Non-target of interest filtering method and apparatus, and device and storage medium

Abstract

The present application relates to the field of machine vision. Provided are a non-target of interest filtering method and apparatus, and a device and a storage medium. The method comprises: acquiring a learning sample image, which is an image captured with respect to a target region, wherein the learning sample image comprises a target of interest; on the basis of the learning sample image, determining pixel filtering parameters, which are used for representing the imaging sizes of the target of interest at different positions of the target region; and on the basis of the pixel filtering parameters, filtering out a non-target of interest from the target region in an image to be processed. The method is applicable to a target detection process, and is used for reducing labor costs incurred by manual calibration for filtering parameters when a non-target of interest is filtered out.

Description

Non-interest target filtering method, device, equipment and storage medium

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on March 4, 2024, with application number 202410245861.X and invention name “Non-interest target filtering method, device, equipment and storage medium”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of machine vision, and in particular to a method, apparatus, device and storage medium for filtering non-interested targets. Background Art

[0003] Object detection products use object detection technology to identify and detect objects of interest (such as people and vehicles) in images or videos. To improve the detection accuracy of these products, you can filter out non-objects of interest in images or videos, analyzing and processing only the objects of interest.

[0004] Currently, filtering out non-objects of interest requires manual configuration of the size of the objects of interest.

[0005] However, manually configuring the size of the target of interest is relatively complicated and requires the joint operation of multiple people, resulting in high labor costs. Summary of the Invention

[0006] The present application provides a method, apparatus, device and storage medium for filtering non-interesting targets, which can filter non-interesting targets through pixel filtering parameters obtained through self-learning, saving labor costs.

[0007] In a first aspect, the present application provides a method for filtering non-targets of interest, the method comprising: obtaining a learning sample image; the learning sample image is an image taken for a target area; the learning sample image includes a target of interest; determining a pixel filtering parameter based on the learning sample image; the pixel filtering parameter is used to characterize the size of the imaging size of the target of interest at different positions in the target area; based on the pixel filtering parameter, filtering non-targets of interest within the target area in the image to be processed; the image to be processed is an image taken for the target area.

[0008] In a second aspect, the present application provides a non-interest target filtering device, which includes various functional modules used in the method described in the first aspect above.

[0009] In a third aspect, the present application provides an electronic device comprising a processor and a memory; the memory stores instructions executable by the processor; when the processor is configured to execute the instructions, the electronic device implements the method described in the first aspect above.

[0010] In a fourth aspect, the present application provides a computer program product. When the computer program product is run in an electronic device, the electronic device executes the steps of the related method described in the first aspect to implement the method described in the first aspect.

[0011] In a fifth aspect, the present application provides a readable storage medium, which includes: software instructions; when the software instructions are executed in an electronic device, the electronic device implements the method described in the first aspect above.

[0012] The non-interest target filtering method provided in the present application can obtain an image of the target of interest in the target area as a learning sample image, and determine the pixel filtering parameters that characterize the imaging size based on the learning sample image, and use the pixel filtering parameters to filter out non-interest targets. Non-interest targets can be filtered out through self-learning, without the need for manual calibration by multiple people, which can save labor costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The drawings described herein are used to provide further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute improper limitations on the present application.

[0014] FIG1 is a schematic diagram of the composition of a non-interest target filtering system provided in an embodiment of the present application;

[0015] FIG2 is a schematic diagram of the composition of an electronic device provided in an embodiment of the present application;

[0016] FIG3a is a schematic diagram of a process of filtering non-interested targets according to an embodiment of the present application;

[0017] FIG3 b is a schematic diagram of an implementation process of a method for filtering non-interested targets provided in an embodiment of the present application;

[0018] FIG4 is a schematic diagram of the same target of interest provided in an embodiment of the present application;

[0019] FIG5 is another schematic diagram of the same target of interest provided in an embodiment of the present application;

[0020] FIG6 is a schematic diagram of the output of the target detection algorithm provided in an embodiment of the present application;

[0021] FIG7 is a schematic diagram of a target chain provided in an embodiment of the present application;

[0022] FIG8 is a schematic diagram of sub-areas provided in an embodiment of the present application;

[0023] FIG9 is a schematic diagram of the 3sigma principle provided in an embodiment of the present application;

[0024] FIG10 is a schematic diagram showing the composition of a non-target-of-interest filtering device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0025] In the following, the terms "first," "second," and "third," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the technical features indicated. Thus, a feature designated as "first," "second," or "third," etc., may explicitly or implicitly include one or more of the features.

[0026] Object detection products use this technology to identify and detect objects of interest in images or videos. To improve accuracy, these products can filter out non-objects of interest in images or videos, analyzing and processing only the objects of interest.

[0027] In smart home or intrusion detection scenarios, target detection products can collect images of specific areas, then filter out irrelevant targets of interest such as pets and sweeping robots in the images, and only prompt detected targets of interest (such as people).

[0028] In vehicle-assisted driving scenarios, target detection products can first acquire images around the vehicle, then filter out irrelevant buildings, billboards, and other non-targets of interest in the images, and only analyze and respond to targets of interest (such as pedestrians, obstacles, or other vehicles).

[0029] In the apron violation detection scenario, human activities are not allowed in certain areas of the apron (such as the runway and other specific areas). Target detection products can obtain images of the aforementioned areas, and then filter out non-interesting targets such as irrelevant apron special vehicles and aircraft in the images, and only prompt targets of interest (such as pedestrians).

[0030] Currently, filtering out non-targets of interest first requires manually configuring the size parameters of the target of interest, and then comparing the size parameters of the detected candidate targets with the manually configured size parameters of the target of interest. If they are consistent, the candidate target is considered to be a target of interest and retained. If they are inconsistent, the candidate target is considered to be a non-target of interest and filtered out.

[0031] Manually configuring the size parameters of an object of interest relies on experience, which can vary between configurers. To avoid errors caused by experience differences, multiple configurers are typically required to jointly configure the size parameters. The final size parameters used for filtering are then combined. However, manually configuring the size of an object of interest is complex, requiring multiple people to work together and resulting in high labor costs.

[0032] Based on this, the embodiments of the present application provide a method, apparatus, device and storage medium for filtering non-interested targets, which can filter non-interested targets through pixel filtering parameters obtained through self-learning, thereby saving labor costs.

[0033] The following is an introduction with reference to the accompanying drawings.

[0034] FIG1 is a schematic diagram illustrating the components of a non-object-of-interest filtering system provided in an embodiment of the present application. As shown in FIG1 , the system includes an image acquisition device 100 and a detection device 200. The image acquisition device 100 and the detection device 200 can be connected via a wired or wireless network. In some scenarios, the image acquisition device 100 and the detection device 200 can be integrated into a single device, and communication between the two can be accomplished via a circuit board or communication line.

[0035] The image acquisition device 100 may include a visible light camera and a thermal imager (e.g., an infrared camera). For example, the image acquisition device 100 may be a binocular camera or a multi-lens camera including an optical camera and a thermal imager. The image acquisition device 100 may be used to capture both visible light and thermal images of a target area. Currently, the image acquisition device 100 may also be a separate visible light camera or thermal imager, and all of these are within the scope of protection of this application.

[0036] In some embodiments, the image acquisition device 100 may also be configured to transmit the captured visible light image and thermal image of the target area to the detection device 200. For example, the image acquisition device 100 may encode the visible light image and thermal image to obtain encoded image data, and then transmit the encoded image data to the detection device 200, which then decodes the encoded data to obtain the visible light image and thermal image.

[0037] As described above, the image acquisition device 100 and the detection device 200 may be connected via a wireless network or a wired network. The wired network or wireless network may include one or more media or devices capable of transmitting the encoded image data from the image acquisition device 100 to the detection device 200.

[0038] In some embodiments, the wired or wireless network may include one or more communication media that enable the image acquisition device 100 to transmit the encoded image data directly to the detection device 200. In this embodiment, the image acquisition device 100 may modulate the encoded image data according to a communication standard (e.g., a wireless communication protocol) and transmit the modulated encoded image data to the detection device 200. The one or more communication media may include wireless and / or wired communication media, such as a radio frequency (RF) spectrum or one or more physical transmission lines.

[0039] Alternatively, the one or more communication media may form part of a packet-based network, such as a local area network, a wide area network, or a global network (such as the Internet).

[0040] Optionally, the one or more communication media may also include a router, a switch, a base station, or other devices that facilitate communication from the image acquisition device 100 to the detection device 200 .

[0041] The detection device 200 can be an electronic device with computing and processing capabilities, such as a computer or a server.

[0042] Among them, the server can be a single server, or it can be a server cluster composed of multiple servers. In some implementations, the server cluster can also be a distributed cluster. Optionally, the server can also be implemented on a cloud platform. For example, the cloud platform can include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an inter-cloud, and a multi-cloud, etc., or any combination thereof. The embodiments of the present application are not limited to this.

[0043] The detection device 200 can filter out non-interesting objects by self-learning pixel filtering parameters based on the visible light image and thermal imaging image collected by the image acquisition device 100. The specific process can be referred to the non-interesting object filtering method provided in the following embodiment, and will not be repeated here.

[0044] In some embodiments, the detection device 200 may also be used to perform target detection of targets of interest on the visible light images and thermal imaging images captured by the image capture device 100 .

[0045] In some embodiments, the detection device 200 may be further connected to a prompting device. When the detection device 200 detects an object of interest in the target area, the prompting device may issue a prompt message.

[0046] It should be noted that the above description uses the example of image acquisition device 100 and detection device 200 being independent devices. Alternatively, image acquisition device 100 and detection device 200 may be integrated into one device. That is, image acquisition device 100 or its corresponding functions and detection device 200 or its corresponding functions may be integrated into a single device. For example, a camera having both image acquisition and object detection functions may be used. This embodiment of the present application is not limited to this.

[0047] The execution subject of the non-interest target filtering method provided in the embodiment of the present application is the above-mentioned detection device 200. As mentioned above, the detection device 200 can be an electronic device with computing and processing functions such as a computer or a server. Optionally, the detection device 200 can also be a processor (such as a central processing unit (CPU)) in the aforementioned electronic device; or, the detection device 200 can also be an application (application, APP) with a non-interest target filtering function installed in the aforementioned electronic device; or, the detection device 200 can also be a software system or platform deployed in the aforementioned electronic device; or, the detection device 200 can also be a functional module that executes the non-interest target filtering method in the aforementioned electronic device, etc. The embodiment of the present application does not limit this.

[0048] For simplicity of description, the following description will be made by taking the detection device 200 as an electronic device as an example.

[0049] FIG2 is a schematic diagram of the composition of an electronic device provided in an embodiment of the present application. As shown in FIG2 , the electronic device may include: a processor 10 , a memory 20 , a communication circuit 30 , a communication interface 40 , and an input / output interface 50 .

[0050] The processor 10 , the memory 20 , the communication interface 40 , and the input / output interface 50 may be connected via a communication line 30 .

[0051] The processor 10 is used to execute the instructions stored in the memory 20 to implement the non-interest target filtering method provided in the following embodiments of the present application. The processor 10 can be a CPU, a general-purpose processor network processor (NP), a digital signal processor (DSP), a microprocessor, a microcontroller (MCU) / single-chip microcomputer / single-chip microcomputer, a programmable logic device (PLD) or any combination thereof. The processor 10 can also be any other device with a processing function, such as a circuit, a device or a software module, which is not limited by the embodiments of the present application. In one example, the processor 10 may include one or more CPUs, such as CPU0 and CPU1 in Figure 2. As an optional implementation method, the electronic device may include multiple processors. For example, in addition to the processor 10, it may also include a processor 60 (shown as an example by a dotted line in Figure 2).

[0052] Memory 20 is used to store instructions. For example, the instruction can be a computer program. Optionally, the memory 20 can be a read-only memory (ROM) or other types of static storage devices that can store static information and / or instructions, or a random access memory (RAM) or other types of dynamic storage devices that can store information and / or instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, etc., and the embodiments of the present application are not limited to this.

[0053] It should be noted that the memory 20 may exist independently of the processor 10 or may be integrated with the processor 10. The memory 20 may be located inside the electronic device or outside the electronic device, which is not limited in the embodiment of the present application.

[0054] The communication line 30 is used to transmit information between the components included in the electronic device.

[0055] Communication interface 40 is used to communicate with other devices (e.g., the image acquisition device 100 and the prompting device) or other communication networks. Such other communication networks may be Ethernet, radio access networks (RAN), wireless local area networks (WLAN), etc. Communication interface 40 may be a module, circuit, transceiver, or any other device capable of communication.

[0056] The input / output interface 50 is used to implement human-computer interaction between a user and the electronic device, for example, to implement action interaction or information interaction between the user and the electronic device.

[0057] For example, the input / output interface 50 may be a mouse, keyboard, display screen, or touch screen screen, etc. Action interaction or information interaction between a user and the electronic device may be achieved through the mouse, keyboard, display screen, or touch screen screen, etc.

[0058] It should be noted that the structure shown in FIG2 does not constitute a limitation on the electronic device. In addition to the components shown in FIG2 , the electronic device may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0059] The following introduces the non-interest target filtering method provided in the embodiment of the present application.

[0060] Figure 3a is a flow chart of a method for filtering non-interested targets provided by an embodiment of the present application. Optionally, the method can be executed by an electronic device having the hardware structure shown in Figure 2 above, as shown in Figure 3a, the method includes S101 to S103.

[0061] S101: Obtain a learning sample image.

[0062] The learning sample images are images captured within the target area and include objects of interest. The target area can be any relevant area. For example, for a smart home scenario, the target area can be an indoor area; for a roadside parking violation detection scenario, the target area can be both sides of the road. The objects of interest are also context-dependent. For example, in the smart home scenario, intrusion detection scenario, and apron violation detection scenario, the objects of interest can be pedestrians, people, or human bodies.

[0063] The learning sample images may be collected in advance. In one example, as shown in the portion corresponding to S101 in FIG. 3 b , the learning sample images may be collected by an image acquisition device.

[0064] Optionally, the learning sample images may be images with relatively high imaging quality. For example, in one possible implementation, the learning sample images may include daytime images captured by an image capture device during the day.

[0065] Optionally, during the day, the lighting conditions are generally good, so the daytime image may include a visible light image and a thermal imaging image.

[0066] It should be understood that the image size of the target of interest learned using a single-channel visible light image or thermal image as a learning sample image may contain errors. The non-target of interest filtering method provided in the embodiment of the application can use a dual-channel visible light image and thermal image as a learning sample image, which can improve the accuracy of the learned image size.

[0067] In another possible implementation, the learning sample images may include night images captured by an image capture device at night.

[0068] Optionally, in the case of night, the lighting condition is generally poor, so the night image may only include a thermal imaging image.

[0069] In another possible implementation, the learning sample images may include both daytime images captured by the image capture device during the day and nighttime images captured by the image capture device at night.

[0070] In some possible embodiments, the electronic device may determine the daytime image and the nighttime image as learning sample images when a high-confidence target of interest is detected in the daytime image and the nighttime image. In this case, the above S101 may specifically include the following steps:

[0071] Step 1a: Under daytime conditions, obtain visible light images and thermal imaging images taken of the target area.

[0072] The daytime condition can be determined based on the timestamp when the visible light image and the thermal image are captured. If the timestamp is within a preset daytime period, the electronic device can determine that it is daytime. For example, the daytime period can be from 8:00 AM to 6:00 PM or other time periods. This embodiment of the application is not limited to this. Alternatively, the daytime condition can also be determined based on the light intensity when the visible light image and the thermal image are captured. This embodiment of the application does not limit the specific criteria for determining the daytime condition.

[0073] For example, as described in the non-interest target filtering system of FIG1 , the electronic device (detection device 200 ) can obtain a visible light image and a thermal imaging image of the target area by receiving the encoded image data sent by the image acquisition device 100 and decoding the encoded image data.

[0074] Step 2a: If the same target of interest is detected in the visible light image and the thermal imaging image, the visible light image and the thermal imaging image are determined as learning sample images.

[0075] For example, the electronic device may use a preset target detection algorithm to perform target detection on the visible light image and the thermal image to determine whether the image contains a target of interest. If the target detection algorithm detects a target of interest in both the dual-channel visible light image and the thermal image, it may further determine whether the detected targets of interest are the same target.

[0076] For example, the preset target detection algorithm may include SIFT (Scale Invariant Feature Transform), SURF (Speeded Up Robust Features), Adaboost (Adaptive Boosting), or a deep learning algorithm. The embodiments of the present application do not limit the specific type of target detection algorithm.

[0077] For example, Figure 4 is a schematic diagram of an identical target of interest provided in an embodiment of the present application. As shown in Figure 4, the electronic device can overlay and align a visible light image and a thermal image. If the image overlap between the target of interest detected in the visible light image and the target of interest detected in the thermal image is high (e.g., greater than a preset overlap threshold), the electronic device can determine that the identical target of interest has been detected in the visible light image and the thermal image.

[0078] The overlap threshold value can be preset in the electronic device by the administrator. For example, the overlap threshold value can be set to 80%, 85%, or 90%, etc. The embodiment of the present application does not limit the specific value of the overlap threshold value. For example, as described above, the electronic device may include an input and output interface 50, which may be a mouse, a keyboard, or a touch screen, etc. The electronic device may receive the overlap threshold value input by the administrator through the mouse, keyboard, or touch screen, etc.

[0079] For another example, Figure 5 is another schematic diagram of the same target of interest provided by an embodiment of the present application. As shown in Figure 5, the electronic device can use detection frames on the visible light image and the thermal image to mark the detected target of interest. In this case, the electronic device can overlay the visible light image and the thermal image and align them. If the detection frame in the visible light image and the detection frame in the thermal image have a high degree of overlap (for example, greater than a preset overlap threshold), the electronic device can determine that the same target of interest has been detected in the visible light image and the thermal image.

[0080] For another example, when the resolutions of the visible light image and the thermal imaging image are the same, the electronic device may also separately record the position coordinates of the detection frame in the visible light image (for example, the coordinates of the center point or the four vertices) and the position coordinates of the detection frame in the thermal imaging image (for example, the coordinates of the center point or the four vertices), and then calculate the difference between the position coordinates of the detection frames. If the difference between the position coordinates is less than a preset coordinate difference threshold, the electronic device may determine that the same target of interest is detected in the visible light image and the thermal imaging image.

[0081] In an embodiment of the present application, after target detection is performed on the visible light image and thermal imaging image of a certain target, if the target is identified as a target of interest by the target detection algorithm in both the day image and the night image (that is, the same target of interest is detected in the visible light image and the thermal imaging image), the electronic device can confirm that the target is a target of interest with high confidence, and determine the corresponding day image and night image as learning sample images.

[0082] Step 3a: Under dark conditions, obtain a thermal imaging image of the target area.

[0083] The darkness condition can be determined based on the timestamp when the visible light image and thermal image were captured. If the timestamp falls within a preset darkness period, the electronic device can determine that it is a darkness condition. For example, the darkness period can be from 6:00 PM to 8:00 AM or other time periods. This embodiment of the application is not limited to this. Alternatively, the darkness condition can be determined based on the light intensity when the visible light image and thermal image were captured. This embodiment of the application does not limit the specific criteria for determining the darkness condition.

[0084] The specific process of step 3a can refer to the above step 1a and will not be repeated here.

[0085] Step 4a: Detect the thermal imaging image using a preset target detection algorithm.

[0086] The target detection algorithm may include SIFT, SURF, Adaboost, or a deep learning algorithm, etc. The embodiment of the present application does not limit the specific type of the target detection algorithm.

[0087] Step 5a: If the target detection algorithm detects a target in the thermal imaging image and the probability that the target is the target of interest is higher than the probability threshold, the thermal imaging image is determined as a learning sample image.

[0088] The probability threshold may be preset in the electronic device by a manager. For example, the probability threshold may be set to 80%, 85%, or 90%, etc. The embodiment of the present application does not limit the specific value of the probability threshold.

[0089] For example, Figure 6 is a schematic diagram of the output of a target detection algorithm provided in an embodiment of the present application. As shown in Figure 6, most target detection algorithms, in addition to detecting a target of interest, can also output a probability of detecting the target as a target of interest. In this case, the electronic device can use the probability output by the target detection algorithm as a confidence level. If the confidence level is high, the target can be determined to be a target of interest.

[0090] It should be understood that during the target detection process, multiple targets may be detected, and these multiple targets may include targets of interest and targets of no interest. The method for filtering targets of no interest provided in the embodiment of the present application, for daytime scenes, if the same target of interest is detected in the visible light image and the thermal imaging image, it means that the confidence that the target is the target of interest is high. For night scenes, if the target is detected in the thermal imaging image of a certain target, and the probability of the target being the target of interest output by the target detection algorithm is higher than the probability threshold, it means that the confidence that the target is the target of interest is also high. Using the image of the target of interest with higher confidence in the above two cases as the learning target can avoid interference from other targets of no interest and improve the accuracy of filtering targets of no interest.

[0091] Optionally, after determining a target of interest with a higher confidence level, the electronic device may also store some relevant information of the target of interest (for example, a unique identifier, the type of target, the probability of detecting it as that type, the pixel coordinates of the detection frame, the imaging height in the image, and the imaging width in the image, etc.) into a target linked list for subsequent application calls.

[0092] For example, FIG7 is a schematic diagram of a target linked list provided in an embodiment of the present application. As shown in FIG7 , the target linked list may include multiple nodes ( FIG7 uses node 1, node 2, and node 3 as an example), each of which may include a pointer field and a data field. The pointer field may be used to store the first address of the next node, and the data field may be used to store some relevant information of the target of interest.

[0093] It should be noted that FIG7 is introduced above using a one-way linked list as an example. Optionally, the target linked list may also be a two-way linked list or other types of linked lists. The embodiment of the present application does not limit the specific type of the target linked list.

[0094] S102: Determine pixel filtering parameters based on the learning sample image.

[0095] The pixel filtering parameters are used to characterize the size of the image of the target of interest at different locations in the target area. Taking the target of interest as a person as an example, the size of the image represented by the pixel filtering parameters can be shown in the portion corresponding to S102 in FIG3b.

[0096] In one possible implementation, when the image acquisition device forms a large angle with the ground plane, and the size of an object of interest at various locations in the target area image is relatively similar, the electronic device can learn a global pixel filtering parameter for the entire target area image. The specific implementation of the global pixel filtering parameter can be referenced in another possible implementation described below and will not be further elaborated here.

[0097] In some application scenarios, due to the existence of the perspective principle, the imaging size of an object of interest at different positions in the target area image may have large differences. In this case, in order to ensure the accuracy of filtering non-objects of interest, different pixel filtering parameters can be learned for different positions of the target area image.

[0098] Therefore, in another possible implementation, the electronic device may divide the image of the target area into sub-areas of multiple rows and columns, and learn a pixel filtering parameter for each sub-area. In this case, the above S102 may specifically include the following steps:

[0099] Step 1b: Divide the image of the target area into multiple sub-areas.

[0100] Optionally, the multiple sub-regions may be distributed in a determinant-column manner, that is, the electronic device may specifically divide the image of the target region into sub-regions of M rows and N columns.

[0101] Wherein, M and N are both positive integers greater than 1. For example, M and N can be 5, 6, or 7. This embodiment of the application does not limit the specific values ​​of M and N. The shape of the sub-region can be a square, a circle, an ellipse, or a parallelogram. This embodiment of the application does not limit the specific shape of the sub-region.

[0102] For example, Figure 8 is a schematic diagram of subregions provided in an embodiment of the present application. As shown in Figure 8, taking a grid-shaped subregion as an example, for an object of interest of the same size, the farther away from the grid in the image (relative to the viewer's position), the smaller the image size; the closer the grid in the image, the larger the image size. Therefore, the image size of the object of interest varies depending on its distance from the grid in the image.

[0103] Optionally, the target area may be divided into multiple sub-areas that are not arranged in a determinant, and may be divided in other arrangements, such as by generating straight lines in any direction for division, etc. This embodiment of the present application does not limit this.

[0104] Step 2b: Determine the pixel filtering parameters corresponding to each of the multiple (eg, M rows and N columns) sub-regions of the object of interest based on the learning sample image.

[0105] In one possible implementation, for any first sub-region in the M rows and N columns of sub-regions, the electronic device can count the imaging heights of multiple targets of interest appearing in the first sub-region in the image of the first sub-region, and determine a confidence interval based on the statistically obtained imaging heights of the multiple targets of interest as the pixel filtering parameter corresponding to the first sub-region.

[0106] The imaging height may also be understood as the number of pixels occupied when the object of interest is imaged in the height direction in the image of the first sub-region.

[0107] For example, the electronic device may calculate the average and standard deviation of the imaging heights of multiple objects of interest, and use the imaging height three times the standard deviation above the average as the upper limit of the confidence interval, and use the imaging height three times the standard deviation below the average as the lower limit of the confidence interval, to obtain the confidence interval as the pixel filtering parameter corresponding to the first sub-region. For another example, the electronic device may use the imaging height that is a preset value above the average as the upper limit of the confidence interval, and use the imaging height that is a preset value below the average as the lower limit of the confidence interval, to obtain the confidence interval as the pixel filtering parameter corresponding to the first sub-region.

[0108] In another possible implementation, for any first sub-region in the M rows and N columns of the sub-region, the electronic device may determine a confidence interval as a pixel filtering parameter based on the ratio between the imaging size and the reference size. In this case, step 2b may specifically include the following steps:

[0109] Step 2.1b: When the target of interest appears in the first sub-region in the learning sample image, record a set of pixel statistics, and obtain multiple sets of pixel statistics corresponding to when multiple targets of interest appear in the first sub-region.

[0110] The pixel statistics include the imaging height of the corresponding object of interest in the first sub-region image.

[0111] For example, whenever an object of interest appears in the first sub-region, the electronic device can record a set of pixel statistics (x, y, w, h) for the first sub-region, where x and y represent the pixel coordinates of the bottom of the detection frame of the object of interest, w represents the image width of the object of interest in the image of the first sub-region, and h represents the image height of the object of interest in the image of the first sub-region.

[0112] Step 2.2b: Determine a plurality of height ratio parameters corresponding to each of the plurality of sets of pixel statistical data based on the plurality of sets of pixel statistical data.

[0113] The height ratio parameter is the ratio of the image height of the object of interest in the first sub-region image to a preset reference height. For example, if the object of interest is a human body, the preset reference height can be set to 1.7 meters, 1.73 meters, or the like. The present embodiment does not limit the specific value of the preset reference height.

[0114] Step 2.3b. Calculate the mean and standard deviation of multiple height scale parameters.

[0115] Optionally, the electronic device may calculate the average value according to the following formula (1):

[0116] In formula (1), μ represents the average value of multiple height ratio parameters. i Represents the height ratio parameter corresponding to the i-th target of interest. i H represents the imaging height of the i-th target of interest in the first sub-region image. i represents the preset reference height. N represents the number of objects of interest appearing in the first sub-area.

[0117] Optionally, the electronic device may first calculate the variance of the plurality of height ratio parameters, and then square the variance to obtain the standard deviation.

[0118] Optionally, the electronic device may calculate the variance according to the following formula (2):

[0119] In formula (2), σ 2 Represents the variance of multiple height scale parameters.

[0120] Step 2.4b: The difference between the mean and three times the standard deviation is used as the lower limit, and the sum of the mean and three times the standard deviation is used as the upper limit to obtain the confidence interval as the pixel filtering parameter corresponding to the first sub-region.

[0121] For example, FIG9 is a schematic diagram of the 3sigma principle provided in an embodiment of the present application. As shown in FIG9 , for a dataset that conforms to a normal distribution, 68.2% of the data falls within the range of (μ-σ, μ+σ), 95.4% of the data falls within the range of (μ-2σ, μ+2σ), and 99.7% of the data falls within the range of (μ-3σ, μ+3σ). Therefore, using three times the standard deviation above and below the mean of the height ratio parameter as the confidence interval can screen out most targets of interest.

[0122] Optionally, the electronic device may also use the difference between the average value and X times the standard deviation as the lower limit and the sum of the average value and X times the standard deviation as the upper limit to obtain a confidence interval as the pixel filtering parameter corresponding to the first sub-region.

[0123] Wherein, X is a natural number, and X is not 3. For example, X can be 1, 2, or 4, etc., which is not limited in the embodiment of the present application.

[0124] It should be noted that for common personnel detection scenarios, people are the target of interest, and the difference between other non-interested targets such as pets and sweeping robots and people is mainly reflected in height. Therefore, the above embodiment is described by taking the pixel filtering parameters obtained according to the imaging height as an example. Optionally, when the difference between other targets of interest and non-interested targets is mainly reflected in width, the electronic device can also obtain the filtering parameters according to the imaging width. The specific process can be referred to the description in the above embodiment and will not be repeated here.

[0125] S103: Filter out non-interested objects within the target area in the image to be processed based on the pixel filtering parameters.

[0126] The image to be processed is an image captured of the target area. The embodiments of the present application do not limit the specific types of non-interested targets. Taking the above-mentioned smart home scenario, intrusion detection scenario, and apron violation detection scenario as examples, non-interested targets can be understood as pets, sweeping robots, vehicles, and airplanes. In one example, as shown in the portion corresponding to S103 in Figure 3b, pixel filtering parameters can be used to determine that a person is an interesting target, while a small robot is a non-interested target.

[0127] In a possible implementation, the electronic device may directly delete non-interested targets in the target area image.

[0128] In another possible implementation, after detecting different targets (targets of interest and targets of non-interest), the electronic device can store some target-related information in a target list. The target list can include multiple nodes, each node is used to store information related to a target. In this case, when the electronic device performs filtering, it can delete the nodes corresponding to non-interesting targets in the target list, or the related information stored in the nodes, to obtain and output the final full-graph filter table.

[0129] The target linked list may refer to FIG7 above, which will not be described in detail here.

[0130] Optionally, as described above, the electronic device can learn a global pixel filtering parameter for the entire target area. In this case, the electronic device can filter out non-interesting objects within the target area based on the global pixel filtering parameter. The specific filtering process can be referred to steps 3c to 6c below and will not be repeated here.

[0131] Optionally, as described above, the electronic device may divide the image of the target area into sub-areas of multiple rows and columns, and learn a confidence interval for each sub-area as a pixel filtering parameter. In this case, the electronic device may obtain an image to be processed taken of the target area, divide the image to be processed into sub-areas of M rows and N columns, and determine the target sub-area where the candidate target is located from the image to be processed. If the size of the candidate target imaged in the target sub-area does not match the target pixel filtering parameter corresponding to the target sub-area, the candidate target is filtered as a non-target of interest. The specific matching process can also be referred to the following steps 3c to 6c, which will not be repeated here.

[0132] For example, when an electronic device detects a candidate target, it can record a set of the above-mentioned pixel statistics (x, y, w, h), where x and y represent the pixel coordinates at the bottom of the detection frame. In this case, the electronic device can determine the target sub-area where the candidate target is located based on the pixel coordinates at the bottom of the detection frame.

[0133] The non-interest target filtering method provided in the embodiment of the present application can also divide the image of the target area into sub-areas of M rows and N columns, and learn a corresponding pixel filtering parameter for each sub-area, and use the pixel filtering parameters corresponding to each sub-area to filter the non-interest targets in the sub-area, which can avoid incorrect filtering caused by the mismatch between the pixel filtering parameters and the sub-areas in the image (for example, the pixel filtering parameters with larger upper and lower limits are used for the sub-areas farther away in the image, and the pixel filtering parameters with smaller upper and lower limits are used for filtering for the sub-areas closer in the image), thereby improving the accuracy of filtering non-interest targets.

[0134] In some possible embodiments, if the corresponding pixel filtering parameters have not been learned for a target sub-region, the electronic device may temporarily use the pixel filtering parameters corresponding to the sub-region in the same row to filter out non-targets of interest in the target sub-region. That is, if the corresponding pixel filtering parameters have not been learned for a target sub-region, and the image size of the candidate target in the target sub-region does not match the reference pixel filtering parameters for the target sub-region, the electronic device may filter the candidate target as a non-target of interest.

[0135] The reference pixel filtering parameters of the target sub-region are determined based on the pixel filtering parameters corresponding to the sub-region in the same row as the target sub-region in the M-row and N-column sub-regions of the image to be processed.

[0136] For example, the electronic device may directly select the pixel filtering parameter corresponding to any sub-region in the same row as the reference pixel filtering parameter.

[0137] For another example, the electronic device may take the average value (of the upper and lower limits), the median value (of the upper and lower limits), the maximum value (of the upper and lower limits), or the minimum value (of the upper and lower limits) of multiple pixel filtering parameters corresponding to multiple sub-regions in the same row as the reference pixel filtering parameter (the upper and lower limits of the confidence interval), etc. The embodiments of the present application are not limited to this.

[0138] It should be understood that, based on the principle of perspective, objects of interest of the same size appear at substantially the same imaging height in the same row of sub-region images. The non-interest object filtering method provided in the embodiments of the present application can, when pixel filtering parameters have not been learned for a first sub-region, determine reference filtering parameters based on the pixel filtering parameters corresponding to sub-regions of substantially the same imaging height in the same row, and use the reference filtering parameters to filter the first sub-region. This can avoid the situation where a sub-region cannot filter non-interest objects when pixel filtering parameters have not been learned, thereby improving the versatility of the solution.

[0139] In some possible embodiments, the user may also preset a size range that the electronic device wishes to detect. In this case, the above S103 may specifically include the following steps:

[0140] Step 1c: Obtain the size range of the target of interest set by the user.

[0141] For example, taking people as the target of interest, the size range can be set to 1.5 meters to 1.8 meters.

[0142] Step 2c: convert the target pixel filtering parameter into a reference size interval according to a preset conversion relationship.

[0143] For example, when the resolution and area of ​​the target area are known, there is a fixed proportional relationship between the number of pixels in the target area image and the actual size of the object. The electronic device can use this fixed proportional relationship as a conversion relationship to convert the imaging height (number of pixels) of the upper and lower limits of the target pixel filtering parameters (confidence interval) into the actual height of the object to obtain a reference size interval.

[0144] Step 3c: When the size interval set by the user does not intersect with the reference size interval, a first imaging height of the candidate object in the first sub-region image is obtained.

[0145] Optionally, when the size interval set by the user and the reference size interval intersect, the electronic device may not filter out non-interested objects.

[0146] For example, if the user sets the detection target to be more than one meter, and the learned reference size interval is [1.5m, 1.8m], that is, the two intervals have an intersection, no filtering will be performed. If the user sets the detection target to be less than one meter, that is, the two intervals do not have an intersection, filtering will be performed.

[0147] Step 4c: Determine a reference height ratio parameter according to the ratio of the first imaging height to the preset reference height.

[0148] The process of calculating the reference height ratio parameter in step 4c can refer to the height ratio parameter in the above formula (1) or formula (2), and will not be repeated here.

[0149] Step 5c: If the reference height ratio parameter is within the confidence interval of the target pixel filtering parameter, the candidate target is retained as the target of interest.

[0150] Step 6c: If the reference height ratio parameter is outside the confidence interval of the target pixel filtering parameter, the candidate target is filtered as a non-target of interest.

[0151] It should be understood that since the manual configuration of the size parameters of the target of interest depends on manual experience, the manual experience of different configuration personnel may be different. For example, in order to avoid errors caused by differences in manual experience, it is usually necessary for multiple configuration personnel to jointly configure the size parameters, and then combine the size parameters jointly configured by multiple configuration personnel to obtain the size parameters used for the final filtering. Such a configuration scheme requires multiple people to work together, and the labor cost is high. The non-interest target filtering method provided in the embodiment of the present application can obtain an image of the target of interest in the target area as a learning sample image, and determine the pixel filtering parameters that characterize the imaging size based on the learning sample image, and use the pixel filtering parameters to filter non-interest targets. Non-interest targets are filtered by self-learning, without the need for manual calibration by multiple people, which can save labor costs.

[0152] The above mainly introduces the solution provided by the embodiment of the present application from the perspective of the method. In order to realize the above functions, it includes hardware structures and / or software modules corresponding to the execution of each function. It should be easy to realize that the technical goals in this field are combined with the units and algorithm steps of each example described in the embodiments disclosed herein, and the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional technical goals can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0153] In an exemplary embodiment, the present application also provides a non-interest target filtering device, which can be used in the above-mentioned detection device 200 or electronic device. Figure 10 is a schematic diagram of the components of the non-interest target filtering device provided in the present application. As shown in Figure 10, the device includes: an acquisition module 1001 and a processing module 1002.

[0154] The acquisition module 1001 is used to acquire a learning sample image; the learning sample image is an image captured for a target area; the learning sample image includes an object of interest.

[0155] Processing module 1002 is used to determine pixel filtering parameters based on the learning sample image; the pixel filtering parameters are used to characterize the size of the image of the target of interest at different positions in the target area; based on the pixel filtering parameters, non-interested targets in the target area in the image to be processed are filtered out; the image to be processed is an image taken of the target area.

[0156] In some possible embodiments, the learning sample images include daytime images and / or nighttime images; daytime images include visible light images and thermal imaging images; and nighttime images include thermal imaging images.

[0157] In some other possible embodiments, the acquisition module 1001 is specifically configured to acquire a visible light image and a thermal image taken of a target area during the day. The processing module 1002 is specifically configured to determine the visible light image and the thermal image as learning sample images if the same target of interest is detected in the visible light image and the thermal image. The acquisition module 1001 is specifically configured to acquire a thermal image taken of a target area at night. The processing module 1002 is specifically configured to detect the thermal image using a preset target detection algorithm; if the target detection algorithm detects a target in the thermal image, and the probability that the target is the target of interest is higher than a probability threshold, the thermal image is determined as a learning sample image.

[0158] In some other possible embodiments, the processing module 1002 is specifically used to divide the image of the target area into sub-areas of M rows and N columns; M and N are both positive integers greater than 1; based on the learning sample image, determine the pixel filtering parameters corresponding to each sub-area of ​​the target of interest in the M rows and N columns sub-area; the pixel filtering parameters are used to characterize the size of the imaging size of the target of interest in each sub-area; obtain the image to be processed taken of the target area; divide the image to be processed into sub-areas of M rows and N columns; determine the target sub-area where the candidate target is located from the image to be processed; if the size of the imaging size of the candidate target in the target sub-area does not match the target pixel filtering parameter corresponding to the target sub-area, the candidate target is filtered as a non-target of interest.

[0159] In some other possible embodiments, the processing module 1002 is further used to filter the candidate target as a non-target of interest if the corresponding pixel filtering parameters have not been learned for the target sub-region and the imaging size of the candidate target in the target sub-region does not match the reference pixel filtering parameters of the target sub-region; wherein the reference pixel filtering parameters of the target sub-region are determined based on the pixel filtering parameters corresponding to the sub-region in the same row as the target sub-region in the M rows and N columns of the image to be processed.

[0160] In some other possible embodiments, for the first sub-region, the first sub-region is any sub-region in the M rows and N columns of sub-regions, and the processing module 1002 is specifically used to record a set of pixel statistics when each target of interest appears in the first sub-region in the learning sample image, and obtain multiple sets of pixel statistics corresponding to each of the multiple targets of interest appearing in the first sub-region; the pixel statistics include the imaging height of the corresponding target of interest in the first sub-region image; based on the multiple sets of pixel statistics, determine multiple height ratio parameters corresponding to each of the multiple sets of pixel statistics; the height ratio parameter is the ratio of the imaging height of the target of interest in the first sub-region image to a preset reference height; calculate the average value and standard deviation of the multiple height ratio parameters; use the difference between the average value and three times the standard deviation as the lower limit, and the sum of the average value and three times the standard deviation as the upper limit to obtain a confidence interval as the pixel filtering parameter corresponding to the first sub-region; or use the difference between the average value and X times the standard deviation as the lower limit, and the sum of the average value and X times the standard deviation as the upper limit to obtain a confidence interval as the pixel filtering parameter corresponding to the first sub-region; X is a natural number, and X is not 3.

[0161] In some other possible embodiments, the processing module 1002 is specifically used to obtain the size interval of the target of interest set by the user; convert the target pixel filtering parameter into a reference size interval according to a preset conversion relationship; when there is no intersection between the size interval set by the user and the reference size interval, obtain the first imaging height of the candidate target in the first sub-area image; determine the reference height ratio parameter according to the ratio of the first imaging height to the preset reference height; if the reference height ratio parameter is within the confidence interval of the target pixel filtering parameter, retain the candidate target as the target of interest; if the reference height ratio parameter is outside the confidence interval of the target pixel filtering parameter, filter the candidate target as a non-target of interest.

[0162] It should be noted that the module division shown in FIG10 is schematic and represents only one logical functional division. In actual implementation, other divisions may be employed. For example, two or more functions may be integrated into a single processing module. Such integrated modules may be implemented in either hardware or software functional modules.

[0163] In an exemplary embodiment, the present application also provides a computer-readable storage medium including computer-executable instructions, which, when executed on an electronic device, enables the electronic device to execute any one of the methods provided in the above embodiments.

[0164] In an exemplary embodiment, the present application also provides a computer program product including computer-executable instructions, which, when executed on an electronic device, enables the electronic device to execute any one of the methods provided in the above embodiments.

[0165] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware, or any combination thereof. When implemented using a software program, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer-executable instructions. When the computer-executable instructions are loaded and executed on a computer, the process or function according to the embodiments of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer-executable instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer-executable instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), or an optical medium (e.g., a DVD).

[0166] Although the present application is described herein in conjunction with various embodiments, in the process of implementing the claimed application, those skilled in the art may understand and implement other variations of the disclosed embodiments by reviewing the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other components or steps, and "one" or "an" does not exclude multiple components. A single processor or other unit may implement several functions listed in the claims. Certain measures are recorded in different dependent claims, but this does not mean that these measures cannot be combined to produce good results.

[0167] Although the present application has been described with reference to specific features and embodiments thereof, it is apparent that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and the drawings are merely illustrative of the present application as defined by the appended claims and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, those skilled in the art may make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, the present application is intended to include such modifications and variations as fall within the scope of the claims of the present application and their equivalents.

[0168] The above is only a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or replacements within the technical scope disclosed in the present application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A method for filtering non-interested targets, characterized in that: The method comprises: Acquire a learning sample image; the learning sample image is an image captured for a target area; the learning sample image includes an object of interest; Determining pixel filtering parameters based on the learning sample image; the pixel filtering parameters are used to characterize the imaging size of the target of interest at different positions in the target area; Based on the pixel filtering parameters, non-interested objects in the target area in the image to be processed are filtered; the image to be processed is an image shot for the target area.

2. The method according to claim 1, characterized in that The learning sample images include daytime images and / or nighttime images; the daytime images include visible light images and thermal imaging images; and the nighttime images include thermal imaging images.

3. The method according to claim 1, characterized in that The acquiring of the learning sample image comprises: In daytime, obtaining a visible light image and a thermal imaging image taken of the target area; If the same target of interest is detected in the visible light image and the thermal imaging image, determining the visible light image and the thermal imaging image as the learning sample images; Acquiring a thermal imaging image of the target area in darkness; Detecting the thermal imaging image using a preset target detection algorithm; If the target detection algorithm detects a target in the thermal imaging image, and the probability that the target is a target of interest is higher than a probability threshold, the thermal imaging image is determined as the learning sample image.

4. The method according to claim 1, wherein The step of determining pixel filtering parameters according to the learning sample image includes: Divide the learning sample image into sub-regions of M rows and N columns; M and N are both positive integers greater than 1; Determine, based on the learning sample image, pixel filtering parameters corresponding to each sub-region of the M rows and N columns of the object of interest; the pixel filtering parameters are used to characterize the imaging size of the object of interest in each sub-region; The filtering of non-interested objects within the target area based on the pixel filtering parameter includes: Acquire an image to be processed that is taken for the target area; Dividing the image to be processed into sub-regions of M rows and N columns; Determine a target subregion where a candidate target is located from the image to be processed; If the imaging size of the candidate target in the target sub-region does not match the target pixel filtering parameter corresponding to the target sub-region, the candidate target is filtered as a non-interested target.

5. The method according to claim 4, characterized in that The method further comprises: If the corresponding pixel filtering parameters are not learned for the target sub-region, and the imaging size of the candidate target in the target sub-region does not match the reference pixel filtering parameters of the target sub-region, the candidate target is filtered as a non-interested target; The reference pixel filtering parameters of the target sub-region are determined based on the pixel filtering parameters corresponding to the sub-region in the same row as the target sub-region in the M-row and N-column sub-regions of the image to be processed.

6. The method according to claim 4, characterized in that For a first subregion, the first subregion is any subregion in the M rows and N columns of subregions; determining, based on the learning sample image, pixel filtering parameters corresponding to each subregion of the object of interest in the M rows and N columns of subregions includes: When each object of interest in the learning sample image appears in the first sub-region, a set of pixel statistics is recorded, thereby obtaining a plurality of sets of pixel statistics corresponding to each of the plurality of objects of interest appearing in the first sub-region; the pixel statistics include an imaging height of the corresponding object of interest in the image of the first sub-region; Determining, based on the multiple sets of pixel statistical data, multiple height ratio parameters corresponding to each of the multiple sets of pixel statistical data; the height ratio parameters are ratios of an imaging height of an object of interest in the first sub-region image to a preset reference height; Calculating the mean and standard deviation of the plurality of height scale parameters; Taking the difference between the average value and three times the standard deviation as a lower limit and the sum of the average value and three times the standard deviation as an upper limit, a confidence interval is obtained as the pixel filtering parameter corresponding to the first sub-region; or, The difference between the average value and X times the standard deviation is used as a lower limit, and the sum of the average value and X times the standard deviation is used as an upper limit to obtain a confidence interval as the pixel filtering parameter corresponding to the first sub-region; X is a natural number, and X is not 3.

7. The method according to claim 6, characterized in that Based on the target pixel filtering parameters corresponding to the target sub-region, it is determined whether to filter the candidate target as a non-interested target in the following manner: Get the size range of the target of interest set by the user; Converting the target pixel filtering parameter into a reference size interval according to a preset conversion relationship; When the size interval set by the user does not intersect with the reference size interval, obtaining a first imaging height of the candidate object in the first sub-region image; determining a reference height ratio parameter according to a ratio of the first imaging height to the preset reference height; If the reference height ratio parameter is within the confidence interval of the target pixel filtering parameter, retaining the candidate target as a target of interest; If the reference height ratio parameter is outside the confidence interval of the target pixel filtering parameter, the candidate target is filtered as a non-target of interest.

8. A non-interest target filtering device, characterized in that: The device comprises: an acquisition module and a processing module; The acquisition module is used to acquire a learning sample image; the learning sample image includes an image of an object of interest in a target area; The processing module is used to determine pixel filtering parameters based on the learning sample image; the pixel filtering parameters are used to characterize the size of the imaging size of the target of interest at different positions in the target area; based on the pixel filtering parameters, non-targets of interest in the target area in the image to be processed are filtered; the image to be processed is an image taken of the target area.

9. An electronic device, characterized in that: include: processor and memory; The memory stores instructions executable by the processor; When the processor is configured to execute the instructions, the electronic device implements the method according to any one of claims 1 to 7.

10. A readable storage medium, characterized in that: include: Software instructions; When the software instructions are executed in an electronic device, the electronic device is enabled to implement the method according to any one of claims 1 to 7.

Images