CN116540790B

CN116540790B - Tripod head stability control method and device, electronic equipment and storage medium

Info

Publication number: CN116540790B
Application number: CN202310818902.5A
Authority: CN
Inventors: 郭光泉
Original assignee: Shenzhen Bolin Images Science Technology Co ltd
Current assignee: Shenzhen Bolin Images Science Technology Co ltd
Priority date: 2023-07-05
Filing date: 2023-07-05
Publication date: 2023-09-08
Anticipated expiration: 2043-07-05
Also published as: CN116540790A

Abstract

The invention relates to an artificial intelligence technology, and discloses a cradle head stability control method, which comprises the following steps: detecting and eliminating abnormal motion data in the multi-axis motion data set of the holder, and performing windowing on the multi-axis motion data subjected to elimination treatment to obtain a multi-axis window data set; carrying out fine granularity prediction on the multi-axis window data set by using a fine granularity prediction model in the stability fusion analysis model to obtain a window fine granularity prediction result; and carrying out data combination on the window fine granularity prediction result and the multi-axis window data set to obtain an input variable set, inputting the input variable set into a stability analysis model to carry out stability analysis, obtaining a holder stability analysis result, and carrying out stability control on the holder according to the holder stability analysis result. The invention also provides a cradle head stability control device, electronic equipment and a computer readable storage medium. The invention can solve the problem of lower accuracy of the stable control of the cradle head.

Description

Tripod head stability control method and device, electronic equipment and storage medium

Technical Field

The present invention relates to the field of artificial intelligence technologies, and in particular, to a method and apparatus for controlling stability of a pan/tilt, an electronic device, and a computer readable storage medium.

Background

With the progress of science and the development of society in recent years, people's life is also better, and various tools are usually used to assist in operation in life, for example, the cradle head becomes a supporting device for installing, fixing a mobile phone, a camera and a video camera in recent years, and not only can be rotated at will, but also is convenient for users to use and carry. However, the stability of the pan-tilt is an important reference index, and the unstable pan-tilt generally causes inaccuracy of the related data obtained later, which affects the result of data processing, so that the stability of the pan-tilt needs to be analyzed, and the stability of the pan-tilt needs to be controlled according to the result of stability analysis. Therefore, a stable control method of the cradle head with high accuracy is needed.

Disclosure of Invention

The invention provides a method and a device for controlling stability of a cradle head and a computer readable storage medium, and mainly aims to solve the problem of low accuracy of the stability control of the cradle head.

In order to achieve the above object, the present invention provides a method for controlling stability of a pan/tilt head, comprising:

the method comprises the steps that a pre-built multi-axis motion data acquisition module is utilized to acquire data of a cradle head in a preset scene, and a multi-axis motion data set is obtained;

Detecting and eliminating abnormal motion data in the multi-axis motion data set based on the historical motion data set of the holder, and performing windowing on the multi-axis motion data subjected to elimination processing to obtain a multi-axis window data set;

constructing a fine granularity prediction model according to a preset two-way long-short term memory network and a short term memory network, constructing a stability analysis model according to a preset support vector machine algorithm, and carrying out model fusion on the fine granularity prediction model and the stability analysis model to obtain a stability fusion analysis model;

carrying out fine granularity prediction on the multi-axis window data set by utilizing a fine granularity prediction model in the stability fusion analysis model to obtain a window fine granularity prediction result;

and carrying out data combination on the window fine granularity prediction result and the multi-axis window data set to obtain an input variable set, inputting the input variable set into the stability analysis model to carry out stability analysis processing to obtain a holder stability analysis result, and carrying out stability control on the holder according to the holder stability analysis result.

Optionally, the detecting and rejecting abnormal motion data in the multi-axis motion data set based on the historical motion data set of the pan-tilt includes:

Acquiring a historical motion data set of the holder, wherein the historical motion data set comprises holder running time and holder motion speed in a preset historical time period;

carrying out activity analysis on the cloud deck running time, and constructing an activity judgment expression according to the result of the activity analysis;

performing similarity calculation processing on the movement speed of the cradle head to obtain a speed similarity judgment expression;

summarizing the activity judging expression and the speed similarity judging expression into a reference expression, and judging whether the multi-axis motion data in the multi-axis motion data set accords with the reference expression or not;

and taking the multi-axis motion data which does not accord with the reference expression as abnormal motion data and executing elimination processing on the abnormal motion data.

Optionally, the performing similarity calculation on the motion speed of the pan-tilt to obtain a speed similarity judgment expression includes:

constructing a corresponding speed data matrix based on the movement speed of the cradle head;

performing Euclidean calculation on any two rows of data in the speed data matrix to obtain a plurality of Euclidean distance values, and constructing a distance matrix according to the Euclidean distance values;

And constructing an average speed formula based on the distance matrix, and constructing a speed similarity judging expression according to the average speed formula and a preset speed threshold.

Optionally, the activity judgment expression is:wherein (1)>Indicating holder run time during a certain period of time,/->Representing a first activity threshold,/->And representing a second activity threshold, wherein the first activity threshold and the second activity threshold are calculated according to the result of activity analysis, and n is the total data amount in the historical motion data set.

Optionally, the inputting the input variable set into the stability analysis model for stability analysis processing to obtain a cradle head stability analysis result includes:

acquiring historical characteristic data of the holder, constructing a hyperplane function according to the historical characteristic data, calculating a distance value from the hyperplane function to each input variable in the input variable set, and constructing a minimum distance function according to the distance value;

constructing constraint conditions, wherein the constraint conditions are that the distance from the coordinates of each historical characteristic data to a hyperplane is greater than or equal to a minimum distance function;

solving the minimum distance function based on the constraint condition by using a preset Lagrangian function, and obtaining the hyperplane according to the minimum distance function;

Classifying the input variable set according to the hyper-plane to obtain a cradle head stability analysis result. Optionally, the constructing a hyperplane function according to the historical feature data of the pan-tilt includes:

taking the number of the historical characteristic data of the holder as a characteristic dimension, and constructing a multidimensional coordinate system consistent with the characteristic dimension;

mapping the historical characteristic data into the multidimensional coordinate system to obtain a characteristic coordinate set;

calculating Euclidean distance between any two feature coordinates in the feature coordinate set, and selecting two feature coordinates with the minimum Euclidean distance as target feature coordinates;

and respectively taking the target feature coordinates as a left boundary and a right boundary, and constructing a hyperplane function in the middle of the left boundary and the right boundary.

Optionally, the solving the minimum distance function based on the constraint condition by using a preset lagrangian function to obtain the hyperplane includes:

constructing the constraint condition and the minimum distance function into a Lagrangian objective function according to the Lagrangian function;

and solving the Lagrangian objective function to obtain the hyperplane.

In order to solve the above problems, the present invention further provides a cradle head stability control device, the device comprising:

The data acquisition module is used for acquiring data of the cradle head in a preset scene by utilizing the pre-constructed multi-axis motion data acquisition module to obtain a multi-axis motion data set;

the windowing module is used for detecting and eliminating abnormal motion data in the multi-axis motion data set based on the historical motion data set of the holder, and performing windowing on the multi-axis motion data subjected to elimination processing to obtain a multi-axis window data set;

the model fusion module is used for constructing a fine granularity prediction model according to a preset two-way long-short term memory network and a short-term memory network, constructing a stability analysis model according to a preset support vector machine algorithm, and carrying out model fusion on the fine granularity prediction model and the stability analysis model to obtain a stability fusion analysis model;

the stability control module is used for carrying out fine granularity prediction on the multi-axis window data set by utilizing the fine granularity prediction model in the stability fusion analysis model to obtain a window fine granularity prediction result, carrying out data combination on the window fine granularity prediction result and the multi-axis window data set to obtain an input variable set, inputting the input variable set into the stability analysis model to carry out stability analysis processing to obtain a holder stability analysis result, and carrying out stability control on the holder according to the holder stability analysis result.

In order to solve the above-mentioned problems, the present invention also provides an electronic apparatus including:

a memory storing at least one instruction; a kind of electronic device with high-pressure air-conditioning system

And the processor executes the instructions stored in the memory to realize the cradle head stability control method.

In order to solve the above-mentioned problems, the present invention further provides a computer readable storage medium, in which at least one instruction is stored, the at least one instruction being executed by a processor in an electronic device to implement the above-mentioned method for controlling pan-tilt stabilization.

In the embodiment of the invention, the data acquisition is carried out on the holder in the preset scene by utilizing the multi-axis motion data acquisition module to obtain a multi-axis motion data set, the multi-axis motion data acquisition module can accurately and rapidly realize data acquisition, reject abnormal motion data detected in the multi-axis motion data set, and carry out windowing processing on the multi-axis motion data subjected to rejection processing to obtain a multi-axis window data set, thereby ensuring the data accuracy of the multi-axis window data set, carrying out model fusion by utilizing a fine-granularity prediction model and a stability analysis model to obtain a stability fusion analysis model to carry out holder stability analysis, obtaining a holder stability analysis result, and carrying out stable control on the holder according to the holder stability analysis result. Therefore, the cradle head stability control method, the cradle head stability control device, the electronic equipment and the computer readable storage medium can solve the problem of low accuracy of cradle head stability control.

Drawings

FIG. 1 is a flow chart of a method for controlling pan/tilt stabilization according to an embodiment of the present application;

FIG. 2 is a functional block diagram of a pan/tilt stabilization control apparatus according to an embodiment of the present application;

fig. 3 is a schematic structural diagram of an electronic device for implementing the method for controlling the stability of the pan-tilt according to an embodiment of the present application.

The achievement of the objects, functional features and advantages of the present application will be further described with reference to the accompanying drawings, in conjunction with the embodiments.

Detailed Description

It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.

The embodiment of the application provides a cradle head stability control method. The execution main body of the pan-tilt stabilization control method comprises at least one of a server, a terminal and the like which can be configured to execute the method provided by the embodiment of the application. In other words, the pan-tilt stabilization control method may be performed by software or hardware installed in a terminal device or a server device, where the software may be a blockchain platform. The service end includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, and the like.

Referring to fig. 1, a flow chart of a method for controlling pan/tilt stabilization according to an embodiment of the invention is shown. In this embodiment, the pan-tilt stabilization control method includes:

s1, acquiring data of a holder in a preset scene by using a pre-constructed multi-axis motion data acquisition module to obtain a multi-axis motion data set.

The cradle head is a supporting device for installing and fixing portable electronic devices such as mobile phones, cameras, video cameras and the like. The cradle head can rotate randomly, so that motion data with different dimensions can be generated, and the multi-axis motion data set comprises but is not limited to a rotation speed, a rotation angle, a carrying capacity and the like.

Further, the preset scene may be a scene photographed in the field.

In the embodiment of the invention, the multi-axis motion data acquisition module is obtained by constructing a motion control hardware module according to a multi-coordinate motion platform, a motion controller and a computer, and embedding a data acquisition system program written on the basis of a software design platform into the motion control hardware module.

The motion controller is used as a lower computer for motion control of the multi-axis motion data acquisition module, the computer is used as an upper computer of the multi-axis motion data acquisition module for data acquisition and data analysis, and the data acquisition system program mainly comprises multi-axis constant setting, initialization function, origin coordinate parameter setting, data acquisition and the like. Therefore, the multi-axis motion data acquisition module has an accurate data acquisition function,

Preferably, in the scheme, a motion function database of the motion controller can be called to realize data acquisition of the cradle head on different axes, including but not limited to acquisition of real-time position, actual position, motion state, starting speed, running speed and motion distance.

S2, detecting and eliminating abnormal motion data in the multi-axis motion data set based on the historical motion data set of the holder, and performing windowing on the multi-axis motion data subjected to elimination processing to obtain a multi-axis window data set.

In the embodiment of the invention, because the cradle head can be operated and operated outdoors and can be influenced by external factors such as environment, weather and the like, abnormal motion data also exist in the collected multi-axis motion data set, and the existence of the abnormal data can influence the accuracy of a subsequent model, so that the embodiment of the invention screens and eliminates the abnormal motion data in the multi-axis motion data set.

Specifically, the detecting and rejecting abnormal motion data in the multi-axis motion data set based on the historical motion data set of the pan-tilt includes:

In detail, the historical motion data set includes a pan-tilt operation time and a pan-tilt motion speed within a preset historical time period, for example, the preset historical time period may be from 1 month of 2022 years to 5 months of 2022 years, the pan-tilt operation time refers to a time period from start to close of the pan-tilt, and the pan-tilt motion speed refers to a speed of the pan-tilt during moving.

Further, the activity analysis refers to analyzing whether the operation time of the cradle head is about the same day, if the operation time of the cradle head is too high or too low, the cradle head may be unstable, and it may be an external factor that affects the operation of the cradle head and what kind of problem occurs in the cradle head itself.

Specifically, the activity judgment expression may be:wherein (1)>Indicating holder run time during a certain period of time,/->Representing a first activity threshold,/->And representing a second activity threshold, wherein the first activity threshold and the second activity threshold are calculated according to the result of activity analysis, and n is the total data amount in the historical motion data set.

Specifically, the performing similarity calculation on the motion speed of the pan-tilt to obtain a speed similarity judgment expression includes:

In detail, if the motion speed of the pan/tilt head in the preset historical time period is:，

constructing a corresponding speed data matrix based on the movement speed of the cradle head as follows:。

further, the present inventionThe embodiment of the invention carries out Euclidean calculation on any two rows of data in the speed data matrix based on the following formula to obtain a plurality of Euclidean distance values: Wherein, the liquid crystal display device comprises a liquid crystal display device,for said Euclidean distance value, < >>And->And for any two rows of data in the speed data matrix, n is the total data in the historical motion data set.

Further, the preset average speed formula is:wherein (1)>For average speed +.>And n is the total data in the historical motion data set for the Euclidean distance value.

In detail, a speed similarity determination expression is constructed according to the average speed formula and the speed threshold value asWherein->Is the speed threshold.

Preferably, since the multi-axis motion data after the culling process may slightly advance or retard the peak point of the average data of the overall motion data thereof, the long-term prediction data may be windowed in order to improve the accuracy of the motion prediction. Specifically, the windowing processing is performed on the multi-axis motion data after the removing processing to obtain a multi-axis window data set, which includes:

and performing windowing processing on the multi-axis motion data by using a preset windowing function to obtain a multi-axis window data set.

In one embodiment of the present invention, the windowing function is:wherein (1)>For the multiaxial window dataset, +. >For the nth window multiaxial data in said multiaxial window dataset,>for the +.>Multiple axis data of individual window->Is->Multiple axis motion data,/->Is the windowing function.

S3, constructing a fine granularity prediction model according to a preset two-way long-short term memory network and a short-term memory network, constructing a stability analysis model according to a preset support vector machine algorithm, and carrying out model fusion on the fine granularity prediction model and the stability analysis model to obtain a stability fusion analysis model.

In the embodiment of the invention, the stability fusion analysis model adopts two models of different types to analyze motion data of different dimensions, and takes the output of one model as the input of the other model to predict. The fine-granularity prediction model is constructed by a BiLSTM+LSTM model, wherein BiLSTM is a two-way long-short-term memory network, LSTM is a long-term and short-term memory network, and LSTM is a cyclic neural network suitable for predicting a medium-long term time sequence. The embodiment of the invention adds a processing unit (CELL) in the LSTM to determine the relevant information of the motion data in the algorithm.

Further, the preset support vector machine algorithm is an SVM model, and the SVM model has excellent two-classification capability and can conduct stability prediction.

In detail, in the scheme, the two-way long-term memory network and the long-term memory network are connected in series, and the output of the two-way long-term memory network is used as the input of the long-term memory network to obtain the fine-grained prediction model. And constructing a stability analysis model according to a preset support vector machine algorithm, wherein the preset support vector machine algorithm is used as a technical support of the stability analysis model.

According to the embodiment of the invention, the fine granularity prediction model and the stability analysis model are subjected to model fusion, and the obtained stability fusion analysis model can be used for more accurately carrying out stability analysis from multiple dimensions.

And S4, carrying out fine granularity prediction on the multi-axis window data set by utilizing a fine granularity prediction model in the stability fusion analysis model to obtain a window fine granularity prediction result.

In detail, in the embodiment of the present invention, the performing fine-grained prediction on the multi-axis window dataset by using the fine-grained prediction model in the stability fusion analysis model to obtain a window fine-grained prediction result includes:

Vectorizing the multi-axis window data set by using an ebedding layer of the fine-granularity prediction model to obtain a multi-axis vector set;

calculating a state value of the multiaxial vector concentrated multiaxial vector through an input gate in a bidirectional long-short-term memory network in the fine granularity prediction model, calculating an activation value of the multiaxial vector concentrated multiaxial vector through a forgetting gate in the bidirectional long-short-term memory network, calculating a state update value of the multiaxial vector concentrated multiaxial vector according to the state value and the activation value, and calculating a coded data set corresponding to the state update value by an output gate in the bidirectional long-short-term memory network;

decoding the encoded data set by utilizing a long-term and short-term memory network in the fine-granularity prediction model to obtain a decoded data set;

and inputting the decoded data set into a preset activation function to obtain an activation probability value, and obtaining a window fine granularity prediction result according to the activation probability value.

In the embodiment of the invention, the vectorization processing can accelerate the data operation speed. Further, the window fine-grained prediction results include data prediction results of the multi-axis window dataset implemented in a refined time dimension.

In detail, the fine-grained prediction model is a common Encoder-Decoder structure, the multi-axis window data set is firstly expanded to a high-dimensional space by utilizing an embedding layer, then a two-way long-short-term memory network is adopted for encoding, then the encoded data is subjected to decoding processing by using the long-short-term memory network, the probability of falling on each interval is obtained after dimension compression by an activation function, and a prediction result is obtained according to the activation probability. Wherein the activation function is a softmax function.

In an alternative embodiment, the method for calculating the state value includes:wherein (1)>Representing status value +_>Indicating bias of cell units in the input gate, < >>Representing the activation factor of the input gate, +.>Representing the peak value of the multiaxial vector at the moment of input gate t-1,/o>Representing the multiaxial vector at time t +.>Representing the weight of the cell units in the input gate.

In an alternative embodiment, the method for calculating the activation value includes:wherein (1)>Representing an activation value +.>Indicating bias of cell units in amnestic gate, < >>An activator representing amnestic gates, +.>Representing the peak value of the multiaxial vector at the moment of the forgetting gate t-1, < >>Representing the multiaxial vector input at time t, +. >Indicating the weight of the cell units in the forgetting gate.

In an alternative embodiment, the method for calculating the state update value includes:wherein (1)>Representing status update values +_>Representing the peak value of the multiaxial vector at the moment of input gate t-1,/o>Representing the peak value of the multiaxial vector at the moment of forgetting gate t-1,/o>Representing status value +_>Representing the activation value.

In an alternative embodiment, the calculating the encoded data set corresponding to the state update value by using an output gate in the two-way long-short term memory network includes:

the encoded data set is calculated using the following formula:wherein (1)>A set of encoded data is represented and,an activation function representing the output gate, +.>Representing a state update value.

Further, the encoded data set is decoded by using a long-short-term memory network in the fine-granularity prediction model to obtain a decoded data set, and the long-short-term memory network is a part of the two-way long-short-term memory network structure, so that the processing steps are similar and are not repeated here.

S5, carrying out data combination on the window fine granularity prediction result and the multi-axis window data set to obtain an input variable set, inputting the input variable set into the stability analysis model to carry out stability analysis processing to obtain a holder stability analysis result, and carrying out stability control on the holder according to the holder stability analysis result.

In the embodiment of the present invention, the data combination of the window fine granularity prediction result and the multi-axis window data set to obtain an input variable set includes:

resolving the window fine-granularity prediction result into a fine-granularity prediction data set, wherein the fine-granularity prediction data set comprises a plurality of prediction data;

and carrying out random combination processing on the prediction data in the fine-granularity prediction data set and window multi-axis data in the multi-axis window data set to obtain an input variable set.

Specifically, the inputting the input variable set into the stability analysis model for stability analysis processing to obtain a cradle head stability analysis result includes:

Classifying the input variable set according to the hyper-plane to obtain a cradle head stability analysis result. Further, the constructing a hyperplane function according to the historical feature data of the pan-tilt includes:

In detail, in this embodiment, the history feature data includes a history tag of whether each pan-tilt is actually stable or not and corresponding data, for example, a category tag of 1 when the pan-tilt is stable and a category tag of 0 when the pan-tilt is unstable. And if two pieces of history feature data exist, the feature dimension is 2, one piece of history feature data in the two pieces of history feature data is taken as a y-axis, the other piece of history feature data in the two pieces of history feature data is taken as an x-axis, a two-dimensional coordinate system is constructed, and the history feature data is mapped onto the two-dimensional coordinate system, so that a feature coordinate set on the two-dimensional coordinate system is obtained. The target feature coordinates are respectively taken as a left boundary and a right boundary, and the function of the left boundary can be that The right boundary function may be +.>Therefore the hyperplane function is +.>Wherein->And->For a preset fixed parameter, +.>And is the horizontal axis.

Further, according to the embodiment of the invention, the distance value from the hyperplane function to the input variable in the input variable set is calculated according to a preset distance formula:wherein (1)>Is distance value>Is a hyperplane function->For the i-th input variable of said set of input variables, a +.>And->Is a preset fixed parameter.

Further, in one embodiment of the present invention, the minimum distance function is:wherein (1)>As a function of minimum distance>And for the distance value from the hyperplane function to the input variable in the input variable set, N is the total number of variables in the input variable set.

In detail, the constraints may be expressed as. Wherein (1)>For the i-th input variable of said set of input variables, a +.>And->For a preset fixed parameter, +.>Is a hyperplane function->As a function of minimum distance.

Further, the solving the minimum distance function based on the constraint condition by using a preset lagrangian function to obtain the hyperplane includes:

And solving the Lagrangian objective function to obtain the hyperplane.

In detail, the lagrangian objective function is:wherein (1)>Is Lagrangian multiplier +.>And->For a preset fixed parameter, +.>Is a hyperplane function->For the i-th input variable of the set of input variables, N is the total number of variables of the input variable set, < >>Is the first activity threshold.

In detail, the finally obtained cradle head stability analysis result comprises conditions of cradle head stability, cradle head instability and the like, and the cradle head is controlled according to the cradle head stability analysis result to realize cradle head stability control. When the cradle head stability analysis result is cradle head instability, further operation can be performed by analyzing the cause of the instability result. For example, the camera can be controlled by the cradle head to move according to a set moving direction, or the posture of the cradle head can be adjusted according to rotation parameter information.

In the embodiment of the invention, the data acquisition is carried out on the holder in the preset scene by utilizing the multi-axis motion data acquisition module to obtain a multi-axis motion data set, the multi-axis motion data acquisition module can accurately and rapidly realize data acquisition, reject abnormal motion data detected in the multi-axis motion data set, and carry out windowing processing on the multi-axis motion data subjected to rejection processing to obtain a multi-axis window data set, thereby ensuring the data accuracy of the multi-axis window data set, carrying out model fusion by utilizing a fine-granularity prediction model and a stability analysis model to obtain a stability fusion analysis model to carry out holder stability analysis, obtaining a holder stability analysis result, and carrying out stable control on the holder according to the holder stability analysis result. Therefore, the cradle head stability control method provided by the invention can solve the problem of lower accuracy of cradle head stability control.

Fig. 2 is a functional block diagram of a pan/tilt stabilization control apparatus according to an embodiment of the present invention.

The cradle head stability control device 100 of the present invention may be installed in an electronic apparatus. According to the functions, the pan-tilt stabilization control apparatus 100 may include a data acquisition module 101, a windowing module 102, a model fusion module 103, and a stabilization control module 104. The module of the invention, which may also be referred to as a unit, refers to a series of computer program segments, which are stored in the memory of the electronic device, capable of being executed by the processor of the electronic device and of performing a fixed function.

In the present embodiment, the functions concerning the respective modules/units are as follows:

the data acquisition module 101 is configured to acquire data of a pan-tilt in a preset scene by using a pre-constructed multi-axis motion data acquisition module, so as to obtain a multi-axis motion data set;

the windowing module 102 is configured to detect and reject abnormal motion data in the multi-axis motion data set based on the historical motion data set of the pan-tilt, and perform windowing processing on the multi-axis motion data subjected to the rejection processing to obtain a multi-axis window data set;

the model fusion module 103 is configured to construct a fine granularity prediction model according to a preset bidirectional long-short term memory network and a preset long-short term memory network, construct a stability analysis model according to a preset support vector machine algorithm, and perform model fusion on the fine granularity prediction model and the stability analysis model to obtain a stability fusion analysis model;

The stability control module 104 is configured to perform fine-grained prediction on the multi-axis window dataset by using a fine-grained prediction model in the stability fusion analysis model to obtain a window fine-grained prediction result, perform data combination on the window fine-grained prediction result and the multi-axis window dataset to obtain an input variable set, input the input variable set into the stability analysis model to perform stability analysis processing, obtain a holder stability analysis result, and perform stability control on the holder according to the holder stability analysis result.

In detail, the specific embodiments of each module of the pan/tilt stabilization control apparatus 100 are as follows:

step one, utilizing a pre-constructed multi-axis motion data acquisition module to acquire data of a cradle head in a preset scene, and obtaining a multi-axis motion data set.

Further, the preset scene may be a scene photographed in the field.

Detecting and eliminating abnormal motion data in the multi-axis motion data set based on the historical motion data set of the holder, and performing windowing on the multi-axis motion data subjected to elimination processing to obtain a multi-axis window data set.

Specifically, the activity judgment expression may be:wherein (1)>Indicating holder run time during a certain period of time,/->Representing a first activity threshold,/->Representing a second activity threshold, wherein the first and second activity thresholds are analyzed according to activityThe result is calculated that n is the total amount of data in the historical motion dataset.

In detail, if the motion speed of the pan/tilt head in the preset historical time period is:，/>

further, the embodiment of the invention performs euclidean calculation on any two rows of data in the velocity data matrix based on the following formula to obtain a plurality of euclidean distance values:wherein, the liquid crystal display device comprises a liquid crystal display device,for said Euclidean distance value, < >>And->And for any two rows of data in the speed data matrix, n is the total data in the historical motion data set.

Further, the preset average speed formula is: Wherein (1)>For average speed +.>And n is the total data in the historical motion data set for the Euclidean distance value.

In one embodiment of the present invention, the windowing function is:wherein (1)>For the multiaxial window dataset, +.>For the +.>Multiple axis data of individual window->For the +.>The multi-axis data of the individual windows,/>is->Multiple axis motion data,/->Is the windowing function.

And thirdly, constructing a fine granularity prediction model according to a preset two-way long-short term memory network and a short-term memory network, constructing a stability analysis model according to a preset support vector machine algorithm, and carrying out model fusion on the fine granularity prediction model and the stability analysis model to obtain a stability fusion analysis model.

And fourthly, carrying out fine granularity prediction on the multi-axis window data set by utilizing a fine granularity prediction model in the stability fusion analysis model to obtain a window fine granularity prediction result.

In an alternative embodiment, the method for calculating the state value includes:wherein (1)>Representing status value +_ >Indicating bias of cell units in the input gate, < >>Representing the activation factor of the input gate, +.>Representing the peak value of the multiaxial vector at the moment of input gate t-1,/o>Representing the multiaxial vector at time t +.>Representing the weight of the cell units in the input gate.

In an alternative embodiment, the method for calculating the activation value includes:wherein (1)>Representing an activation value +.>Indicating bias of cell units in amnestic gate, < >>An activator representing amnestic gates, +.>Representing the peak value of the multiaxial vector at the moment of the forgetting gate t-1, < >>Representing the multiaxial vector input at time t, +.>Indicating the weight of the cell units in the forgetting gate.

the encoded data set is calculated using the following formula:wherein (1)>A set of encoded data is represented and,an activation function representing the output gate, +. >Representing a state update value.

And fifthly, carrying out data combination on the window fine granularity prediction result and the multi-axis window data set to obtain an input variable set, inputting the input variable set into the stability analysis model to carry out stability analysis processing to obtain a holder stability analysis result, and carrying out stability control on the holder according to the holder stability analysis result.

In detail, in this embodiment, the history feature data includes a history tag of whether each pan-tilt is actually stable or not and corresponding data, for example, a category tag of 1 when the pan-tilt is stable and a category tag of 0 when the pan-tilt is unstable. The number of the history feature data is taken as a feature dimension, if two history feature data exist, the feature dimension is 2, and two calendars are taken as the feature dimensionOne of the history feature data is a y-axis, the other of the two history feature data is an x-axis, a two-dimensional coordinate system is constructed, and the history feature data is mapped onto the two-dimensional coordinate system to obtain a feature coordinate set on the two-dimensional coordinate system. The target feature coordinates are respectively taken as a left boundary and a right boundary, and the function of the left boundary can be thatThe right boundary function may be +. >Therefore the hyperplane function is +.>Wherein->And->For a preset fixed parameter, +.>And is the horizontal axis.

And solving the Lagrangian objective function to obtain the hyperplane.

Fig. 3 is a schematic structural diagram of an electronic device for implementing a method for controlling pan/tilt stabilization according to an embodiment of the present invention.

The electronic device may comprise a processor 10, a memory 11, a communication interface 12 and a bus 13, and may further comprise a computer program, such as a pan/tilt stabilization control program, stored in the memory 11 and executable on the processor 10.

The memory 11 includes at least one type of readable storage medium, including flash memory, a mobile hard disk, a multimedia card, a card memory (e.g., SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. The memory 11 may in some embodiments be an internal storage unit of the electronic device, such as a mobile hard disk of the electronic device. The memory 11 may in other embodiments also be an external storage device of the electronic device, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) or the like, which are provided on the electronic device. Further, the memory 11 may also include both an internal storage unit and an external storage device of the electronic device. The memory 11 may be used to store not only application software installed in an electronic device and various data, such as codes of a pan/tilt stabilization control program, but also temporarily store data that has been output or is to be output.

The processor 10 may be comprised of integrated circuits in some embodiments, for example, a single packaged integrated circuit, or may be comprised of multiple integrated circuits packaged with the same or different functions, including one or more central processing units (Central Processing unit, CPU), microprocessors, digital processing chips, graphics processors, combinations of various control chips, and the like. The processor 10 is a Control Unit (Control Unit) of the electronic device, and connects various components of the entire electronic device using various interfaces and lines, and executes various functions of the electronic device and processes data by running or executing programs or modules (e.g., a cradle head stabilization Control program, etc.) stored in the memory 11, and calling data stored in the memory 11.

The communication interface 12 is used for communication between the electronic device and other devices, including network interfaces and user interfaces. Optionally, the network interface may include a wired interface and/or a wireless interface (e.g., WI-FI interface, bluetooth interface, etc.), typically used to establish a communication connection between the electronic device and other electronic devices. The user interface may be a Display (Display), an input unit such as a Keyboard (Keyboard), or alternatively a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like. The display may also be referred to as a display screen or display unit, as appropriate, for displaying information processed in the electronic device and for displaying a visual user interface.

The bus 13 may be a peripheral component interconnect standard (peripheral component interconnect, PCI) bus, or an extended industry standard architecture (extended industry standard architecture, EISA) bus, among others. The bus 13 may be classified into an address bus, a data bus, a control bus, and the like. The bus 13 is arranged to enable a connection communication between the memory 11 and at least one processor 10 etc.

Fig. 3 shows only an electronic device with components, and it will be understood by those skilled in the art that the structure shown in fig. 3 is not limiting of the electronic device and may include fewer or more components than shown, or may combine certain components, or a different arrangement of components.

For example, although not shown, the electronic device may further include a power source (such as a battery) for supplying power to the respective components, and preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that functions of charge management, discharge management, power consumption management, and the like are implemented through the power management device. The power supply may also include one or more of any of a direct current or alternating current power supply, recharging device, power failure detection circuit, power converter or inverter, power status indicator, etc. The electronic device may further include various sensors, bluetooth modules, wi-Fi modules, etc., which are not described herein.

Further, the electronic device may also include a network interface, optionally, the network interface may include a wired interface and/or a wireless interface (e.g., WI-FI interface, bluetooth interface, etc.), typically used to establish a communication connection between the electronic device and other electronic devices.

Optionally, the electronic device may further comprise a user interface, which may be a Display, an input unit, such as a Keyboard (Keyboard), or a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like. The display may also be referred to as a display screen or display unit, as appropriate, for displaying information processed in the electronic device and for displaying a visual user interface.

It should be understood that the embodiments described are for illustrative purposes only and are not limited to this configuration in the scope of the patent application.

The cradle head stability control program stored in the memory 11 of the electronic device is a combination of a plurality of instructions, and when running in the processor 10, it can be implemented:

Specifically, the specific implementation method of the above instructions by the processor 10 may refer to the description of the relevant steps in the corresponding embodiment of fig. 1, which is not repeated herein.

Further, the electronic device integrated modules/units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. The computer readable storage medium may be volatile or nonvolatile. For example, the computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM).

The present invention also provides a computer readable storage medium storing a computer program which, when executed by a processor of an electronic device, can implement:

In the several embodiments provided in the present invention, it should be understood that the disclosed apparatus, device and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical function division, and there may be other manners of division when actually implemented.

The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical units, may be located in one place, or may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional module in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units can be realized in a form of hardware or a form of hardware and a form of software functional modules.

It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof.

The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.

The blockchain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, encryption algorithm and the like. The Blockchain (Blockchain), which is essentially a decentralised database, is a string of data blocks that are generated by cryptographic means in association, each data block containing a batch of information of network transactions for verifying the validity of the information (anti-counterfeiting) and generating the next block. The blockchain may include a blockchain underlying platform, a platform product services layer, an application services layer, and the like.

Furthermore, it is evident that the word "comprising" does not exclude other elements or steps, and that the singular does not exclude a plurality. A plurality of units or means recited in the system claims can also be implemented by means of software or hardware by means of one unit or means. The terms second, etc. are used to denote a name, but not any particular order.

Finally, it should be noted that the above-mentioned embodiments are merely for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications and equivalents may be made to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. The method for controlling the stability of the cradle head is characterized by comprising the following steps of:

acquiring a historical motion data set of the tripod head, wherein the historical motion data set comprises tripod head running time and tripod head moving speed in a preset historical time period, carrying out activity analysis on the tripod head running time, constructing an activity judgment expression according to the result of the activity analysis, carrying out similarity calculation processing on the tripod head moving speed to obtain a speed similarity judgment expression, summarizing the activity judgment expression and the speed similarity judgment expression into a reference expression, judging whether multiaxial motion data in the multiaxial motion data set accords with the reference expression, taking multiaxial motion data which does not accord with the reference expression as abnormal motion data, carrying out rejection processing on the abnormal motion data, and carrying out windowing processing on the multiaxial motion data subjected to the rejection processing to obtain a multiaxial window data set;

2. The method of claim 1, wherein the step of performing similarity calculation on the motion speed of the pan-tilt to obtain a speed similarity judgment expression comprises:

3. The pan-tilt stabilization control method of claim 1, wherein the activity determination expression is:wherein (1)>Indicating holder run time during a certain period of time,/->Representing a first activity threshold,/->And representing a second activity threshold, wherein the first activity threshold and the second activity threshold are calculated according to the result of activity analysis, and n is the total data amount in the historical motion data set.

4. The method of claim 1, wherein the inputting the input variable set into the stability analysis model for stability analysis processing to obtain a cradle head stability analysis result comprises:

acquiring historical characteristic data of the holder, constructing a hyperplane function according to the historical characteristic data, calculating a distance value from the hyperplane function to each input variable in the input variable set, and constructing a minimum distance function according to the distance value; constructing constraint conditions, wherein the constraint conditions are that the distance from the coordinates of each historical characteristic data to a hyperplane is greater than or equal to a minimum distance function;

Solving the minimum distance function based on the constraint condition by using a preset Lagrangian function, and obtaining the hyperplane according to the minimum distance function; classifying the input variable set according to the hyper-plane to obtain a cradle head stability analysis result.

5. The pan-tilt stabilization control method of claim 4, wherein constructing a hyperplane function from the historical feature data comprises:

calculating Euclidean distance between any two feature coordinates in the feature coordinate set, and selecting two feature coordinates with the smallest Euclidean distance as target feature coordinates;

6. The pan-tilt stabilization control method of claim 4, wherein solving the minimum distance function based on the constraint condition using a preset lagrangian function to obtain the hyperplane comprises:

and solving the Lagrangian objective function to obtain the hyperplane.

7. A pan-tilt stabilization control apparatus, the apparatus comprising:

a windowing module, configured to obtain a historical motion data set of the pan-tilt, where the historical motion data set includes pan-tilt running time and pan-tilt moving speed in a preset historical time period, perform activity analysis on the pan-tilt running time, construct an activity judgment expression according to a result of the activity analysis, perform similarity calculation processing on the pan-tilt moving speed to obtain a speed similarity judgment expression, collect the activity judgment expression and the speed similarity judgment expression as a reference expression, determine whether multi-axis motion data in the multi-axis motion data set accords with the reference expression, take multi-axis motion data that does not accord with the reference expression as abnormal motion data, perform rejection processing on the abnormal motion data, and perform windowing processing on the multi-axis motion data after the rejection processing to obtain a multi-axis window data set;

8. An electronic device, the electronic device comprising:

at least one processor; the method comprises the steps of,

a memory communicatively coupled to the at least one processor; wherein, the liquid crystal display device comprises a liquid crystal display device,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the pan-tilt-stability control method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the pan/tilt stabilization control method according to any one of claims 1 to 6.