be used to train the classification algorithm. A suitable EEG signal set can be selected from the EEG signals generated by the recognition events of the target images during training as a model of the image recognition system. The events that are recognized by the training images and the events that are recognized by the images are of the same type of events. The event potential signals of the events that are recognized by the training images and the event potential signals of the events that are recognized by the images can have similar or identical event potential signals of the same type.

[0064] Based on the image display time, before extracting multiple target signal segments from the EEG signal, filtering and noise reduction processing can be performed on the event potential signal. Filtering the EEG signal segments can remove noise mixed in with the EEG signal, making the image recognition and analysis results of the extracted target signal segments more accurate. For example, the EEG signals of individual members in the target group can be filtered independently to remove noise mixed in with the EEG signal, such as eye point signals, electromyography signals, power frequency noise, etc., as well as 250Hz downsampling. FIR filtering can be used.

[0065] By training the image classification algorithm, the goal of obtaining an image classification model was achieved, thereby improving the accuracy of image recognition in the system.

[0066] Optionally, determining the target recognition result of the image based on the recognition results corresponding to multiple target signal segments includes: determining the weights of multiple EEG devices; and determining the recognition result of the image based on the weights of the multiple EEG devices and the corresponding EEG signal recognition results.

[0067] The target recognition result of an image can be determined based on the recognition results corresponding to multiple target signal segments. The target recognition result of an image can be determined by setting the weights of multiple EEG devices. The final target recognition result of the image can be obtained by calculating the target recognition results of multiple EEG devices and their corresponding weights. The weights of multiple EEG devices can be determined based on the accuracy of the target object wearing the EEG device in recognizing the image, thereby increasing the accuracy of the recognition result.

[0068] By assigning weights to EEG devices to participate in the determination of image recognition results, the accuracy of image recognition is increased, thus achieving the technical effect of improving the accuracy rate of image recognition.

[0069] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0070] It should be noted that this embodiment also provides an optional implementation method, which will be described in detail below.

[0071] Brain-computer interfaces (BCIs) are technologies that allow the human brain to interact with the outside world, bypassing peripheral muscles and nerves. The brain generates corresponding electroencephalograms (EEGs) in response to different types of stimuli designed in time and space (such as visual, auditory, and olfactory stimuli). By analyzing and processing these EEG signals using signal processing methods, the information the brain wants to express can be obtained and transformed into other types of instructions (such as instructions to control a wheelchair or robotic arm, or image discrimination results instructions within the system). Traditional odd-ball BCI systems collect and analyze the EEG signals of a single individual to complete a series of discrimination tasks. However, with interference from environmental noise and factors such as attention deficit and mental fatigue caused by prolonged work, the system's discrimination performance deteriorates.

[0072] This embodiment provides an image classification system and method based on a high-precision synchronous group brain-computer interface. Brain-computer interface technology has the ability to analyze high-speed brain activity in real time. This embodiment utilizes brain-computer interface technology to effectively combine human visual processing capabilities with computer data processing capabilities to achieve operations such as image classification and filtering.

[0073] The aforementioned group brain-computer interface requires that multiple audiovisual events used to elicit EEG signals from multiple members within the group occur synchronously. The EEG signals of multiple members stimulated by these events must be collected synchronously. Machine learning methods are then used to collaboratively process and classify the group EEG signals corresponding to each image, thereby determining whether the image is a target image. Before using the system, multiple members within the group are informed which image category belongs to the target image. When performing the image classification task, all members must concentrate on determining whether the image displayed on the screen is a target image. If a member subjectively judges it to be a target image, they silently count the number of target images in their mind.

[0074] The following description, in conjunction with the illustrations, illustrates this implementation method.

[0075] Figure 2 This is a schematic diagram of the framework of an image recognition system based on a swarm brain-computer interface according to an embodiment of this application, as shown below. Figure 2As shown, the system mainly consists of four parts: (1) System control terminal: which is equivalent to the central control console of the overall system operation, controls when the visual stimulation terminal starts visual stimulation, receives the processing results sent by the data processing terminal, and then presents the classification results through the visual stimulation terminal, so that the whole system forms a closed loop. (2) Visual stimulation presentation terminal: the visual stimulation of this system adopts the Rapid Series Visual Presentation (RSVP) paradigm to induce event-related potentials of each member in the group, or more accurately, low-probability target event potentials. (3) EEG acquisition terminal: the EEG amplifier is used to collect the response of the cerebral cortex potential generated by visual stimulation in real time, and the signal is sent to the data processing terminal through the network protocol. (4) Data processing terminal: the EEG signals of multiple people in the group that are collected here through the network protocol are processed collaboratively, and the processing results are sent to the system control terminal through the network protocol.

[0076] For image recognition systems using group brain-computer interfaces, the core is to achieve synchronous occurrence of visual stimuli and synchronous acquisition of EEG signals among all members of the group. Figure 3 The schematic diagram of audiovisual stimulus synchronization and signal acquisition synchronization provided in the embodiments of this application is as follows: Figure 3 As shown, this embodiment presents a high-precision audiovisual event synchronization presentation and multi-member EEG signal synchronization acquisition technology solution: A single workstation renders a large visual stimulus interface, which is composed of the individual visual stimulus interfaces of each member within the group. Through screen expansion, each member's portion of the visual stimulus interface is displayed on their respective monitor. To ensure synchronized EEG signal acquisition among group members, while multiple visual stimuli of the same image are presented on the same workstation via expansion, a parallel port multi-split technology is used to send signal synchronization tags to multiple independent EEG acquisition devices. This ensures synchronization of both visual stimulation and signal acquisition, which is the foundation of a group brain-computer interface image recognition system.

[0077] The following sections will describe the system control terminal, visual stimulus presentation terminal, EEG acquisition terminal, and data processing terminal.

[0078] System control terminal: The system control terminal mainly controls the overall operation sequence of the system, including communication with the visual stimulus presentation terminal, communication with multiple signal acquisition terminals, and communication with the data processing terminal. Figure 4 The flowchart of the system control terminal provided in the embodiments of this application is as follows: Figure 4As shown, this task assumes there are 10 batches of images to be classified, with 100 images in each batch. Communication with the visual stimulus terminal involves sending the images to be classified from the upstream source to the visual stimulus presentation terminal, and controlling when the visual stimulus presentation terminal starts and stops presenting the visual stimulus. Communication with multiple signal acquisition terminals controls when to start and stop signal acquisition based on the presentation status of the visual stimulus terminal, ensuring the integrity of the acquired signals. Communication with the data processing terminal is to obtain the classification results for each image, ultimately completing the task of binary classification of target and non-target images in the image sequence.

[0079] Visual Stimulation Terminal: The visual stimulation terminal primarily presents the visual stimulation interface to multiple members within the group. The start and stop of stimulation for each batch of images are determined by the system control terminal. The presentation method used for each batch is a rapid sequence visual presentation paradigm, which presents images at a fixed frequency (within a certain range to ensure that members can recognize and distinguish whether the current image is a target image while looking at the screen; the frequency can be selected between 6-12Hz). The images contain two types of images: target images (e.g., images containing cars) and non-target images (e.g., images that do not contain cars). The frequency of target images is much lower than that of non-target images. Under this model, a low-probability target event potential can be induced in the brain, thereby achieving the classification of target and non-target. Figure 5 The flowchart of the visual stimulation end according to the embodiments of this application is as follows: Figure 5 As shown, in a set of images, a synchronization tag is sent to multiple EEG acquisition terminals for each image presented, and each member of the group looks at the monitor in front of them. This ensures that the EEG signals of multiple members in the group are synchronized when each image is presented.

[0080] Signal acquisition end: The signal acquisition end uses a 64-channel EEG cap ( Figure 6 The distribution map of 64-channel EEG acquisition points provided in the embodiments of this application is as follows: Figure 6 (As shown) It is used in conjunction with some commercially available EEG signal amplifiers such as the Neuroscan device, which has a sampling rate of up to 1000Hz, to collect EEG signals under visual stimulation. Figure 7 The flowchart of the signal acquisition terminal provided in the embodiments of this application is as follows: Figure 7 As shown, the system control unit determines when the signal acquisition device starts and stops acquiring data. One signal acquisition unit is responsible for acquiring the EEG signal of one member of the group, amplifying, filtering, and performing analog-to-digital conversion on the acquired EEG data. Then, based on the synchronization tags sent by the visual stimulation unit when each image is presented, it extracts the EEG data corresponding to each image from the continuous, uninterrupted EEG signal. Figure 8 This is a schematic diagram of EEG signals and synchronization tag signals provided according to embodiments of this application, such as... Figure 8 As shown, the bottom column contains the signal synchronization tags sent from the visual stimulation end via the parallel port extension. When an image is first presented, the synchronization tag corresponding to that image is sent to the acquisition end via the parallel port. The EEG signals of 64 channels within the time period from 0.1 seconds before the image is presented to 1 second after the image is presented are selected as the EEG signals evoked by that image. Then, the EEG signal set corresponding to the batch of 100 images is sent to the data processing end via the network protocol.

[0081] Data processing end: Collaboratively process the EEG signals of multiple people under synchronized visual stimulation, and finally classify the images into target and non-target images, and send them to the system control end for statistics and summary through network protocol. Figure 9 The flowchart of the data processing terminal provided in the embodiments of this application is as follows: Figure 9 As shown, the process mainly includes collecting group EEG data, processing the group EEG data, and sending the processing results. The collaborative processing of group EEG data is as follows, taking the EEG data of each member in a group induced by a certain image as an example:

[0082] (1) Preprocessing of the EEG signals of each member in the group. This includes independent filtering of the EEG signals of each member in the group to remove noise mixed in the EEG signals, such as eye point signals, electromyography signals, power frequency noise, etc., and downsampling to 250Hz. The FIR (Finite Impulse Response) filter is selected in this system.

[0083] (2) Feature extraction from preprocessed signals. Based on research on low-probability target event potentials induced by the odd-ball paradigm, features are mainly concentrated within 400ms after event stimulation. This system selects the corresponding EEG set from 0-400ms after the preprocessed image presentation as features for feature extraction.

[0084] (3) The signal is discriminated based on its features, and the discrimination result is obtained. This system uses Linear Discriminant Analysis (LDA) and Logistic Regression (LR) for classification.

[0085] (4) Weighted decision-making yields the final judgment result. Figure 10 This is a flowchart of collaborative processing of group multi-person EEG signals provided in the embodiments of this application. The collaborative processing flow of group multi-person EEG signals is as follows: Figure 10As shown, the EEG signals of each member under synchronous visual stimulation have been preprocessed, feature extracted, and pattern discriminated to obtain the discrimination results. Then, the discrimination results of each member are weighted to obtain the final group discrimination result.

[0086] The usage method of this image classification system based on high-definition synchronization group brain-computer interface is as follows:

[0087] (1) All members of the group wear EEG caps to ensure that the impedance between the EEG cap and the scalp of each lead is less than 15 kΩ, thereby ensuring the accuracy of the data.

[0088] (2) Inform all members of the group which type of image in the image set is the target image. If they realize that the image on the screen is the target image, they should silently count the number of target images in their minds.

[0089] (3) Run the system for the first round of experiments, and collect the EEG signal sets induced by target images and non-target images of all members in the group. Use the EEG signal sets obtained from the experiment to train the classification algorithm linear discriminant analysis and logistic regression model, and select the model that is more suitable for this image recognition system.

[0090] (4) Once the models of the classification algorithms corresponding to all members in the group are trained, the system is run to perform more accurate image classification tasks by utilizing the cooperation between the groups.

[0091] This embodiment relates to the fields of artificial intelligence, brain science, and computer technology. In particular, it relates to an image recognition system based on a high-precision synchronized group brain-computer interface, which utilizes collaboration among groups to achieve more accurate classification of target and non-target images.

[0092] The key to this embodiment lies in the fact that, compared with the prior art, the beneficial effects of this invention are as follows: This invention utilizes the novel brain-computer interface technology to classify target images and non-target images in a large image set; this embodiment proposes a high-precision synchronous (audiovisual event synchronization, signal acquisition synchronization) brain-computer interface system, and based on this system, improves the more accurate classification of target images and non-target images through collaborative cooperation among groups.

[0093] This application also provides an image processing apparatus. It should be noted that the image processing apparatus of this application can be used to execute the image processing method provided in this application. The image processing apparatus provided in this application will be described below.

[0094] Figure 11 This is a schematic diagram of an image processing apparatus according to an embodiment of this application. Figure 11As shown, the device includes: a data acquisition module 110, an interception module 112, a processing module 114, and a determination module 116. The device will be described in detail below.

[0095] Acquisition module 110 acquires brainwave signals from a target object receiving image stimulation via an EEG device, wherein the EEG device is worn on the target object viewing the image, and the image is displayed on a display device; interception module 112, connected to acquisition module 110, performs interception operations on the EEG signals according to the display time of the image to obtain multiple target signal segments; processing module 114, connected to interception module 112, performs waveform processing on the multiple target signal segments respectively to obtain recognition results corresponding to the multiple target signal segments respectively, wherein the recognition results are recognition results of the image; determination module 116, connected to processing module 114, determines the target recognition result of the image based on the recognition results corresponding to the multiple target signal segments.

[0096] The aforementioned device acquires brainwave signals from a target object receiving image stimulation via an EEG device (EEG device), where the EEG device is worn on the target object viewing the image, and the image is displayed on a display device. A segmentation module 112 performs segmentation on the EEG signals based on the image display time, obtaining multiple target signal segments. A processing module performs waveform processing on each of the multiple target signal segments to obtain recognition results corresponding to each segment, where the recognition results are for the image. A determination module 118 determines the target recognition result of the image based on the recognition results corresponding to the multiple target signal segments. By combining EEG device and manual recognition, the device achieves rapid and accurate classification of target and non-target images, improving image recognition accuracy and efficiency. This solves the problems of low accuracy with computer-based image classification and low efficiency with manual image classification in related technologies.

[0097] Optionally, before acquiring the EEG signals of the target object receiving image stimulation through the EEG device, the method further includes: sending images and display instructions to multiple display devices, wherein the multiple display devices correspond one-to-one with multiple EEG devices; and sending a signal synchronization tag to the corresponding EEG device after the display device displays the image, wherein the signal synchronization tag includes the display time of the image displayed by the display device.

[0098] Optionally, acquiring the EEG signals of the target object receiving image stimulation through the EEG device includes: receiving event potential signals acquired by multiple EEG devices, wherein the event potential signals are the potential signals of events in which the EEG device acquires the target object's image recognition; and receiving a signal synchronization tag sent by the display device to the EEG device.

[0099] Optionally, based on the display time of the image, the EEG signal is truncated to obtain multiple target signal segments, including: determining the triggering time period of the event potential triggered by the recognition result corresponding to the event; determining the image display time based on the signal synchronization tag; and truncating the event potential signal according to the image display time and the triggering time period to obtain the target signal segments.

[0100] Optionally, waveform processing is performed on multiple target signal segments to obtain recognition results corresponding to the multiple target signal segments, including: processing the waveforms of multiple target signal segments using a classification algorithm to obtain classification results corresponding to the multiple target signals; and using the classification results as recognition results corresponding to the multiple target signal segments.

[0101] Optionally, before performing waveform processing on multiple target signal segments to obtain recognition results corresponding to the multiple target signal segments, the method further includes: training the classification algorithm using training images, wherein the events recognized by the training images and the events recognized by the images are of the same type; before performing a truncating operation on the EEG signal according to the display time of the image to obtain multiple target signal segments, the method further includes: filtering and denoising the event potential signal.

[0102] Optionally, determining the target recognition result of the image based on the recognition results corresponding to multiple target signal segments includes: determining the weights of multiple EEG devices; and determining the recognition result of the image based on the weights of the multiple EEG devices and the corresponding EEG signal recognition results.

[0103] The image processing apparatus provided in this application embodiment acquires brain signals from a target object receiving image stimulation via an EEG device through an acquisition module 110. The EEG device is worn on the target object viewing the image, and the image is displayed on a display device. The interception module 112 performs an interception operation on the EEG signal according to the display time of the image to obtain multiple target signal segments. The processing module performs waveform processing on the multiple target signal segments to obtain recognition results corresponding to the multiple target signal segments, wherein the recognition results are the recognition results of the image. The determination module 118 determines the target recognition result of the image based on the recognition results corresponding to the multiple target signal segments. By combining EEG device and manual recognition, the apparatus achieves the purpose of quickly and accurately classifying target images and non-target images, thereby improving the accuracy and efficiency of image recognition. This solves the problems of low accuracy in image classification methods relying solely on computers and low efficiency in image classification relying solely on manual methods in related technologies.

[0104] The image processing device includes a processor and a memory. The acquisition module 110, the cropping module 112, the processing module 114, the determination module 118, etc., are all stored in the memory as program units. The processor executes the program units stored in the memory to realize the corresponding functions.

[0105] The processor contains a kernel, which retrieves the corresponding program units from memory. One or more kernels can be configured, and adjusting kernel parameters can address the issues of low accuracy in computer-based image classification and low efficiency in manual image classification.

[0106] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0107] This invention provides a computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the image processing method.

[0108] This invention provides a processor for running a program, wherein the program executes the image processing method during runtime.

[0109] like Figure 12As shown, this application embodiment provides an electronic device, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs the following steps: acquiring brainwave signals of a target object receiving image stimulation through a brainwave device, wherein the brainwave device is worn on the target object viewing the image, and the image is displayed on a display device; performing a segmentation operation on the brainwave signals according to the display time of the image to obtain multiple target signal segments; performing waveform processing on the multiple target signal segments respectively to obtain recognition results corresponding to the multiple target signal segments respectively, wherein the recognition results are the recognition results of the image; and determining the target recognition result of the image based on the recognition results corresponding to the multiple target signal segments.

[0110] Optionally, before acquiring the EEG signals of the target object receiving image stimulation through the EEG device, the method further includes: sending images and display instructions for displaying the images to multiple display devices, wherein the multiple display devices correspond one-to-one with multiple EEG devices; and sending a signal synchronization tag to the corresponding EEG device after the display device displays the image, wherein the signal synchronization tag includes the display time of the image displayed by the display device.

[0111] Optionally, acquiring the EEG signals of the target object receiving image stimulation through the EEG device includes: receiving event potential signals acquired by multiple EEG devices, wherein the event potential signals are the potential signals of events in which the EEG device acquires the target object's image recognition; and receiving a signal synchronization tag sent by the display device to the EEG device.

[0112] Optionally, based on the display time of the image, the EEG signal is truncated to obtain multiple target signal segments, including: determining the triggering time period of the event potential triggered by the recognition result corresponding to the event; determining the image display time based on the signal synchronization tag; and truncating the event potential signal according to the image display time and the triggering time period to obtain the target signal segments.

[0113] Optionally, waveform processing is performed on multiple target signal segments to obtain recognition results corresponding to the multiple target signal segments, including: processing the waveforms of multiple target signal segments using a classification algorithm to obtain classification results corresponding to the multiple target signal segments; and using the classification results as the recognition results corresponding to the target signal segments.

[0114] Optionally, before performing waveform processing on multiple target signal segments to obtain recognition results corresponding to the multiple target signal segments, the method further includes: training the classification algorithm using training images, wherein the events recognized by the training images and the events recognized by the images are of the same type; before performing a truncating operation on the EEG signal according to the display time of the image to obtain multiple target signal segments, the method further includes: filtering and denoising the event potential signal.

[0115] Optionally, determining the target recognition result of the image based on the recognition results corresponding to multiple target signal segments includes: determining the weights of multiple EEG devices; and determining the recognition result of the image based on the weights of the multiple EEG devices and the corresponding EEG signal recognition results.

[0116] The devices mentioned in this article can be servers, PCs, tablets, mobile phones, etc.

[0117] This application also provides a computer program product, which, when executed on a data processing device, is suitable for executing an initialization program having the following method steps: acquiring brainwave signals of a target object receiving image stimulation through an EEG device, wherein the EEG device is worn on the target object viewing the image, and the image is displayed on a display device; performing a segmentation operation on the EEG signals according to the display time of the image to obtain multiple target signal segments; performing waveform processing on the multiple target signal segments respectively to obtain recognition results corresponding to the multiple target signal segments respectively, wherein the recognition results are recognition results of the image; and determining the target recognition result of the image based on the recognition results corresponding to the multiple target signal segments.

[0118] Optionally, before acquiring the EEG signals of the target object receiving image stimulation through the EEG device, the method further includes: sending images and display instructions for displaying the images to multiple display devices, wherein the multiple display devices correspond one-to-one with multiple EEG devices; and sending a signal synchronization tag to the corresponding EEG device after the display device displays the image, wherein the signal synchronization tag includes the display time of the image displayed by the display device.

[0119] Optionally, acquiring the EEG signals of the target object receiving image stimulation through the EEG device includes: receiving event potential signals acquired by multiple EEG devices, wherein the event potential signals are the potential signals of events in which the EEG device acquires the target object's image recognition; and receiving a signal synchronization tag sent by the display device to the EEG device.

[0120] Optionally, based on the display time of the image, the EEG signal is truncated to obtain multiple target signal segments, including: determining the triggering time period of the event potential triggered by the recognition result corresponding to the event; determining the image display time based on the signal synchronization tag; and truncating the event potential signal according to the image display time and the triggering time period to obtain the target signal segments.

[0121] Optionally, waveform processing is performed on multiple target signal segments to obtain recognition results corresponding to the multiple target signal segments, including: processing the waveforms of multiple target signal segments using a classification algorithm to obtain classification results corresponding to the multiple target signal segments; and using the classification results as the recognition results corresponding to the target signal segments.

[0122] Optionally, before performing waveform processing on multiple target signal segments to obtain recognition results corresponding to the multiple target signal segments, the method further includes: training the classification algorithm using training images, wherein the events recognized by the training images and the events recognized by the images are of the same type; before performing a truncating operation on the EEG signal according to the display time of the image to obtain multiple target signal segments, the method further includes: filtering and denoising the event potential signal.

[0123] Optionally, determining the target recognition result of the image based on the recognition results corresponding to multiple target signal segments includes: determining the weights of multiple EEG devices; and determining the recognition result of the image based on the weights of the multiple EEG devices and the corresponding EEG signal recognition results.

[0124] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0125] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0126] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0127] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0128] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0129] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0130] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0131] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0132] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0133] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. An image processing method, characterized in that, include: Acquiring brainwave signals from a target object receiving image stimulation via a brainwave device includes: receiving event potential signals acquired by multiple brainwave devices, wherein the event potential signals are potential signals acquired by the brainwave devices of the target object recognizing the image; receiving a signal synchronization tag sent by a display device to the brainwave device; wherein the brainwave device is worn on the target object viewing the image, and the image is displayed on the display device; Based on the display time of the image, a segmentation operation is performed on the EEG signal to obtain multiple target signal segments, including: determining the triggering time period of the event potential triggered by the recognition result corresponding to the event; determining the display time of the image based on the signal synchronization tag; and segmenting the event potential signal according to the display time of the image and the triggering time period to obtain the target signal segments. Waveform processing is performed on the plurality of target signal segments respectively to obtain recognition results corresponding to the plurality of target signal segments respectively, wherein the recognition results are the recognition results of the image, which are obtained by classification through linear discriminant analysis and logistic regression; The target recognition result of the image is determined based on the recognition results corresponding to the multiple target signal segments.

2. The method according to claim 1, characterized in that, Before acquiring the electroencephalogram (EEG) signals of a target object receiving image stimulation via an EEG device, the method further includes: Send the image and display instructions to display the image to multiple display devices, wherein the multiple display devices correspond one-to-one with multiple EEG devices; After displaying the image, the display device sends a signal synchronization tag to the corresponding EEG device, wherein the signal synchronization tag includes the display time of the image displayed by the display device.

3. The method according to claim 1, characterized in that, Waveform processing is performed on the plurality of target signal segments to obtain recognition results corresponding to the plurality of target signal segments, including: The waveforms of the multiple target signal segments are processed by a classification algorithm to obtain the classification results corresponding to the multiple target signal segments; The classification result is used as the identification result corresponding to the multiple target signal segments.

4. The method according to claim 3, characterized in that, Before performing waveform processing on the plurality of target signal segments to obtain recognition results corresponding to the plurality of target signal segments, the method further includes: The classification algorithm is trained using training images, wherein the events that are identified in the training images are of the same type as the events that are identified in the images. Before performing a segmentation operation on the EEG signal based on the display time of the image to obtain multiple target signal segments, the method further includes: The event potential signal is filtered and denoised.

5. The method according to any one of claims 1 to 4, characterized in that, Determining the target recognition result of the image based on the recognition results corresponding to the multiple target signal segments includes: Determine the weights of multiple EEG devices; The recognition result of the image is determined based on the weights of the multiple EEG devices and the recognition results of the corresponding EEG signals.

6. An image processing apparatus, characterized in that, include: The acquisition module acquires the electroencephalogram (EEG) signals of the target object receiving image stimuli via an EEG device, including: The device receives event potential signals collected by multiple brainwave devices, wherein the event potential signals are potential signals collected by the brainwave devices when the target object recognizes the image; it also receives a signal synchronization tag sent by a display device to the brainwave device; wherein the brainwave device is worn on the target object viewing the image, and the image is displayed on the display device. The interception module performs an interception operation on the EEG signal according to the display time of the image to obtain multiple target signal segments, including: determining the triggering time period of the event potential triggered by the recognition result corresponding to the event; determining the display time of the image according to the signal synchronization tag; and intercepting the event potential signal according to the display time of the image and the triggering time period to obtain the target signal segments. The processing module performs waveform processing on the plurality of target signal segments respectively to obtain recognition results corresponding to the plurality of target signal segments respectively, wherein the recognition result is the recognition result of the image; The determination module determines the target recognition result of the image based on the recognition results corresponding to the multiple target signal segments.

7. A computer-readable storage medium, characterized in that, The storage medium includes a stored program, wherein the program executes the image processing method according to any one of claims 1 to 5.

8. An electronic device, characterized in that, It includes one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the image processing method according to any one of claims 1 to 5.

Images