Camera device adjustment method, device, electronic device and computer readable medium

Abstract

The embodiments of the present disclosure disclose a camera device adjustment method, device, electronic device and computer-readable medium. A specific implementation of the method includes: acquiring a first captured image; in response to receiving replacement information, determining to replace the camera device as a second camera device; for the second camera device, executing a device adjustment step: capturing a second captured image; generating image difference information; determining a device position relationship; generating device adjustment parameter information; in response to determining that the device adjustment parameter information does not meet the target adjustment condition, generating adjustment information; in response to determining that the target adjustment condition is met, performing device adjustment on the second camera device to obtain an adjusted second camera device; using the adjusted second camera device as the second camera device, executing a device adjustment step. This implementation can accurately and efficiently achieve the replacement of the camera device, so that the area image captured by the replacement camera device is consistent with the area image captured by the original camera device.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of computer technology, and in particular to a camera device adjustment method, device, electronic device, and computer-readable medium. Background Art

[0002] At present, using a camera to monitor an area has become one of the main monitoring methods. For shooting a target area, the camera is usually adjusted by manually adjusting the position and camera parameters of the camera so that the image captured by the adjusted camera is roughly consistent with the image captured by the original camera.

[0003] However, when the above method is used to adjust the camera device, the following technical problems often occur:

[0004] First, manual adjustment is inefficient and cannot effectively ensure that the images captured by the adjusted camera device are completely consistent with the images captured by the original camera device, which may lead to inconsistent shooting focus.

[0005] Second, it is crucial to accurately generate device adjustment parameters corresponding to the second shooting device according to the image difference and the positional relationship between the camera devices. The accurate generation of the device adjustment parameters determines the adjustment accuracy and adjustment times of the second shooting device.

[0006] The above information disclosed in this Background section is only for enhancement of understanding of the background of the inventive concept and therefore it may contain information that does not form the prior art that is already known in this country to a person of ordinary skill in the art. Summary of the invention

[0007] The content of this disclosure is used to introduce concepts in a brief form, which will be described in detail in the detailed implementation section below. The content of this disclosure is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.

[0008] Some embodiments of the present disclosure provide a camera device adjustment method, device, electronic device, and computer-readable medium to solve one or more of the technical problems mentioned in the above background technology section.

[0009] In a first aspect, some embodiments of the present disclosure provide a camera device adjustment method, comprising: acquiring a first captured image of a target shooting area captured by a first camera device; in response to receiving replacement information for the first camera device, determining a replacement camera device corresponding to the first camera device as a second camera device; performing the following device adjustment steps for the second camera device: using the second camera device to capture a second captured image of the target shooting area; generating image difference information for the first captured image and the second captured image; determining a device position relationship between the first camera device and the second camera device; generating device adjustment parameter information for the second camera device based on the image difference information and the device position relationship; in response to determining that the device adjustment parameter information does not meet the target adjustment condition, generating adjustment information indicating that the second camera device is not adjusted; in response to determining that the device adjustment parameter information meets the target adjustment condition, adjusting the second camera device based on the device adjustment parameter information to obtain an adjusted second camera device; using the adjusted second camera device as the second camera device, and continuing to perform the device adjustment steps.

[0010] In a second aspect, some embodiments of the present disclosure provide a camera device adjustment device, comprising: an acquisition unit, configured to acquire a first captured image of a target captured area captured by a first camera device; a determination unit, configured to determine, in response to receiving replacement information for the first camera device, a replacement camera device corresponding to the first camera device as a second camera device; a first execution unit, configured to execute the following device adjustment steps for the second camera device: using the second camera device to capture a second captured image of the target captured area; generating image difference information for the first captured image and the second captured image; determining the replacement camera device corresponding to the first camera device as a second camera device. a device position relationship between a first camera device and a second camera device; generating device adjustment parameter information for the second camera device based on picture difference information and the device position relationship; in response to determining that the device adjustment parameter information does not satisfy a target adjustment condition, generating adjustment information indicating that the second camera device is not to be adjusted; a device adjustment unit, configured to, in response to determining that the device adjustment parameter information satisfies the target adjustment condition, perform device adjustment on the second camera device according to the device adjustment parameter information to obtain an adjusted second camera device; a second execution unit, configured to use the adjusted second camera device as the second camera device and continue to execute the device adjustment step.

[0011] In a third aspect, some embodiments of the present disclosure provide an electronic device comprising: one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner in the first aspect.

[0012] In a fourth aspect, some embodiments of the present disclosure provide a computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the method described in any implementation manner in the first aspect is implemented.

[0013] The above-mentioned various embodiments of the present disclosure have the following beneficial effects: through the camera device adjustment method of some embodiments of the present disclosure, the camera device can be replaced accurately and efficiently, so that the regional picture taken by the replacement camera device is consistent with the regional picture taken by the original camera device. Specifically, the reason why the relevant camera device is not accurate and efficient is that the manual adjustment efficiency is low, and it cannot effectively guarantee that the picture taken by the adjusted camera device is completely consistent with the picture content taken by the original camera device, and the problem of inconsistent shooting focus may occur. Based on this, the camera device adjustment method of some embodiments of the present disclosure first obtains the first shooting picture taken by the first camera device for the target shooting area as the real comparison picture of the subsequent replacement camera device, so as to facilitate the subsequent replacement of the camera device. Then, in response to receiving the replacement information for the above-mentioned first camera device, the replacement camera device corresponding to the above-mentioned first camera device is determined as the second camera device, so as to facilitate the subsequent adjustment and replacement of the second camera device. Secondly, for the second camera device, the following device adjustment steps are performed: the first step is to use the second camera device to shoot the second shooting picture for the above-mentioned target shooting area, so as to compare the second shooting picture with the first shooting picture later, so as to accurately adjust the camera device and ensure that the regional picture shot by the replacement camera device is consistent with the regional picture shot by the original camera device. The second step is to generate the picture difference information for the above-mentioned first shooting picture and the second shooting picture, so as to consider how to adjust the device parameters of the second camera device from the perspective of the picture difference later. The third step is to determine the device position relationship between the above-mentioned first camera device and the second camera device, so as to consider how to adjust the device parameters of the second camera device from the perspective of the device position relationship later. The fourth step is to consider the influencing factors from various aspects according to the picture difference information and the device position relationship, so as to accurately generate the device adjustment parameter information for the second camera device. The fifth step is to generate the adjustment information indicating that the second camera device is not adjusted in response to determining that the device adjustment parameter information does not meet the target adjustment condition. Here, the target adjustment condition can effectively constrain the adjustment force of the second camera device to avoid violating the device adjustment limit of the second camera device for the sake of accuracy. Further, in response to determining that the above-mentioned device adjustment parameter information meets the above-mentioned target adjustment condition, the second camera device is adjusted according to the device adjustment parameter information to obtain the adjusted second camera device. Finally, the adjusted second camera device is used as the second camera device, and the above-mentioned device adjustment steps are continued. In summary, the second camera device is cyclically adjusted through the angle of the picture difference and the angle of the device position relationship to achieve the replacement of the camera device, so that the area picture taken by the replacement camera device is consistent with the area picture taken by the original camera device. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that components and elements are not necessarily drawn to scale.

[0015] Figure 1 is a flow chart of some embodiments of the camera device adjustment method according to the present disclosure;

[0016] Figure 2 is a schematic structural diagram of some embodiments of the camera device adjustment device according to the present disclosure;

[0017] Figure 3 It is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0018] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0019] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present disclosure can be combined with each other.

[0020] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0021] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0022] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0023] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0024] refer to Figure 1, shows a process 100 of some embodiments of the camera device adjustment method according to the present disclosure. The camera device adjustment method comprises the following steps:

[0025] Step 101: Acquire a first captured image of a target capturing area captured by a first camera device.

[0026] In some embodiments, the execution subject (e.g., electronic device) of the above-mentioned camera device adjustment method can obtain the first captured image of the target shooting area captured by the first camera device through a wired connection or a wireless connection. The first camera device can be a camera device that originally captured the image of the target shooting area. That is, the first camera device can be a camera device that has not been replaced before. The target shooting area can be the shooting area of ​​the camera device.

[0027] Step 102: In response to receiving replacement information for the first camera device, determining a replacement camera device corresponding to the first camera device as a second camera device.

[0028] In some embodiments, in response to receiving replacement information for the first camera device, the execution subject may determine a replacement camera device corresponding to the first camera device as the second camera device. The replacement information may be a camera device replacement request. For example, the replacement information may be a request information to replace the first camera device with the second camera device. The second camera device may be a camera device to be used for photographing the target photographing area.

[0029] Step 103: For the second camera device, perform the following device adjustment steps:

[0030] Step 1031: Use a second camera device to capture a second picture of the target shooting area.

[0031] In some embodiments, the execution entity may utilize a second camera device to capture a second picture of the target shooting area.

[0032] Step 1032: Generate image difference information for the first captured image and the second captured image.

[0033] In some embodiments, the execution subject may generate image difference information for the first shot image and the second shot image, wherein the image difference information may represent the content difference between the image content corresponding to the first shot image and the image content corresponding to the second shot image.

[0034] In some optional implementations of some embodiments, the generating of the picture difference information for the first shot picture and the second shot picture may include the following steps:

[0035] The first step is to determine the initial picture difference information between the first shot picture and the second shot picture using target image processing software. The target image processing software may be OpenCV image processing software. The initial picture difference information may be pixel difference information between the first shot picture and the second shot picture.

[0036] In the second step, in response to determining that the initial picture difference information satisfies the pixel difference regularity condition, the initial picture difference information is determined as the picture difference information. The pixel difference regularity condition may be the information of the difference between the first shot picture and the second shot picture corresponding to the partial area, indicating that the pixel difference information is the misaligned pixel difference between the first shot picture and the second shot picture. The area ratio between the partial area and the target shot area is greater than the target value. The pixel difference regularity condition may be satisfied to indicate that the first shot picture and the second shot picture have content deviations between the two pictures.

[0037] In the third step, in response to determining that the initial picture difference information does not satisfy the pixel difference regularity condition, the first shot picture is divided into a predetermined ratio to generate a first shot sub-picture set, and the second shot picture is divided into a predetermined ratio to generate a second shot sub-picture set. For example, the predetermined ratio may be divided into 9 equal ratios.

[0038] In the fourth step, the first captured sub-picture in the first captured sub-picture set and the second captured sub-picture in the second captured sub-picture set are combined according to the segmentation sequence to generate a captured sub-picture group set. The segmentation regions corresponding to the two captured sub-pictures in the captured sub-picture group set are in the same position. The segmentation sequence may be a segmentation sequence of a predetermined ratio.

[0039] Step 5: for each captured sub-picture group in the captured sub-picture group set, perform the following first generation step:

[0040] Sub-step 1, input each captured sub-picture in the above-mentioned captured sub-picture group into a pre-trained picture feature information generation model to generate first picture feature information and second picture feature information. The picture feature information generation model can be a neural network model for generating picture feature information. In practice, the picture feature information generation model can be a convolutional neural network model with multiple layers connected in series. The captured sub-picture group includes: a first captured sub-picture for the first captured picture and a second captured sub-picture for the second captured picture. The first picture feature information is information that characterizes the picture feature semantics corresponding to the first captured sub-picture. The second picture feature information is information that characterizes the picture feature semantics corresponding to the second captured sub-picture.

[0041] Sub-step 2, respectively inputting the first picture feature information and the second picture feature information into a picture information recognition model based on feature points and feature objects, so as to generate a first feature point information set and a first feature object information set for the first picture feature information, and a second feature point information set and a second feature object information set for the second picture feature information. The first feature point information may be the name information and position information corresponding to the first feature point in the first captured sub-picture. The second feature point information may be the name information and position information corresponding to the second feature point in the second captured sub-picture. The first feature object information may be the name information and position information corresponding to the first feature object in the first captured sub-picture. The second feature object information may be the name information and position information corresponding to the second feature object in the second captured sub-picture. The first feature object and the second feature object may be objects in the picture. The picture information recognition model based on feature points and feature objects may be a neural network model that generates feature point information corresponding to each feature point in the picture and feature object information corresponding to each feature object.

[0042] Sub-step 3, generating captured sub-picture difference information based on the first feature point information set, the first feature object information set, the second feature point information set and the second feature object information set.

[0043] As an example, first, the execution subject may determine a first information difference between the first feature point information in the first feature point information set and the second feature point information in the second feature point information set. Then, determine a second information difference between the first feature object information in the first feature object information set and the second feature object information in the second feature object information set. Finally, determine the first information difference and the second information difference as the captured sub-picture difference information.

[0044] The fifth step is to generate the above picture difference information according to the obtained captured sub-picture difference information set.

[0045] As an example, the execution subject may directly determine the captured sub-picture difference information set as the picture difference information.

[0046] Optionally, the above-mentioned picture information recognition model based on feature points and feature objects includes: an information recognition model based on feature points and an information recognition model based on feature objects, wherein the above-mentioned information recognition model based on feature points includes: a feature depth extraction model and a feature point information output layer, and the above-mentioned information recognition model based on feature objects includes: the above-mentioned feature depth extraction model and a feature object information output layer. The information recognition model based on feature points can be a neural network model that generates feature point information. Feature point information can be information for identifying feature point names and feature point positions. The information recognition model based on feature objects can be a neural network model that generates feature object information. Feature object information can be information for identifying feature object names and feature object positions. The feature depth extraction model can be a neural network model that further extracts feature information (depth feature information). For example, the feature depth extraction model can be a 21-layer convolutional neural network model in series. The feature object information output layer can be a neural network model that outputs feature object information. In practice, the feature object information output layer can be a convolutional network model with a first number of layers + a fully connected layer with a second number of layers. The feature point information output layer can be a neural network model that outputs feature point information. In practice, the feature point information output layer can be a third number of layers of convolutional network model + a fourth number of layers of fully connected layers.

[0047] Optionally, the step of inputting the first picture feature information and the second picture feature information into a picture information recognition model based on feature points and feature objects, respectively, to generate a first feature point information set and a first feature object information set for the first picture feature information, and a second feature point information set and a second feature object information set for the second picture feature information, may include the following steps:

[0048] In the first step, the first picture feature information and the second picture feature information are input into the feature depth extraction model to generate the first picture feature depth extraction information and the second picture feature depth extraction information. The first picture feature depth extraction information can represent the deep feature information corresponding to the first picture feature information. The second picture feature depth extraction information can represent the deep feature information corresponding to the second picture feature information.

[0049] In the second step, the first picture feature depth extraction information and the second picture feature depth extraction information are input into the feature point information output layer to output the first feature point information set and the second feature point information set.

[0050] In the third step, the first picture feature depth extraction information and the second picture feature depth extraction information are input into the feature object information output layer to output the first feature object information set and the second feature object information set.

[0051] Optionally, the above-mentioned picture information recognition model based on feature points and feature objects is trained by the following steps:

[0052] The first step is to obtain a training data set. The training data includes: a first picture, a second picture, a first label feature point information set, a second label feature point information set, a first label feature object information set, and a second label feature object information set. The first label feature point information set can represent the actual label feature point information set in the first picture. The second label feature point information set can represent the actual label feature point information set in the second picture. The first label feature object information set can represent the actual label feature object information set in the first picture. The second label feature object information set can represent the actual label feature object information set in the second picture.

[0053] The second step is to select target training data from the training data set.

[0054] In the third step, for the target training data, perform the following training steps:

[0055] Sub-step 1: determine a first picture and a second picture corresponding to the target training data as a first target picture and a second target picture, respectively.

[0056] Sub-step 2: generating first preliminary picture feature information for the first target picture and second preliminary picture feature information for the second target picture.

[0057] Sub-step 3: input the first preliminary picture feature information and the second preliminary picture feature information into an initial feature depth extraction model included in an initial picture information recognition model to generate first depth picture feature information and second depth picture feature information. The initial picture information recognition model may be a picture information recognition model that has not yet been trained.

[0058] Sub-step 4: input the first depth picture feature information and the second depth picture feature information into an initial feature point information output layer included in an initial picture information recognition model to generate a first initial feature point information set and a second initial feature point information set.

[0059] Sub-step 5: inputting the first depth picture feature information and the second depth picture feature information into an initial feature object information output layer included in an initial picture information recognition model to generate a first initial feature object information set and a second initial feature object information set.

[0060] Sub-step 6: generating first feature point loss information based on the first initial feature point information set and the first label feature point information set.

[0061] As an example, the execution entity may generate first feature point loss information based on the first initial feature point information set and the first label feature point information set using a cross entropy loss function.

[0062] Sub-step 7: generating second feature point loss information based on the second initial feature point information set and the second label feature point information set.

[0063] As an example, the execution entity may generate second feature point loss information based on the second initial feature point information set and the second label feature point information set using a cross entropy loss function.

[0064] Sub-step 8, generating first feature object loss information based on the first initial feature object information set and the first label feature object information set.

[0065] As an example, the execution entity may generate first feature object loss information based on the first initial feature object information set and the first label feature object information set using a cross entropy loss function.

[0066] Sub-step 9, generating second feature object loss information based on the second initial feature object information set and the third label feature object information set.

[0067] As an example, the execution entity may generate second feature object loss information based on the second initial feature object information set and the second label feature object information set using a cross entropy loss function.

[0068] Sub-step 10, based on the first initial feature point information set and the first initial feature object information set, using the point-surface different position loss function, generates the first point-surface loss information. The second point-surface loss information may be loss information indicating whether the second initial feature point information set overlaps with the second initial feature object information set. The point-surface different position loss function may be an IoU Loss loss function.

[0069] As an example, the execution entity may use an IoU Loss loss function to generate first point-surface loss information according to the first initial feature point information set and the first initial feature object information set.

[0070] Sub-step 11, based on the second initial feature point information set and the second initial feature object information set, using the point and surface different position loss function, generates second point and surface loss information.

[0071] As an example, the execution entity may use the IoU Loss loss function to generate the second point-surface loss information according to the second initial feature point information set and the second initial feature object information set.

[0072] Sub-step 12, generating total loss information based on the first feature point loss information, the second feature point loss information, the first feature object loss information, the second feature object loss information, the first point-surface loss information and the second point-surface loss information.

[0073] As an example, the above-mentioned execution entity can perform weighted summation processing on the first feature point loss information, the second feature point loss information, the first feature object loss information, the second feature object loss information, the first point-surface loss information and the second point-surface loss information to generate weighted loss information as total loss information.

[0074] Sub-step 13, in response to determining that the total loss information is less than a fifth value, determining the initial picture information recognition model as the picture information recognition model, wherein the fifth value may be a predetermined value.

[0075] In the fourth step, in response to determining that the total loss information is greater than or equal to the fifth value, the model parameters of the initial picture information recognition model are updated according to the total loss information to generate an updated picture information recognition model.

[0076] As an example, the execution subject may update the model parameters of the initial picture information recognition model by back propagation to generate an updated picture information recognition model.

[0077] The fifth step is to remove the target training data from the above training data set to obtain the removed training data set.

[0078] Step 6: Reselect training data from the above removed training data set as candidate training data.

[0079] In the seventh step, the candidate training data is used as the target training data, the updated picture information recognition model is used as the initial picture information recognition model, and the above model training steps are continued.

[0080] In some optional implementations of some embodiments, generating the picture difference information according to the obtained captured sub-picture difference information set may include the following steps:

[0081] The first step is to determine the difference ratio information of the sub-picture content corresponding to each sub-picture difference information in the sub-picture difference information set. The sub-picture content difference ratio information may represent the difference ratio between the sub-picture content in the first sub-picture and the sub-picture content in the second sub-picture. For example, the sub-picture content difference ratio information may be 0.8. The larger the sub-picture content difference ratio, the greater the difference in the sub-picture content corresponding to the two sub-pictures.

[0082] The second step is to determine at least one of the obtained picture essential content difference ratio information sets corresponding to the picture essential content difference ratio information greater than a first value, wherein the first value may be a preset value, for example, 50%.

[0083] The third step is to determine the information ratio corresponding to the at least one information about the substantial content difference ratio of the above-mentioned picture. The information ratio can be the ratio of the number of information corresponding to the at least one information about the substantial content difference ratio of the picture divided by the number of information corresponding to the sub-picture difference information set. For example, if the number of information corresponding to the at least one information about the substantial content difference ratio of the picture is 10, and the number of information corresponding to the sub-picture difference information set of the picture substantial content difference ratio information set is 100, then the information ratio can be 0.1.

[0084] In step 4, in response to determining that the above information ratio is less than a second value, at least one picture substantial content difference ratio information is removed from the above picture substantial content difference ratio information set to obtain a picture substantial content difference ratio information subset after removal. The second value may be a predetermined value. For example, the second value may be 0.6.

[0085] The fifth step is to generate picture difference information according to the first captured sub-picture difference information subset corresponding to the above-mentioned picture substantial content difference ratio information subset after removal.

[0086] As an example, the execution subject may determine the information summary information corresponding to the first captured sub-picture difference information subset as the picture difference information.

[0087] Optionally, the generating the picture difference information according to the obtained captured sub-picture difference information set further includes:

[0088] In the first step, in response to determining that the information ratio is greater than or equal to the second value and less than a third value, determining whether there is an information proximity relationship between the second captured sub-picture difference information subsets corresponding to the sub-picture substantial content difference ratio information subset after removal, wherein the third value is greater than the second value.

[0089] In a second step, in response to determining that at least two second captured sub-picture difference information in the second captured sub-picture difference information subset are in an information proximity relationship, image difference information is generated according to at least two second captured sub-picture difference information corresponding to the at least two second captured sub-picture difference information. The information proximity relationship may be that image positions corresponding to the two captured sub-picture sets corresponding to the at least two second captured sub-picture difference information are in a position proximity relationship.

[0090] As an example, the execution subject may directly determine at least two pieces of second captured sub-picture difference information as the picture difference information.

[0091] In the third step, in response to determining that the above information ratio is greater than the above third value, based on the above first shot picture and the second shot picture, the picture difference information generation model is used to generate picture difference information. The picture difference information generation model can be a neural network model for generating picture difference information. In practice, the picture difference information generation model can be a multi-layer series convolution model.

[0092] Optionally, the steps further include:

[0093] The first step is to determine the same pixel area pictures for the first shooting picture and the second shooting picture, and obtain a pixel area picture group set. The pixel area group includes: at least one first pixel area picture in the first shooting picture and at least one second pixel area picture in the second shooting picture. The pixel similarity corresponding to each first pixel area picture in the at least one first pixel area picture and each second pixel area picture in the at least one second pixel area picture is higher than a fourth value. The same pixel area picture can be a picture of the area with the same pixels between the first shooting picture and the second shooting picture. The fourth value can be a preset value. For example, the fourth value can be 75%.

[0094] In the second step, for each pixel area picture group in the above pixel area picture group set, the following second generation step is performed:

[0095] Sub-step 1: determining at least one first pixel area picture and at least one second pixel area picture included in the above pixel area picture group as at least one first target pixel area picture and at least one second target pixel area picture, respectively.

[0096] Sub-step 2: determining at least one first picture position information corresponding to the at least one first target pixel region picture and at least one second picture position information corresponding to the at least one second target pixel region picture.

[0097] Sub-step 3, inputting the at least one first target pixel area picture, the at least one first picture position information, the at least one second target pixel area picture and the at least one second picture position information into the picture difference information generation model to generate picture difference sub-information. In practice, the picture difference information generation model can be a multi-head attention mechanism model based on a convolutional neural network.

[0098] The third step is to generate the above picture difference information according to the obtained picture difference sub-information set.

[0099] As an example, the above execution subject may directly determine the picture difference sub-information set as the picture difference information.

[0100] Step 1033: Determine the device position relationship between the first camera device and the second camera device.

[0101] In some embodiments, the execution subject may determine the device position relationship between the first camera device and the second camera device, wherein the device position relationship may represent the relative position relationship between the corresponding placement position of the first camera device and the corresponding placement position of the second camera device.

[0102] Step 1034: Generate device adjustment parameter information for the second camera device according to the picture difference information and the device position relationship.

[0103] In some embodiments, the execution subject may generate device adjustment parameter information for the second camera device based on the image difference information and the device position relationship. The device adjustment parameter may represent parameter information for device adjustment of the second camera device. In practice, the device adjustment parameter information may include but is not limited to at least one of the following: device position adjustment parameter, device shooting parameter information. The device shooting parameter information may include: device aperture and device focal length.

[0104] In some optional implementations of some embodiments, the generating of the device adjustment parameter information for the second camera device according to the picture difference information and the device position relationship includes:

[0105] The first step is to determine the first device parameter information corresponding to the first camera device and the second device parameter information corresponding to the second camera device.

[0106] The second step is to determine parameter difference information between the first device parameter information and the second device parameter information.

[0107] The third step is to determine the pixel change information corresponding to the above picture difference information, wherein the pixel change information may be difference information representing the picture pixel difference corresponding to the picture difference information.

[0108] In the fourth step, the pixel change information, the device position relationship and the parameter difference information are input into a pre-trained device adjustment parameter information generation model to generate device adjustment parameter information for the second camera device. The device adjustment parameter information generation model may be a neural network model for generating device adjustment parameter information. In practice, the device adjustment parameter information generation model may be a residual model with multiple layers connected in series.

[0109] In some optional implementations of some embodiments, the device adjustment parameter information generation model includes: a first adjustment parameter information generation model based on pixel change information, a second adjustment parameter information generation model based on device position relationship, a third adjustment parameter information generation model based on parameter difference information, a feature information summary model and an adjustment parameter information generation layer. Among them. The first adjustment parameter information generation model based on pixel change information may be a neural network model that outputs first adjustment parameter feature information for pixel change information. The first adjustment parameter feature information may characterize the semantics of information corresponding to the adjustment parameter information. In practice, the first adjustment parameter information generation model may be a convolutional layer of the first number of layers in series. The second adjustment parameter information generation model based on device position relationship may be a neural network model that outputs second adjustment parameter feature information for device position relationship. The second adjustment parameter feature information may characterize the semantics of information corresponding to the second adjustment parameter information. In practice, the second adjustment parameter information generation model may be a convolutional layer of the second number of layers in series. The third adjustment parameter information generation model based on parameter difference information may be a neural network model that outputs third adjustment parameter feature information for parameter difference information. The third adjustment parameter feature information may characterize the semantics of information corresponding to the second adjustment parameter information. In practice, the third adjustment parameter information generation model may be a convolutional layer of the third number of layers in series. The feature information aggregation model may be a neural network model that aggregates the first adjustment parameter feature information, the second adjustment parameter feature information, and the third adjustment parameter feature information. In practice, the feature information aggregation model may be an attention mechanism model based on a convolutional model. The adjustment parameter information generation layer may be a fully connected layer that outputs the adjustment parameter information.

[0110] Optionally, the pixel change information, the device position relationship and the parameter difference information are input into a pre-trained device adjustment parameter information generation model to generate device adjustment parameter information for the second camera device, comprising the following steps:

[0111] In the first step, the pixel change information is input into a first adjustment parameter information generation model to generate first adjustment parameter feature information.

[0112] In the second step, the device position relationship is input into a second adjustment parameter information generation model to generate second adjustment parameter characteristic information.

[0113] The third step is to input the parameter difference information into the third adjustment parameter information generation model to generate the third adjustment parameter characteristic information.

[0114] In a fourth step, the first adjustment parameter feature information, the second adjustment parameter feature information and the third adjustment parameter feature information are input into a feature information summary model to generate feature summary information.

[0115] In the fifth step, the feature summary information is input into the adjustment parameter information generation layer to generate the adjustment parameter information.

[0116] Optionally, the device adjustment parameter information generation model can be trained by the following steps:

[0117] The first step is to obtain a preset training data set, wherein the preset training data includes: a first initial picture, a second initial picture, target pixel change information for the first initial picture and the second initial picture, a target device position relationship for the first initial picture and the second initial picture, target parameter difference information for the first initial picture and the second initial picture, first adjustment parameter label information for the first initial picture and the second initial picture, second adjustment parameter label information for the first initial picture and the second initial picture, third adjustment parameter label information for the first initial picture and the second initial picture, and actual adjustment parameter label information.

[0118] The second step is to select target preset training data from the preset training data set.

[0119] In the third step, for the target preset training data, perform the following model training steps:

[0120] Sub-step 1: inputting the target pixel change information into an initial first adjustment parameter information generation model included in the initial device adjustment parameter information generation model to generate first initial adjustment parameter feature information.

[0121] Sub-step 2: inputting the target device position relationship into an initial second adjustment parameter information generation model included in the initial device adjustment parameter information generation model to generate second initial adjustment parameter characteristic information.

[0122] Sub-step 3: inputting the target parameter difference information into an initial third adjustment parameter information generation model included in the initial device adjustment parameter information generation model to generate third initial adjustment parameter characteristic information.

[0123] Sub-step 4, inputting the first initial adjustment parameter feature information into the first adjustment parameter information generation model to generate the first adjustment parameter information for the first initial picture and the second initial picture. The first adjustment parameter information generation model may be a model that generates corresponding adjustment parameter information for pixel change information. For example, the first adjustment parameter information generation model may be a convolutional neural network model.

[0124] Sub-step 5, inputting the second initial adjustment parameter feature information into the second adjustment parameter information generation model to generate the second adjustment parameter information for the first initial screen and the second initial screen. The second adjustment parameter information generation model may be a model that generates corresponding adjustment parameter information for the position relationship of the target device. For example, the second adjustment parameter information generation model may be a convolutional neural network model.

[0125] Sub-step 6, inputting the third initial adjustment parameter feature information into a third adjustment parameter information generation model to generate third adjustment parameter information for the first initial screen and the second initial screen. The third adjustment parameter information generation model may be a model that generates corresponding adjustment parameter information for target parameter difference information. For example, the third adjustment parameter information generation model may be a convolutional neural network model.

[0126] Sub-step 7: input the first initial adjustment parameter characteristic information, the second initial adjustment parameter characteristic information and the third initial adjustment parameter characteristic information into the initial characteristic information summary model included in the initial device adjustment parameter information generation model to generate initial characteristic summary information.

[0127] Sub-step 8: Generate first model loss information between the first adjustment parameter information and the above-mentioned first adjustment parameter label information.

[0128] Sub-step 9, generating second model loss information between the second adjustment parameter information and the above-mentioned second adjustment parameter label information.

[0129] Sub-step 10: generating third model loss information between the third adjustment parameter information and the third adjustment parameter label information.

[0130] Sub-step 11, inputting the initial feature summary information into the initial adjustment parameter information generation layer to generate initial adjustment parameter information.

[0131] Sub-step 11, generating fourth model loss information for the above-mentioned initial adjustment parameter information and actual adjustment parameter label information.

[0132] Sub-step 12, in response to determining that the first model loss information is less than the first model value, the second model loss information is less than the second model loss value, the third model loss information is less than the third model loss value, and the fourth model loss information is less than the fourth model loss value, the initial device adjustment parameter information generation model is determined as the device adjustment parameter information generation model. The first model value, the second model value, the third model value, and the fourth model value may be pre-set values.

[0133] In the fourth step, in response to determining that at least one of the first model loss information is less than the first model value, the second model loss information is less than the second model loss value, the third model loss information is less than the third model loss value, and the fourth model loss information is less than the fourth model loss value is not satisfied, based on the first model value, the second model value, the third model value and the fourth model value, the model parameters of the initial device adjustment parameter information generation model are updated by back propagation to generate an updated device adjustment parameter information generation model.

[0134] The fifth step is to remove the target preset training data from the preset training data set to obtain the removed training data set.

[0135] The sixth step is to reselect the preset training data from the removed training data set as candidate preset training data.

[0136] In the seventh step, the updated device adjustment parameter information generation model is used as the initial device adjustment parameter information generation model, and the candidate preset training data is used as the target preset training data, and the above model training steps are continued.

[0137] The content in "some optional implementation methods of some embodiments" as an inventive point of the present disclosure solves the technical problem mentioned in the background technology: "How to accurately generate the device adjustment parameters corresponding to the second shooting device according to the positional relationship between the picture difference and the camera device is crucial. The accurate generation of the device adjustment parameters determines the adjustment accuracy and adjustment times of the second shooting device." Based on this, the present disclosure, through the device adjustment parameter information generation model including the first adjustment parameter information generation model based on pixel change information, the second adjustment parameter information generation model based on the device position relationship, the third adjustment parameter information generation model based on parameter difference information, the feature information summary model and the adjustment parameter information generation layer, under the model structure of the device adjustment parameter information generation model, can effectively take into account the picture difference feature information and the device position relationship feature information to accurately generate the adjustment parameter information. In addition, the model-targeted training of the device adjustment parameter information generation model can ensure the model output accuracy of the device adjustment parameter information generation model.

[0138] Step 1035 : In response to determining that the device adjustment parameter information does not satisfy the target adjustment condition, generate adjustment information indicating that the second camera device is not to be adjusted.

[0139] In some embodiments, in response to determining that the device adjustment parameter information does not meet the target adjustment condition, the execution subject may generate adjustment information indicating that the second camera device is not adjusted. The target adjustment condition indicates that the change in parameter information of each parameter corresponding to the device adjustment parameter information is less than the corresponding adjustment value. That is, each parameter has a corresponding adjustment value. The adjustment value may be a maximum adjustment value.

[0140] Step 104 , in response to determining that the device adjustment parameter information satisfies the target adjustment condition, the second camera device is adjusted according to the device adjustment parameter information to obtain an adjusted second camera device.

[0141] In some embodiments, in response to determining that the device adjustment parameter information satisfies the target adjustment condition, the execution subject may perform device adjustment on the second camera device according to the device adjustment parameter information to obtain an adjusted second camera device.

[0142] As an example, the execution subject may perform device adjustment on the second camera device according to parameter information of various parameters corresponding to the device adjustment parameter information to obtain an adjusted second camera device.

[0143] Step 105, using the adjusted second camera device as the second camera device, and continuing to perform the above device adjustment steps.

[0144] In some embodiments, the execution subject may use the adjusted second camera device as the second camera device and continue to execute the device adjustment steps.

[0145] The above-mentioned various embodiments of the present disclosure have the following beneficial effects: through the camera device adjustment method of some embodiments of the present disclosure, the camera device can be replaced accurately and efficiently, so that the regional picture taken by the replacement camera device is consistent with the regional picture taken by the original camera device. Specifically, the reason why the relevant camera device is not accurate and efficient is that the manual adjustment efficiency is low, and it cannot effectively guarantee that the picture taken by the adjusted camera device is completely consistent with the picture content taken by the original camera device, and the problem of inconsistent shooting focus may occur. Based on this, the camera device adjustment method of some embodiments of the present disclosure first obtains the first shooting picture taken by the first camera device for the target shooting area as the real comparison picture of the subsequent replacement camera device, so as to facilitate the subsequent replacement of the camera device. Then, in response to receiving the replacement information for the above-mentioned first camera device, the replacement camera device corresponding to the above-mentioned first camera device is determined as the second camera device, so as to facilitate the subsequent adjustment and replacement of the second camera device. Secondly, for the second camera device, the following device adjustment steps are performed: the first step is to use the second camera device to shoot the second shooting picture for the above-mentioned target shooting area, so as to compare the second shooting picture with the first shooting picture later, so as to accurately adjust the camera device and ensure that the regional picture shot by the replacement camera device is consistent with the regional picture shot by the original camera device. The second step is to generate the picture difference information for the above-mentioned first shooting picture and the second shooting picture, so as to consider how to adjust the device parameters of the second camera device from the perspective of the picture difference later. The third step is to determine the device position relationship between the above-mentioned first camera device and the second camera device, so as to consider how to adjust the device parameters of the second camera device from the perspective of the device position relationship later. The fourth step is to consider the influencing factors from various aspects according to the picture difference information and the device position relationship, so as to accurately generate the device adjustment parameter information for the second camera device. The fifth step is to generate the adjustment information indicating that the second camera device is not adjusted in response to determining that the device adjustment parameter information does not meet the target adjustment condition. Here, the target adjustment condition can effectively constrain the adjustment force of the second camera device to avoid violating the device adjustment limit of the second camera device for the sake of accuracy. Further, in response to determining that the above-mentioned device adjustment parameter information meets the above-mentioned target adjustment condition, the second camera device is adjusted according to the device adjustment parameter information to obtain the adjusted second camera device. Finally, the adjusted second camera device is used as the second camera device, and the above-mentioned device adjustment steps are continued. In summary, the second camera device is cyclically adjusted through the angle of the picture difference and the angle of the device position relationship to achieve the replacement of the camera device, so that the area picture taken by the replacement camera device is consistent with the area picture taken by the original camera device.

[0146] Further references Figure 2As an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of a camera device adjustment device, and these device embodiments are Figure 1 Corresponding to the method embodiments shown, the camera device adjustment device can be specifically applied to various electronic devices.

[0147] like Figure 2 As shown, a camera device adjustment device 200 includes: an acquisition unit 201, a determination unit 202, a first execution unit 203, a device adjustment unit 204, and a second execution unit 205. The acquisition unit 201 is configured to acquire a first captured image of a target captured area captured by a first camera device; the determination unit 202 is configured to determine, in response to receiving replacement information for the first camera device, a replacement camera device corresponding to the first camera device as a second camera device; the first execution unit 203 is configured to perform the following device adjustment steps for the second camera device: using the second camera device to capture a second captured image of the target captured area; generating image difference information for the first captured image and the second captured image; determining the difference between the first camera device and the second camera device; the device position relationship between them; generating device adjustment parameter information for the second camera device according to the picture difference information and the device position relationship; in response to determining that the device adjustment parameter information does not meet the target adjustment condition, generating adjustment information indicating that the second camera device is not adjusted; the device adjustment unit 204 is configured to, in response to determining that the above-mentioned device adjustment parameter information meets the above-mentioned target adjustment condition, perform device adjustment on the second camera device according to the device adjustment parameter information to obtain an adjusted second camera device; the second execution unit 205 is configured to use the adjusted second camera device as the second camera device and continue to execute the above-mentioned device adjustment step.

[0148] It is understandable that the units described in the camera adjustment device 200 are similar to those described in the reference Figure 1 Therefore, the operations, features and beneficial effects described above for the method are also applicable to the camera device adjustment device 200 and the units included therein, and will not be described in detail here.

[0149] Reference below Figure 3 , which shows a structural schematic diagram of an electronic device (eg, an electronic device) 300 suitable for implementing some embodiments of the present disclosure. Figure 3 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0150] like Figure 3As shown, the electronic device 300 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the electronic device 300 are also stored. The processing device 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0151] Typically, the following devices may be connected to the I / O interface 305: input devices 306 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 308 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 309. The communication devices 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 3 The electronic device 300 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 3 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0152] In particular, according to some embodiments of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, some embodiments of the present disclosure include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In some such embodiments, the computer program can be downloaded and installed from the network through the communication device 309, or installed from the storage device 308, or installed from the ROM 302. When the computer program is executed by the processing device 301, the above-mentioned functions defined in the method of some embodiments of the present disclosure are executed.

[0153] It should be noted that the computer-readable medium in some embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In some embodiments of the present disclosure, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In some embodiments of the present disclosure, the computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0154] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0155] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device. The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: obtains a first captured image of a target shooting area captured by a first camera device; in response to receiving replacement information for the first camera device, determines a replacement camera device corresponding to the first camera device as a second camera device; performs the following device adjustment steps for the second camera device: uses the second camera device to capture a second captured image of the target shooting area; generates image difference information for the first captured image and the second captured image; determines a device position relationship between the first camera device and the second camera device; generates device adjustment parameter information for the second camera device according to the image difference information and the device position relationship; in response to determining that the device adjustment parameter information does not meet the target adjustment condition, generates adjustment information indicating that the second camera device is not adjusted; in response to determining that the device adjustment parameter information meets the target adjustment condition, performs device adjustment on the second camera device according to the device adjustment parameter information to obtain an adjusted second camera device; uses the adjusted second camera device as the second camera device, and continues to perform the device adjustment steps.

[0156] Computer program code for performing the operations of some embodiments of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0157] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0158] The units described in some embodiments of the present disclosure may be implemented by software or by hardware. The described units may also be provided in a processor, for example, may be described as: a processor including an acquisition unit, a determination unit, a first execution unit, a device adjustment unit, and a second execution unit. The names of these units do not, in some cases, constitute limitations on the units themselves, for example, the acquisition unit may also be described as a "unit for acquiring a first captured image of a target capture area captured by a first camera device".

[0159] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0160] The above descriptions are only some preferred embodiments of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the above-mentioned inventive concept. For example, the above-mentioned features are replaced with the technical features with similar functions disclosed in the embodiments of the present disclosure (but not limited to) and the technical solutions formed.

Claims

1. A camera device adjustment method, comprising: Acquire a first captured image of a target shooting area captured by a first camera device; In response to receiving replacement information for the first camera device, determining a replacement camera device corresponding to the first camera device as a second camera device; For the second camera device, perform the following device adjustment steps: Using a second camera device, capturing a second captured image of the target shooting area; Generate picture difference information for the first shot picture and the second shot picture, wherein generating the picture difference information for the first shot picture and the second shot picture includes: using target image processing software, determining initial picture difference information between the first shot picture and the second shot picture; in response to determining that the initial picture difference information satisfies a pixel difference regularity condition, determining the initial picture difference information as picture difference information; in response to determining that the initial picture difference information does not satisfy the pixel difference regularity condition, performing a predetermined ratio division on the first shot picture to generate a first shot sub-picture set, and performing a predetermined ratio division on the second shot picture to generate a second shot sub-picture set; performing sub-picture combination on the first shot sub-picture in the first shot sub-picture set and the second shot sub-picture in the second shot sub-picture set according to a division order to generate a shot sub-picture group set; Determining a device position relationship between the first camera device and the second camera device; generating device adjustment parameter information for the second camera device according to the picture difference information and the device position relationship; In response to determining that the device adjustment parameter information does not satisfy the target adjustment condition, generating adjustment information indicating that the second camera device is not adjusted; In response to determining that the device adjustment parameter information satisfies the target adjustment condition, performing device adjustment on the second camera device according to the device adjustment parameter information to obtain an adjusted second camera device; The adjusted second camera device is used as the second camera device, and the device adjustment step is continued.

2. The method according to claim 1, wherein: The generating device adjustment parameter information for the second camera device according to the picture difference information and the device position relationship includes: Determine first device parameter information corresponding to the first camera device and second device parameter information corresponding to the second camera device; Determining parameter difference information between the first device parameter information and the second device parameter information; Determine pixel change information corresponding to the picture difference information; The pixel change information, the device position relationship and the parameter difference information are input into a pre-trained device adjustment parameter information generation model to generate device adjustment parameter information for the second camera device.

3. The method according to claim 1, wherein: After performing sub-picture combination on the first captured sub-picture set and the second captured sub-picture set according to the segmentation order to generate a captured sub-picture group set, the method further includes: For each captured sub-picture group in the captured sub-picture group set, the following first generation step is performed: Inputting each captured sub-picture in the captured sub-picture group into a pre-trained picture feature information generation model to generate first picture feature information and second picture feature information; Inputting the first picture feature information and the second picture feature information into a picture information recognition model based on feature points and feature objects respectively, so as to generate a first feature point information set and a first feature object information set for the first picture feature information, and a second feature point information set and a second feature object information set for the second picture feature information; generating captured sub-picture difference information according to the first feature point information set, the first feature object information set, the second feature point information set and the second feature object information set; The picture difference information is generated according to the obtained captured sub-picture difference information set.

4. The method according to claim 3, wherein: The step of generating the picture difference information according to the obtained captured sub-picture difference information set comprises: For each piece of captured sub-picture difference information in the captured sub-picture difference information set, determining picture substantial content difference ratio information corresponding to the captured sub-picture difference information; Determine at least one piece of picture substantial content difference ratio information in the obtained picture substantial content difference ratio information set, the corresponding picture substantial content difference ratio information being greater than a first value; Determine the information proportion corresponding to the at least one picture substantial content difference ratio information; In response to determining that the information proportion is less than a second value, removing at least one piece of picture substantial content difference ratio information from the picture substantial content difference ratio information set to obtain a picture substantial content difference ratio information subset after removal; The picture difference information is generated according to the first shooting sub-picture difference information subset corresponding to the removed picture substantial content difference ratio information subset.

5. The method according to claim 4, wherein: The step of generating the picture difference information according to the obtained captured sub-picture difference information set further comprises: In response to determining that the information proportion is greater than or equal to the second value and less than a third value, determining whether there is an information proximity relationship between the second captured sub-picture difference information subsets corresponding to the removed picture substantial content difference ratio information subsets; In response to determining that at least two second captured sub-picture difference information in the second captured sub-picture difference information subset have an information proximity relationship, generating picture difference information according to at least two second captured sub-picture difference information corresponding to the at least two second captured sub-picture difference information; In response to determining that the information proportion is greater than the third value, the picture difference information is generated according to the first shooting picture and the second shooting picture using a picture difference information generation model.

6. The method according to claim 5, wherein: Said method further comprises: Determine the same pixel area pictures for the first shooting picture and the second shooting picture, and obtain a pixel area picture group set, wherein the pixel area group includes: at least one first pixel area picture in the first shooting picture and at least one second pixel area picture in the second shooting picture, and the pixel similarity corresponding to each first pixel area picture in the at least one first pixel area picture and each second pixel area picture in the at least one second pixel area picture is higher than a fourth value; For each pixel area picture group in the pixel area picture group set, the following second generating step is performed: Determine at least one first pixel area picture and at least one second pixel area picture included in the pixel area picture group as at least one first target pixel area picture and at least one second target pixel area picture respectively; Determine at least one first picture position information corresponding to the at least one first target pixel area picture and at least one second picture position information corresponding to the at least one second target pixel area picture; Inputting the at least one first target pixel area picture, the at least one first picture position information, the at least one second target pixel area picture and the at least one second picture position information into the picture difference information generation model to generate picture difference sub-information; The picture difference information is generated according to the obtained picture difference sub-information set.

7. The method according to claim 3, wherein: The picture information recognition model based on feature points and feature objects includes: an information recognition model based on feature points and an information recognition model based on feature objects, wherein the information recognition model based on feature points includes: a feature depth extraction model and a feature point information output layer, and the information recognition model based on feature objects includes: the feature depth extraction model and a feature object information output layer; and The step of inputting the first picture feature information and the second picture feature information into a picture information recognition model based on feature points and feature objects respectively to generate a first feature point information set and a first feature object information set for the first picture feature information, and a second feature point information set and a second feature object information set for the second picture feature information, comprises: Inputting the first picture feature information and the second picture feature information into the feature depth extraction model to generate first picture feature depth extraction information and second picture feature depth extraction information; Inputting the first picture feature depth extraction information and the second picture feature depth extraction information into the feature point information output layer to output the first feature point information set and the second feature point information set; The first picture feature depth extraction information and the second picture feature depth extraction information are input into the feature object information output layer to output the first feature object information set and the second feature object information set, The image information recognition model based on feature points and feature objects is trained by the following steps: Acquire a training data set, wherein the training data includes: a first picture, a second picture, a first label feature point information set, a second label feature point information set, a first label feature object information set, and a second label feature object information set; Select target training data from the training data set; For the target training data, perform the following training steps: Determine a first picture and a second picture corresponding to the target training data as a first target picture and a second target picture, respectively; generating first preliminary picture feature information for the first target picture and second preliminary picture feature information for the second target picture; Inputting the first preliminary picture feature information and the second preliminary picture feature information into an initial feature depth extraction model included in an initial picture information recognition model to generate first depth picture feature information and second depth picture feature information; Inputting the first depth picture feature information and the second depth picture feature information into an initial feature point information output layer included in an initial picture information recognition model to generate a first initial feature point information set and a second initial feature point information set; Inputting the first depth picture feature information and the second depth picture feature information into an initial feature object information output layer included in an initial picture information recognition model to generate a first initial feature object information set and a second initial feature object information set; Generate first feature point loss information according to the first initial feature point information set and the first label feature point information set; Generate second feature point loss information according to the second initial feature point information set and the second label feature point information set; Generate first feature object loss information according to the first initial feature object information set and the first label feature object information set; generating second feature object loss information according to the second initial feature object information set and the second label feature object information set; Generate first point-surface loss information by using point-surface different position loss functions according to the first initial feature point information set and the first initial feature object information set; Generate second point-surface loss information by using point-surface different position loss functions according to the second initial feature point information set and the second initial feature object information set; Generate total loss information according to the first feature point loss information, the second feature point loss information, the first feature object loss information, the second feature object loss information, the first point-surface loss information, and the second point-surface loss information; In response to determining that the total loss information is less than a fifth value, determining the initial picture information recognition model as the picture information recognition model; In response to determining that the total loss information is greater than or equal to the fifth value, updating the model parameters of the initial picture information recognition model according to the total loss information to generate an updated picture information recognition model; The target training data is removed from the training data set to obtain a training data set after removal; training data is reselected from the training data set after removal as candidate training data; the candidate training data is used as the target training data, the updated picture information recognition model is used as the initial picture information recognition model, and the model training step is continued.

8. A camera device adjustment device, comprising: An acquisition unit is configured to acquire a first captured image of a target shooting area captured by a first camera device; a determining unit configured to, in response to receiving replacement information for the first camera device, determine a replacement camera device corresponding to the first camera device as a second camera device; The first execution unit is configured to execute the following device adjustment steps for the second camera device: using the second camera device to capture a second capture picture for the target capture area; Generate picture difference information for the first shot picture and the second shot picture, wherein generating the picture difference information for the first shot picture and the second shot picture includes: using target image processing software, determining initial picture difference information between the first shot picture and the second shot picture; in response to determining that the initial picture difference information satisfies a pixel difference regularity condition, determining the initial picture difference information as picture difference information; in response to determining that the initial picture difference information does not satisfy the pixel difference regularity condition, performing a predetermined ratio division on the first shot picture to generate a first shot sub-picture set, and performing a predetermined ratio division on the second shot picture to generate a second shot sub-picture set; performing sub-picture combination on the first shot sub-picture set and the second shot sub-picture set in the second shot sub-picture set according to a division order to generate a shot sub-picture group set; determining a device position relationship between the first camera device and the second camera device; generating device adjustment parameter information for the second camera device according to the picture difference information and the device position relationship; in response to determining that the device adjustment parameter information does not satisfy a target adjustment condition, generating adjustment information indicating that the second camera device is not adjusted; a device adjustment unit configured to, in response to determining that the device adjustment parameter information satisfies the target adjustment condition, perform device adjustment on the second camera device according to the device adjustment parameter information to obtain an adjusted second camera device; The second execution unit is configured to use the adjusted second camera device as the second camera device and continue to execute the device adjustment step.

9. An electronic device, comprising: one or more processors; a storage device having one or more programs stored thereon, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

10. A computer readable medium having a computer program stored thereon, wherein: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images