A method and device for upgrading an intelligent model

Abstract

The present application is about an upgrade method and device for an intelligent model, and belongs to the field of intelligent analysis. The method includes: obtaining input data belonging to a first environment, the first environment is different from a second environment of a first intelligent model to be upgraded, and the first intelligent model is trained based on first sample data belonging to the second environment; inputting the input data into the first intelligent model, obtaining output data corresponding to the input data processed and output by the first intelligent model, and the output data includes confidence and target box information; training the first intelligent model according to the first sample data and the output data to obtain a second intelligent model. The present application can improve the scene generalization performance of the intelligent model.

Description

Technical Field

[0001] The present application relates to the field of intelligent analysis, and in particular to a method and device for upgrading an intelligent model. Background Art

[0002] The intelligent model is obtained by training the deep learning algorithm. For example, the face detection model is obtained by training the deep learning algorithm with a large number of training samples. The training samples include two categories. One category of training samples includes face images and manually annotated annotation information, and the annotation information can be "yes". The other category of training samples includes non-face images and manually annotated annotation information, and the annotation information can be "no". The trained face detection model can detect whether the image captured by the camera includes a face image.

[0003] When an intelligent model trained in a certain environment is usually only applied to that environment, the performance of the intelligent model will be reduced if it is applied to a new environment. For example, for a face detection model, the face images in the training samples used to train the face detection model are Chinese face images. When the trained face detection model is applied to an environment for detecting Chinese faces, the performance of the face detection model is higher, resulting in a higher accuracy of the face detection model in detecting face images. When the face detection model is applied to a new environment in other countries, the scene generalization performance of the face detection model is reduced, resulting in a decrease in the accuracy of detecting face images. Summary of the invention

[0004] The present application provides a method and device for upgrading an intelligent model to improve the scene generalization performance of the intelligent model. The technical solution is as follows:

[0005] In one aspect, the present application provides a method for upgrading an intelligent model, which is applied to a first device, and the method includes:

[0006] Acquire input data belonging to a first environment, the first environment being different from a second environment of a first intelligent model to be upgraded, the first intelligent model being trained based on first sample data belonging to the second environment;

[0007] Inputting the input data into the first intelligent model, obtaining output data corresponding to the input data output by the first intelligent model after processing the input data, wherein the output data includes confidence and target box information;

[0008] The first intelligent model is trained according to the first sample data and the output data to obtain a second intelligent model.

[0009] Optionally, the training the first intelligent model according to the first sample data and the output data to obtain a second intelligent model includes:

[0010] According to the output data, setting annotation information in the input data corresponding to the output data to obtain second sample data belonging to the first environment, wherein the annotation information is used to indicate whether the target in the input data is a real target;

[0011] The first intelligent model is trained according to the second sample data and the first sample data to obtain a second intelligent model.

[0012] Optionally, the first device sets annotation information in input data corresponding to the output data according to the output data, including:

[0013] Calculating a value score of the output data according to the output data, the value score being used to indicate the suitability of the input data corresponding to the output data as the second sample data;

[0014] According to the value score of each output data, select output data that meets a preset condition from each output data;

[0015] According to the selected output data, annotation information is set in the input data corresponding to the selected output data.

[0016] Optionally, after setting annotation information in input data corresponding to the output data according to the output data, the method further includes:

[0017] The second sample data is displayed, and when a correction operation performed by a user on the annotation information in the second sample data is detected, the annotation information is corrected according to the correction operation.

[0018] Optionally, the step of training the first intelligent model according to the second sample data and the first sample data to obtain a second intelligent model includes:

[0019] The first intelligent model is trained according to the second sample data, the output data corresponding to the second sample data and the first sample data to obtain a second intelligent model.

[0020] Optionally, before inputting the input data into the first intelligent model, the method further includes:

[0021] Input data belonging to the first environment and sent by the second device is received, where the input data is collected by the second device.

[0022] Optionally, after training the first intelligent model according to the first sample data and the output data to obtain the second intelligent model, the method further includes:

[0023] The second smart model is sent to the second device so that the second device upgrades the first smart model installed in the second device to the second smart model.

[0024] On the other hand, the present application provides an upgrade device for an intelligent model, the device comprising:

[0025] A first acquisition module, configured to acquire input data belonging to a first environment, the first environment being different from a second environment to which a first intelligent model to be upgraded is applied, the first intelligent model being trained based on first sample data belonging to the second environment;

[0026] A second acquisition module, used to input the input data into the first intelligent model, and obtain output data corresponding to the input data output by the first intelligent model after processing the input data, wherein the output data includes confidence and target frame information;

[0027] A training module is used to train the first intelligent model according to the first sample data and the output data to obtain a second intelligent model.

[0028] Optionally, the training module includes:

[0029] a setting unit, configured to set annotation information in input data corresponding to the output data according to the output data, so as to obtain second sample data belonging to the first environment, wherein the annotation information is used to indicate whether a target in the input data is a real target;

[0030] A training unit is used to train the first intelligent model according to the second sample data and the first sample data to obtain a second intelligent model.

[0031] Optionally, the setting unit is used to:

[0032] Calculating a value score of the output data according to the output data, the value score being used to indicate the suitability of the input data corresponding to the output data as the second sample data;

[0033] According to the value score of each output data, select output data that meets a preset condition from each output data;

[0034] According to the selected output data, annotation information is set in the input data corresponding to the selected output data.

[0035] Optionally, the device further comprises:

[0036] The correction module is used to display the second sample data, and when detecting a correction operation performed by the user on the annotation information in the second sample data, correct the annotation information according to the correction operation.

[0037] Optionally, the training unit is used to:

[0038] The first intelligent model is trained according to the second sample data, the output data corresponding to the second sample data and the first sample data to obtain a second intelligent model.

[0039] Optionally, the device further comprises:

[0040] The receiving module is used to receive input data belonging to the first environment sent by the second device, where the input data is collected by the second device.

[0041] Optionally, the device further comprises:

[0042] A sending module is used to send the second intelligent model to the second device so that the second device upgrades the first intelligent model installed in the second device to the second intelligent model.

[0043] On the other hand, the present application provides an electronic device, including:

[0044] processor;

[0045] a memory for storing executable instructions for the processor;

[0046] Among them, the processor is used to execute the executable instructions to implement the instructions of the above-mentioned intelligent model upgrade method.

[0047] On the other hand, the present application provides a computer-readable storage medium for storing a computer program, which is loaded and executed by a processor to implement instructions for the above-mentioned intelligent model upgrade method.

[0048] The technical solution provided by the embodiments of the present application may have the following beneficial effects:

[0049] By obtaining the output data of the first intelligent model after processing the input data belonging to the first environment, the first intelligent model is trained according to the first sample data and each output data to obtain the second intelligent model. The second intelligent model is made applicable to the first environment, and the scene generalization performance of the second intelligent model is improved.

[0050] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0052] Figure 1 It is a flow chart of an upgrade method of an intelligent model provided in an embodiment of the present application;

[0053] Figure 2 It is a flow chart of another method for upgrading an intelligent model provided in an embodiment of the present application;

[0054] Figure 3 It is a schematic diagram of a system architecture provided by an embodiment of the present application;

[0055] Figure 4 It is another system architecture schematic diagram provided by an embodiment of the present application;

[0056] Figure 5 It is a schematic diagram of the structure of an upgrade device for an intelligent model provided in an embodiment of the present application;

[0057] Figure 6 It is a schematic diagram of a terminal structure provided in an embodiment of the present application.

[0058] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0059] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0060] The intelligent model is obtained by training a machine learning algorithm. The machine learning algorithm may be a deep learning algorithm or the like, for example, a convolutional neural grid or the like. For example, the intelligent model may be at least one of a human body detection model, a face detection model, a face recognition model, a vehicle detection model, or an object detection model or the like. When training the machine learning algorithm, the first sample data belonging to the second environment is usually used to train the deep learning algorithm, so that the trained intelligent model can process the input data belonging to the second environment.

[0061] For example, taking the training of a face detection model as an example, assuming that the input data belonging to the second environment is face image data, the trained face detection model is used for face images. The face detection model is obtained by training the deep learning algorithm using the first sample data belonging to the second environment. The first sample data is image data, including two categories. One category of first sample data includes face images and annotation information. The annotation information can be "yes", and the face images in this category of first sample data are manually annotated. Another category of first sample data includes images that are non-face images, and the included annotation information can be "no". Another category of first sample data includes images that can be at least one of vehicle images, building images, or animal images. Using the first sample data to train the machine learning algorithm can obtain a face detection model for detecting face images.

[0062] When the camera is installed with the face detection model, the camera captures an image and inputs the captured image into the face detection model as input data of the second environment. The face detection model processes the input data to obtain output data corresponding to the input data, and the output data may include content such as confidence and target frame information, and the target frame information includes the target frame position and target frame size. Based on the confidence, it can be determined whether the image in the input data that is located in the target frame corresponding to the target frame information is a face image.

[0063] Although the intelligent model is used to process input data belonging to the second environment, sometimes the intelligent model is also used to process data belonging to the new environment. For the sake of convenience, the new environment is referred to as the first environment, that is, the intelligent model is applied to the first environment. When the intelligent model is applied to the first environment, the performance of the intelligent model may be reduced, resulting in a reduction in the accuracy of the processing results output by the intelligent model when the intelligent model processes the input data belonging to the first environment.

[0064] In order to enable the intelligent model to adapt to the first environment, the present application can automatically upgrade the intelligent model through any of the following embodiments, so that when the upgraded intelligent model can be applied to the first environment, there will be no performance degradation, and the input data of the first environment can be processed to improve the accuracy of processing the input data of the first environment.

[0065] See also Figure 1 The present application embodiment provides an upgrade method for an intelligent model, which can be an online upgrade or an offline upgrade and is applied to a first device. The method includes:

[0066] Step 101: Obtain input data belonging to a first environment, where the first environment is different from a second environment of a first intelligent model to be upgraded, and the first intelligent model is trained based on first sample data belonging to the second environment.

[0067] Step 102: Input the input data to the first intelligent model, and obtain output data corresponding to the input data output by the first intelligent model after processing the input data, wherein the output data includes confidence and target frame information.

[0068] Step 103: Train the first intelligent model according to the first sample data and the output data to obtain the second intelligent model.

[0069] In the embodiment of the present application, the output data of the first intelligent model processing the input data belonging to the first environment is obtained, and the first intelligent model is trained according to the first sample data and each output data to obtain the second intelligent model. The second intelligent model is made applicable to the first environment, and the scene generalization performance of the second intelligent model is improved.

[0070] See also Figure 2 The embodiment of the present application provides a method for upgrading an intelligent model, which can be an online upgrade or an offline upgrade. The method includes:

[0071] Step 201: The first device obtains at least one input data belonging to a first environment, where the first environment is different from a second environment of a first intelligent model to be upgraded, and the first intelligent model is obtained by training a machine learning algorithm based on first sample data belonging to the second environment.

[0072] The first device may be a camera, a server, a desktop computer, or a tablet computer. The first device may be installed with a first intelligent model, and the first device may form a network with other devices or may not form a network with other devices. Figure 3 , the first device can form a network with at least one second device, and the second device can be installed with a first intelligent model. In this case, the first device and the second device can be devices such as cameras. The first device can collect at least one input data belonging to the first environment. When the first device and the second device form a network, the first device can also receive at least one input data belonging to the first environment collected by the second device. Among them, when the first device and the second device are cameras, the input data collected by the first device is image data, and the input data collected by the second device is image data.

[0073] See also Figure 4, a network consisting of a management device and at least one terminal device, the management device may be a server, a desktop computer or a tablet computer, etc., in which a first intelligent model may be stored. The terminal device may be a camera, etc., in which a first intelligent model is installed. The first device may be a terminal device, or may be a management device. When the first device is a management device, the second device is at least one terminal device, the second device collects input data of the first environment, sends the input data to the first device, and the first device receives the input data. The first device is a terminal device, the second device may be another terminal device, and the first device may collect input data or receive input data sent by the second device.

[0074] The input data may be image data or the like.

[0075] Step 202: The first device inputs the multiple input data into the first intelligent model, and obtains output data corresponding to each input data processed and output by the first intelligent model, where the output data at least includes confidence and target box information.

[0076] For each input data input to the first intelligent model, the first intelligent model processes the input data and outputs output data corresponding to the input data, which is essentially the processing result obtained by the first intelligent model processing the input data. The output data includes confidence and target box information.

[0077] Optionally, the target frame information may include the target frame position and the target frame size, and the output data may also include at least one feature such as data category, high-level semantic feature, time, point or description.

[0078] The confidence is obtained by the first intelligent model based on high-level semantic features.

[0079] For example, the first camera inputs the image data belonging to the first environment into the face detection model, and the face detection model detects each input image data and outputs output data corresponding to each image data. For any image data, the output data includes confidence and target frame information, and the target frame information includes the target frame position and the target frame size. A target frame can be determined in the image data based on the target frame position and the target frame size. The image framed by the target frame is the face image detected by the face detection model, and the confidence indicates the probability that the detected face image is a real face image.

[0080] Step 203: For each output data, the first device sets annotation information for the input data corresponding to the output data according to the output data, so as to obtain second sample data belonging to the first environment.

[0081] The annotation information can be "yes" or "no". If the annotation information in the second sample data is "yes", it means that the target image framed by the target box in the second sample data is a real target. If the annotation information in the second sample data is "no", it means that the target image framed by the target box in the second sample data is not a real target.

[0082] This step can be achieved by the following operations from 2031 to 2033, which are:

[0083] 2031: The first device calculates a value score of the output data based on the output data, where the value score is used to indicate the suitability of input data corresponding to the output data as second sample data.

[0084] The first device may be installed with an intelligent analysis model for calculating the value. The first device may input the output data into the intelligent analysis model, and the intelligent analysis model calculates the value score of the output data based on the confidence level and target box information included in the output data.

[0085] The intelligent analysis model is obtained by training algorithms such as deep learning algorithms in advance. For example, the intelligent analysis model is obtained by training algorithms such as convolutional neural networks in advance. Multiple training samples can be set in advance, and each training sample includes the output data output by the intelligent module and the value score manually marked for the output data. The training sample is input into the deep learning algorithm, and then the deep learning algorithm is trained. The training process is as follows:

[0086] The deep learning algorithm calculates the value score of the output data based on the output data, compares the calculated value score with the value score manually marked on the output data, obtains the score difference, and adjusts the parameters of the deep learning algorithm based on the score difference. Repeat the above training process to train the intelligent analysis model.

[0087] The lower the confidence level included in the output data, the higher the value score of the output data output by the intelligent analysis model may be. Conversely, the higher the confidence level included in the output data, the lower the value score of the output data output by the intelligent analysis network may be.

[0088] 2032: The first device selects output data that meets a preset condition from each output data according to the value score of each output data.

[0089] The first device may select output data having a value score exceeding a preset score threshold from each output data, or the first device may select a preset number of output data having the largest value scores from each output data.

[0090] Optionally, the output data may also be classified into data categories, and the output data may be classified according to the data categories included in each output data. For any data category, a preset number of output data meeting preset conditions are selected from the output data corresponding to the data category according to the value score of each output data. In this way, the output data of each data category may be selected in a balanced manner.

[0091] 2033: For each selected output data, the first device sets annotation information in the input data corresponding to the output data according to the output data, so as to obtain second sample data belonging to the first environment.

[0092] For each selected output data, the output data includes confidence and target frame information, and the target frame information may include the target frame position and target size. According to the target frame information in the output data, a target frame is determined in the input data corresponding to the output data. According to the confidence, it is determined whether the target framed by the target frame is a real target, and annotation information is set for the target according to the determination result.

[0093] When the confidence exceeds the preset confidence threshold, it can be determined that the target framed by the target box in the input data is a real target, and the label information set for the target is "yes", and the label information "yes" is used to indicate that the target is a real target. When the confidence does not exceed the preset confidence threshold, it can be determined that the target framed by the target box in the input data is not a real target, and the label information set for the target is "no", and the label information "no" is used to indicate that the target is not a real target.

[0094] For example, taking the first intelligent model as a face detection model, the input data corresponding to each output data is image data. For each output data, a target frame is determined in the image data corresponding to the output data according to the target frame information included in the output data, and whether the target enclosed by the target frame is a face image is determined according to the confidence, and annotation information is set for the target according to the determination result.

[0095] The labeling information set in the second sample data may be wrong, that is, the target in the second sample data may not be a real target, but the labeling information marked is "yes", or the target in the second sample data may be a real target, but the labeling information marked is "no".

[0096] In order to improve the accuracy of setting the annotation information and the accuracy of the second sample data, the second sample data can be displayed for the user to confirm. When the user confirms that the annotation information in the second sample data is incorrect, the user can correct the annotation information in the second sample data.

[0097] During implementation: the second sample data is displayed, and when a correction operation performed by the user on the annotation information in the second sample data is detected, the annotation information in the second sample data is corrected according to the correction operation.

[0098] When displaying the second sample data, a correction button may also be displayed. If the user finds that the annotation information in the second sample data is incorrect, the user may click the modification button to trigger the modification operation. If the annotation information in the second sample data is "yes", the annotation information in the second sample data is modified to "no". If the annotation information in the second sample data is "no", the annotation information in the second sample data is modified to "yes".

[0099] Step 204: The first device trains the first intelligent model according to the first sample data and the second sample data to obtain the second intelligent model.

[0100] The first intelligent model is obtained by training a machine learning algorithm based on first sample data belonging to the second environment. The first sample data is obtained by manual annotation, and the first sample data can be saved after the first intelligent model is trained using the first sample data. The first sample data can be saved in a server or directly saved in a first device.

[0101] When the first sample data is stored in the server, the first device may download the first sample data from the server. When the first sample data is stored in the first device, the first device may directly obtain the stored first sample data.

[0102] When saving the first sample data, the version number of the first intelligent model can also be saved, that is, there can be a correspondence between the first sample data and the version number of the first intelligent model. In this correspondence, the first sample data corresponding to the version number of the first intelligent model can be image data, or it can be a high-level semantic feature obtained by converting the image data through the first intelligent model. Since this correspondence may be installed on different devices as the first intelligent model, in order to increase privacy, the first sample data corresponding to the version number of the first intelligent model in the correspondence exists in the form of high-level semantic features.

[0103] In this step, the corresponding first sample data is queried from the corresponding relationship according to the version number of the first intelligent model, and the first intelligent model is trained according to the queried first sample data and the second sample data to obtain the second intelligent model.

[0104] If the corresponding first sample data is not found from the corresponding relationship, an error message may be given. Alternatively, a version number may be determined, which is lower than the version number of the first intelligent model, and the corresponding first sample data may be found from the corresponding relationship according to the determined version number. The queried first sample data is input into a conversion model, which is used to convert the input first sample data into first sample data corresponding to the version number of the first intelligent model. The first sample data corresponding to the version number of the first intelligent model output by the conversion model is obtained, and the first intelligent model is trained according to the obtained first sample data and second sample data to obtain a second intelligent model.

[0105] Optionally, the determined version number is the version number that has the smallest difference from the version number of the first intelligent model.

[0106] The conversion model is obtained by training the deep learning algorithm with training data in advance. The training data includes sample data corresponding to the first version number and sample data corresponding to the second version number, and the first version number is less than the second version number. The training data is input into the deep learning algorithm. The deep learning algorithm converts the sample data corresponding to the first version number to obtain sample data corresponding to the second version number. The deep learning algorithm adjusts the parameters according to the sample data corresponding to the second version number obtained by the conversion and the sample data corresponding to the second version number input. The deep learning algorithm repeats the above process and continuously adjusts its own parameters. The deep learning algorithm obtained after stopping the training is the conversion model.

[0107] The first device inputs the first sample data and the second sample data into the first intelligent model, and the first intelligent model is trained according to the first sample data and the second sample data. The training process is:

[0108] The first intelligent model uses both the first sample data and the second sample data as sample data, detects the target in the sample data according to the sample data, compares the detected target with the annotation information in the sample data, obtains difference information, and adjusts the parameters of the first intelligent model according to the difference information. Repeat the above training process to train the second intelligent model.

[0109] Optionally, the first sample data, the output data corresponding to the second sample data and the second sample data can be input into the first intelligent model, so that the first intelligent model can be trained according to the second sample data, the output data corresponding to the second sample data and the first sample data to obtain the second intelligent model.

[0110] When the first device has the first smart model installed, the first device can upgrade the first smart model installed on itself to the second smart model.

[0111] The first device can also send the first smart model to the second device, and the second device can also upgrade the first smart model installed on itself to the second smart model.

[0112] Optional, for Figure 3 In the scenario shown, the first device and each second device can obtain the second intelligent model in the above manner. Figure 3 In the scenario shown, the management device and each terminal device can obtain the second intelligent model in the above manner.

[0113] For any device (which may be the first device, the second device, the management device, or the endpoint device), multiple test data may be set in the device in advance. For each test data, the test data includes input data belonging to the first environment and output data corresponding to the input data. The device may input the input data included in each test data into the second intelligent model, and obtain output data corresponding to each input data included in each test data output by the second intelligent model. According to the output data included in each test data and the output data corresponding to the input data included in each test data, the accuracy is obtained.

[0114] The device can send the accuracy rate to other devices, and receive the accuracy rate sent by other devices. If the accuracy rate obtained by the device is greater than the accuracy rate sent by each other device, the installed first intelligent model is upgraded to the second intelligent model obtained by itself, and the second intelligent model obtained by itself is sent to each other device. Each other device receives the second intelligent model and upgrades the first intelligent model installed by itself to the received second intelligent model.

[0115] If the acquisition accuracy of the device is lower than the accuracy rates sent by one or more other devices, the device receives a second intelligent model sent by one of the other devices with the highest accuracy, and upgrades the first intelligent model installed by itself to the received second intelligent model.

[0116] In an embodiment of the present application, the output data of the first intelligent model processed and outputted by the input data belonging to the first environment is obtained, and the annotation information is set in the input data corresponding to each output data according to each output data, so as to obtain the second sample data belonging to the first environment. The first intelligent model is trained according to the second sample data and the first sample data to obtain the second intelligent model. The second intelligent model is made applicable to the first environment, and the performance of the second intelligent model is improved. The first intelligent model can be automatically upgraded, human resources can be reduced, the upgrade efficiency can be improved, and the generalization adaptability of the intelligent model to the environment can be improved.

[0117] The following are device embodiments of the present application, which can be used to execute the method embodiments of the present application. For details not disclosed in the device embodiments of the present application, please refer to the method embodiments of the present application.

[0118] See also Figure 5 The embodiment of the present application provides an upgrade device for an intelligent model, and the device 300 includes:

[0119] A first acquisition module 301 is used to acquire input data belonging to a first environment, where the first environment is different from a second environment of a first intelligent model to be upgraded, and the first intelligent model is obtained by training a machine learning algorithm based on first sample data belonging to the second environment;

[0120] A second acquisition module 302 is used to input the input data into the first intelligent model, and obtain output data corresponding to the input data output by the first intelligent model after processing the input data, wherein the output data includes confidence and target box information;

[0121] The training module 303 is used to train the first intelligent model according to the first sample data and the output data to obtain a second intelligent model.

[0122] Optionally, the training module 303 includes:

[0123] a setting unit, configured to set annotation information in input data corresponding to the output data according to the output data, so as to obtain second sample data belonging to the first environment, wherein the annotation information is used to indicate whether a target in the input data is a real target;

[0124] A training unit is used to train the first intelligent model according to the second sample data and the first sample data to obtain a second intelligent model.

[0125] Optionally, the setting unit is used to:

[0126] Calculating a value score of the output data according to the output data, the value score being used to indicate the suitability of the input data corresponding to the output data as the second sample data;

[0127] According to the value score of each output data, select output data that meets a preset condition from each output data;

[0128] According to the selected output data, annotation information is set in the input data corresponding to the selected output data.

[0129] Optionally, the device 300 further includes:

[0130] The correction module is used to display the second sample data, and when detecting a correction operation performed by the user on the annotation information in the second sample data, correct the annotation information according to the correction operation.

[0131] Optionally, the training unit is used to:

[0132] The first intelligent model is trained according to the second sample data, the output data corresponding to the second sample data and the first sample data to obtain a second intelligent model.

[0133] Optionally, the device 300 further includes:

[0134] The receiving module is used to receive input data belonging to the first environment sent by the second device, where the input data is collected by the second device.

[0135] Optionally, the device 300 further includes:

[0136] A sending module is used to send the second intelligent model to the second device so that the second device upgrades the first intelligent model installed in the second device to the second intelligent model.

[0137] In the embodiment of the present application, the second acquisition module acquires the output data of the first intelligent model after processing the input data belonging to the first environment, and the training module trains the first intelligent model according to the first sample data and each output data to obtain the second intelligent model. The second intelligent model is made applicable to the first environment, and the scene generalization performance of the second intelligent model is improved.

[0138] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0139] Figure 6 FIG. 4 is a block diagram showing a structure of a terminal 400 provided by an exemplary embodiment of the present invention. The terminal 400 may be a portable mobile terminal, such as a tablet computer, a notebook computer or a desktop computer.

[0140] Typically, the terminal 400 includes a processor 401 and a memory 402 .

[0141] The processor 401 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 401 may be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 401 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a CPU (Central Processing Unit); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor 401 may be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 401 may also include an AI (Artificial Intelligence) processor, which is used to process computing operations related to machine learning.

[0142] The memory 402 may include one or more computer-readable storage media, which may be non-transitory. The memory 402 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In some embodiments, the non-transitory computer-readable storage medium in the memory 402 is used to store at least one instruction, which is used to be executed by the processor 401 to implement the upgrade method of the intelligent model provided in the method embodiment of the present application.

[0143] In some embodiments, the terminal 400 may further optionally include: a peripheral device interface 403 and at least one peripheral device. The processor 401, the memory 402 and the peripheral device interface 403 may be connected via a bus or a signal line. Each peripheral device may be connected to the peripheral device interface 403 via a bus, a signal line or a circuit board. Specifically, the peripheral device includes: at least one of a radio frequency circuit 404, a touch display screen 405, a camera 406, an audio circuit 407, a positioning component 408 and a power supply 409.

[0144] The peripheral device interface 403 may be used to connect at least one peripheral device related to I / O (Input / Output) to the processor 401 and the memory 402. In some embodiments, the processor 401, the memory 402, and the peripheral device interface 403 are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor 401, the memory 402, and the peripheral device interface 403 may be implemented on a separate chip or circuit board, which is not limited in this embodiment.

[0145] The radio frequency circuit 404 is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The radio frequency circuit 404 communicates with the communication network and other communication devices through electromagnetic signals. The radio frequency circuit 404 converts the electrical signal into an electromagnetic signal for transmission, or converts the received electromagnetic signal into an electrical signal. Optionally, the radio frequency circuit 404 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a user identity module card, etc. The radio frequency circuit 404 can communicate with other terminals through at least one wireless communication protocol. The wireless communication protocol includes but is not limited to: the World Wide Web, a metropolitan area network, an intranet, various generations of mobile communication networks (2G, 3G, 4G and 5G), a wireless local area network and / or a WiFi (Wireless Fidelity) network. In some embodiments, the radio frequency circuit 404 may also include circuits related to NFC (Near Field Communication), which is not limited in this application.

[0146] The display screen 405 is used to display the UI (User Interface). The UI may include graphics, text, icons, videos and any combination thereof. When the display screen 405 is a touch display screen, the display screen 405 also has the ability to collect touch signals on the surface or above the surface of the display screen 405. The touch signal can be input to the processor 401 as a control signal for processing. At this time, the display screen 405 can also be used to provide virtual buttons and / or virtual keyboards, also known as soft buttons and / or soft keyboards. In some embodiments, the display screen 405 can be one, and the front panel of the terminal 400 is set; in other embodiments, the display screen 405 can be at least two, which are respectively set on different surfaces of the terminal 400 or are folded; in some other embodiments, the display screen 405 can be a flexible display screen, which is set on the curved surface or folded surface of the terminal 400. Even, the display screen 405 can also be set to a non-rectangular irregular shape, that is, a special-shaped screen. The display screen 405 can be made of materials such as LCD (Liquid Crystal Display), OLED (Organic Light-Emitting Diode) and the like.

[0147] The camera assembly 406 is used to capture images or videos. Optionally, the camera assembly 406 includes a front camera and a rear camera. Typically, the front camera is arranged on the front panel of the terminal, and the rear camera is arranged on the back of the terminal. In some embodiments, there are at least two rear cameras, which are any one of a main camera, a depth of field camera, a wide-angle camera, and a telephoto camera, so as to realize the fusion of the main camera and the depth of field camera to realize the background blur function, the fusion of the main camera and the wide-angle camera to realize the panoramic shooting and VR (Virtual Reality) shooting function or other fusion shooting functions. In some embodiments, the camera assembly 406 may also include a flash. The flash can be a monochrome temperature flash or a dual-color temperature flash. A dual-color temperature flash refers to a combination of a warm light flash and a cold light flash, which can be used for light compensation at different color temperatures.

[0148] The audio circuit 407 may include a microphone and a speaker. The microphone is used to collect sound waves from the user and the environment, and convert the sound waves into electrical signals and input them into the processor 401 for processing, or input them into the radio frequency circuit 404 to achieve voice communication. For the purpose of stereo acquisition or noise reduction, there may be multiple microphones, which are respectively arranged at different parts of the terminal 400. The microphone may also be an array microphone or an omnidirectional acquisition microphone. The speaker is used to convert the electrical signal from the processor 401 or the radio frequency circuit 404 into sound waves. The speaker may be a traditional film speaker or a piezoelectric ceramic speaker. When the speaker is a piezoelectric ceramic speaker, it can not only convert the electrical signal into sound waves audible to humans, but also convert the electrical signal into sound waves inaudible to humans for purposes such as ranging. In some embodiments, the audio circuit 407 may also include a headphone jack.

[0149] Positioning component 408 is used to locate the current geographical location of terminal 400 to implement navigation or LBS (Location Based Service). Positioning component 408 can be a positioning component based on the US GPS (Global Positioning System), China's Beidou system or Russia's Galileo system.

[0150] The power supply 409 is used to power various components in the terminal 400. The power supply 409 can be an alternating current, a direct current, a disposable battery, or a rechargeable battery. When the power supply 409 includes a rechargeable battery, the rechargeable battery can be a wired rechargeable battery or a wireless rechargeable battery. A wired rechargeable battery is a battery charged through a wired line, and a wireless rechargeable battery is a battery charged through a wireless coil. The rechargeable battery can also be used to support fast charging technology.

[0151] In some embodiments, the terminal 400 further includes one or more sensors 410 , including but not limited to: an acceleration sensor 411 , a gyroscope sensor 412 , a pressure sensor 413 , a fingerprint sensor 414 , an optical sensor 415 , and a proximity sensor 416 .

[0152] The acceleration sensor 411 can detect the magnitude of acceleration on the three coordinate axes of the coordinate system established by the terminal 400. For example, the acceleration sensor 411 can be used to detect the components of gravity acceleration on the three coordinate axes. The processor 401 can control the touch display screen 405 to display the user interface in a horizontal view or a vertical view according to the gravity acceleration signal collected by the acceleration sensor 411. The acceleration sensor 411 can also be used for collecting game or user motion data.

[0153] The gyro sensor 412 can detect the body direction and rotation angle of the terminal 400, and the gyro sensor 412 can cooperate with the acceleration sensor 411 to collect the user's 3D actions on the terminal 400. The processor 401 can implement the following functions based on the data collected by the gyro sensor 412: motion sensing (such as changing the UI according to the user's tilt operation), image stabilization during shooting, game control, and inertial navigation.

[0154] The pressure sensor 413 can be set on the side frame of the terminal 400 and / or the lower layer of the touch display screen 405. When the pressure sensor 413 is set on the side frame of the terminal 400, it can detect the user's holding signal of the terminal 400, and the processor 401 performs left and right hand recognition or shortcut operation according to the holding signal collected by the pressure sensor 413. When the pressure sensor 413 is set on the lower layer of the touch display screen 405, the processor 401 controls the operability controls on the UI interface according to the user's pressure operation on the touch display screen 405. The operability controls include at least one of a button control, a scroll bar control, an icon control, and a menu control.

[0155] The fingerprint sensor 414 is used to collect the user's fingerprint, and the processor 401 identifies the user's identity based on the fingerprint collected by the fingerprint sensor 414, or the fingerprint sensor 414 identifies the user's identity based on the collected fingerprint. When the user's identity is identified as a trusted identity, the processor 401 authorizes the user to perform relevant sensitive operations, which include unlocking the screen, viewing encrypted information, downloading software, paying, and changing settings. The fingerprint sensor 414 can be set on the front, back, or side of the terminal 400. When a physical button or a manufacturer logo is set on the terminal 400, the fingerprint sensor 414 can be integrated with the physical button or the manufacturer logo.

[0156] The optical sensor 415 is used to collect the ambient light intensity. In one embodiment, the processor 401 can control the display brightness of the touch display screen 405 according to the ambient light intensity collected by the optical sensor 415. Specifically, when the ambient light intensity is high, the display brightness of the touch display screen 405 is increased; when the ambient light intensity is low, the display brightness of the touch display screen 405 is reduced. In another embodiment, the processor 401 can also dynamically adjust the shooting parameters of the camera assembly 406 according to the ambient light intensity collected by the optical sensor 415.

[0157] The proximity sensor 416, also called a distance sensor, is usually arranged on the front panel of the terminal 400. The proximity sensor 416 is used to collect the distance between the user and the front of the terminal 400. In one embodiment, when the proximity sensor 416 detects that the distance between the user and the front of the terminal 400 is gradually decreasing, the processor 401 controls the touch display screen 405 to switch from the screen-on state to the screen-off state; when the proximity sensor 416 detects that the distance between the user and the front of the terminal 400 is gradually increasing, the processor 401 controls the touch display screen 405 to switch from the screen-off state to the screen-on state.

[0158] Those skilled in the art will understand that Figure 6 The structure shown in the figure does not constitute a limitation on the terminal 400, and the terminal 400 may include more or less components than those shown in the figure, or combine some components, or adopt a different component arrangement.

[0159] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the application disclosed herein. The present application is intended to cover any variations, uses or adaptations of the present application, which follow the general principles of the present application and include common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0160] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A method for upgrading an intelligent model, applied to a first device, characterized in that: The method comprises: Acquire input data belonging to a first environment, the first environment being different from a second environment of a first intelligent model to be upgraded, the first intelligent model being trained based on first sample data belonging to the second environment; Inputting the input data into the first intelligent model, obtaining output data corresponding to the input data output by the first intelligent model after processing the input data, wherein the output data includes confidence and target box information; The first intelligent model is trained according to the first sample data and the output data to obtain the second intelligent model, including: according to the output data, setting annotation information in the input data corresponding to the output data, including: calculating the value score of the output data according to the output data, the value score is used to indicate the suitability of the input data corresponding to the output data as the second sample data; according to the value score of each output data, selecting output data that meets the preset conditions from each output data; according to the selected output data, setting annotation information in the input data corresponding to the selected output data; so as to obtain the second sample data belonging to the first environment, the annotation information is used to indicate whether the target in the input data is a real target; wherein the intelligent analysis model calculates the value score of the output data according to the confidence and target box information included in the output data, and the intelligent analysis model is obtained by training the deep learning algorithm in advance; The first intelligent model is trained according to the second sample data and the first sample data to obtain a second intelligent model.

2. The method according to claim 1, characterized in that After setting the annotation information in the input data corresponding to the output data according to the output data, the method further includes: The second sample data is displayed, and when a correction operation performed by a user on the annotation information in the second sample data is detected, the annotation information is corrected according to the correction operation.

3. The method according to any one of claims 1 to 2, characterized in that: The step of training the first intelligent model according to the second sample data and the first sample data to obtain a second intelligent model includes: The first intelligent model is trained according to the second sample data, the output data corresponding to the second sample data and the first sample data to obtain a second intelligent model.

4. The method according to any one of claims 1 to 2, characterized in that: Before inputting the input data into the first intelligent model, the method further includes: Input data belonging to the first environment and sent by the second device is received, where the input data is collected by the second device.

5. The method according to any one of claims 1 to 2, characterized in that: After the first intelligent model is trained according to the first sample data and the output data to obtain the second intelligent model, the method further includes: The second smart model is sent to the second device so that the second device upgrades the first smart model installed in the second device to the second smart model.

6. An upgrade device for an intelligent model, characterized in that: The device comprises: A first acquisition module, configured to acquire input data belonging to a first environment, the first environment being different from a second environment of a first intelligent model to be upgraded, the first intelligent model being trained based on first sample data belonging to the second environment; A second acquisition module, used to input the input data into the first intelligent model, and obtain output data corresponding to the input data output by the first intelligent model after processing the input data, wherein the output data includes confidence and target frame information; A training module, used for training the first intelligent model according to the first sample data and the output data to obtain a second intelligent model; the training module includes: a setting unit, configured to set annotation information in input data corresponding to the output data according to the output data, so as to obtain second sample data belonging to the first environment, wherein the annotation information is used to indicate whether a target in the input data is a real target; A training unit, configured to train the first intelligent model according to the second sample data and the first sample data to obtain a second intelligent model; The setting unit is used to: calculate a value score of the output data based on the output data, the value score being used to indicate the suitability of the input data corresponding to the output data as the second sample data; select output data that meets preset conditions from each output data based on the value score of each output data; and set annotation information in the input data corresponding to the selected output data based on the selected output data; wherein the intelligent analysis model calculates the value score of the output data based on the confidence and target box information included in the output data, and the intelligent analysis model is obtained by training the deep learning algorithm in advance.

7. The device according to claim 6, characterized in that The device also includes: The correction module is used to display the second sample data, and when detecting a correction operation performed by the user on the annotation information in the second sample data, correct the annotation information according to the correction operation.

8. The device according to any one of claims 6 to 7, characterized in that The training unit is used to: The first intelligent model is trained according to the second sample data, the output data corresponding to the second sample data and the first sample data to obtain a second intelligent model.

9. The device according to any one of claims 6 to 7, characterized in that The device also includes: The receiving module is used to receive input data belonging to the first environment sent by the second device, where the input data is collected by the second device.

10. The device according to any one of claims 6 to 7, characterized in that The device also includes: A sending module is used to send the second intelligent model to the second device so that the second device upgrades the first intelligent model installed in the second device to the second intelligent model.

11. An electronic device, characterized in that: include: processor; a memory for storing executable instructions for the processor; The processor is used to execute the executable instructions to implement the method according to any one of claims 1 to 5.

12. A computer-readable storage medium, characterized in that: Used to store a computer program, wherein the computer program is loaded and executed by a processor to implement instructions of the method according to any one of claims 1 to 5.

Images