A kind of anti-shake camera module

Abstract

The present application relates to the technical field of anti-shake camera, and discloses an anti-shake camera module, which comprises a frame structure module, a first rotating module, a second rotating module and a third rotating module.The frame structure module comprises a frame structure column and a support frame.The first rotating module comprises a rotating body, a rotating wheel, a first belt, a first motor and a first driving wheel.The first belt is engaged with the rotating wheel and the first driving wheel.The second rotating module comprises a second motor, a second belt, a second driving wheel and a second driven wheel.The second belt is engaged with the second driving wheel and the second driven wheel.The third rotating module comprises a camera block, a rotating shaft, a third belt, a third driven wheel, a third driving wheel, a third motor and a rotating bearing.The third belt is engaged with the third driven wheel and the third driving wheel.The camera acquisition module is used for shooting, and the shaking control module is used for controlling the operation of the first rotating module, the second rotating module and the third rotating module.The present application ensures the anti-shake reliability of the camera acquisition module when shooting.

Description

Technical Field

[0001] This invention relates to the field of image stabilization camera technology, and more specifically, to an image stabilization camera module. Background Technology

[0002] With the development of artificial intelligence and Internet of Things technologies, cameras, as image acquisition terminals, are widely used in consumer electronics, security monitoring, industrial inspection, medical equipment, and many other fields. As their application scenarios continue to expand, the requirements for image acquisition accuracy, stability, and clarity in various fields are also facing challenges. Currently, cameras often cannot avoid external interference, such as vibration and wind disturbance. Moreover, these interferences are not unidirectional; they are usually multi-dimensional rotations, tilts, and other movements. This directly causes instantaneous changes in the relative position between the camera and the target, resulting in quality defects such as blurriness, ghosting, and edge blurring in the final image acquired by the camera, thus compromising the image's detail integrity and pixel accuracy.

[0003] Therefore, it is necessary to design a camera stabilization module to solve the problems existing in the current technology. Summary of the Invention

[0004] In view of this, the present invention proposes a camera stabilization module, which aims to solve the problems of blurry and ghosting images caused by camera shake, which cannot meet the requirements for high-precision image acquisition.

[0005] This invention proposes a camera stabilization module, comprising: A frame structure module includes frame structure columns and supporting frames, wherein the frame structure columns and supporting frames are fixedly connected; The first rotating module includes a rotating body, a rotating wheel, a first belt, a first motor, and a first drive wheel. The rotating body and the first motor are fixedly connected. The first motor and the support frame are fixedly connected. The first motor, the support frame, and the first drive wheel are fixedly connected. The rotating body is sleeved on the outside of the rotating wheel. The first belt meshes with the rotating wheel and the first drive wheel. The second rotating module is fixedly connected to the first rotating module. The second rotating module includes a second motor, a second belt, a second driving pulley, and a second driven pulley. The second motor and the second driving pulley are fixedly connected. The second belt meshes with the second driving pulley and the second driven pulley. The support frame and the second motor are fixedly connected. The third rotating module includes a camera block, a rotating shaft, a third belt, a third driven wheel, a third driving wheel, a third motor, and a rotating bearing. The camera block and the rotating shaft are fixedly connected. The third driven wheel is sleeved on the rotating shaft. The rotating bearing is sleeved on the rotating shaft. The third motor is fixedly connected to the support frame and the third driving wheel. The third belt meshes with the third driven wheel and the third driving wheel. A camera acquisition module is disposed inside the first rotating module, and the camera acquisition module is used for taking pictures; A jitter control module is provided, wherein the first rotation module, the second rotation module, and the third rotation module are electrically connected to the jitter control module, and the jitter control module is used to control the operation of the first rotation module, the second rotation module, and the third rotation module.

[0006] Furthermore, the rotating body is provided with several sets of rotating units inside, and each set of rotating units includes several rotating balls. The first rotating module also includes a support frame, which is fixedly connected to the rotating wheel. The rotating body is provided with several positioning grooves. The camera acquisition module includes a support plate and a camera unit, which is fixedly connected to the support frame. The camera unit is fixedly connected to the upper surface of the support plate.

[0007] Furthermore, the second rotating module also includes a second tilting bearing and a second tilting linkage column. The second tilting bearing and the support frame are fixedly connected. The second driven wheel is fitted with the second tilting linkage column. The second tilting bearing is fitted with the second tilting linkage column, and the end of the second tilting linkage column away from the second driven wheel is engaged with a positioning groove.

[0008] Furthermore, the image-stabilized camera module also includes an auxiliary second rotation module, which includes a first flip bearing and a first flip linkage column. The first flip bearing is fixedly connected to the support frame. The first flip bearing is sleeved on the first flip linkage column, and the end of the first flip linkage column away from the first flip bearing is engaged with a positioning groove. The first flip linkage column and the second flip linkage column are coaxial.

[0009] Furthermore, the jitter control module includes a data acquisition and analysis module, a data processing module, and a jitter adjustment module; The acquisition and analysis module is configured to acquire jitter data for each axis, determine the correlation relationship of all jitter data based on the association rule algorithm, and determine the drive time of the first motor, the second motor and the third motor based on the jitter camera model, all jitter data and the correlation relationship. The data processing module is configured to construct a humidity monitoring area based on the camera module, determine the ambient humidity based on the humidity monitoring area, and make a qualification judgment on the camera stabilization environment of the camera module based on the relationship between the ambient humidity and the first belt, the third belt and the second belt. The shake adjustment module is configured to, when the camera stabilization environment of the camera module is unqualified, compare the ambient humidity with the historical shake database to determine the ambient shake coefficient, adjust the drive time based on the ambient shake coefficient to determine the target drive time, and start the first motor, the second motor and the third motor according to the target drive time.

[0010] Furthermore, when determining the association relationships of all jitter data based on association rule algorithms, the following steps are included: Based on the Eclat algorithm, all jitter data is divided into several candidate itemsets. Frequent itemsets are determined based on the support of the candidate itemsets, and the association results between each jitter data are determined based on the frequent itemsets.

[0011] Furthermore, when determining the drive times of the first motor, the second motor, and the third motor based on the shake camera model, all shake data, and the aforementioned correlation, the process includes: The LSTM neural network model and model dataset are obtained in advance. The model dataset is divided into a model training set and a model test set. The LSTM neural network model is trained based on the training set and tested based on the model test set to determine the error index. If the error index value of the currently trained LSTM neural network model is less than or equal to the error index value of the previously trained LSTM neural network model, then training is stopped, and the currently trained LSTM neural network model is determined as the shaking camera model. If the error metric of the currently trained LSTM neural network model is greater than the error metric of the previously trained LSTM neural network model, then the learning rate of the currently trained LSTM neural network model is adjusted, and training continues. Substitute all the jitter data and all the correlation results into the jitter camera model to determine the driving time of the first motor, the second motor and the third motor.

[0012] Furthermore, when constructing a humidity monitoring area based on a camera module and determining the ambient humidity based on the humidity monitoring area, the method includes: The size data of the camera module is magnified in all directions, and the magnification boundary is determined based on the focus range of the camera unit to construct a three-dimensional monitoring space. The module space of the camera module is removed from the three-dimensional monitoring space to determine the humidity monitoring area. The humidity monitoring area is divided into several humidity collection areas, and humidity data for each humidity collection area is obtained. The average value of all humidity data is then determined as the ambient humidity.

[0013] Furthermore, when determining the pass / fail status of the camera stabilization environment for the camera module based on the relationship between the ambient humidity and the first belt, third belt, and second belt, the following steps are included: The standard ambient humidity is determined based on the first belt, the third belt, and the second belt, and then compared with the ambient humidity. If the ambient humidity is greater than the standard ambient humidity, the camera module's image stabilization environment is deemed unqualified. If the ambient humidity is less than or equal to the standard ambient humidity, the camera module is deemed to have a qualified image stabilization environment, and the first motor, the second motor, and the third motor are started according to the driving time.

[0014] Furthermore, when comparing the ambient humidity with a historical jitter database to determine the ambient jitter coefficient, and adjusting the drive time based on the ambient jitter coefficient to determine the target drive time, the process includes: The historical jitter database includes several historical environmental jitter coefficients and several historical environmental humidity values, with each historical environmental jitter coefficient corresponding to a historical environmental humidity value. When the historical ambient humidity is the same as the ambient humidity in the historical jitter database, the historical ambient jitter coefficient corresponding to the historical ambient humidity is determined as the ambient jitter coefficient. When there is no historical ambient humidity in the historical jitter database that is the same as the ambient humidity, the ambient jitter coefficient is determined based on the ambient humidity model and the ambient humidity. The target drive time is the product of the environmental jitter coefficient and the drive time.

[0015] Compared with the prior art, the beneficial effect of the present invention is that the coordinated operation of the first rotating module, the second rotating module and the third rotating module can cope with the jitter interference from different directions. The first rotating module, based on its rotating body, rotating wheels, and first belt, enables the rotating body to rotate around the Y-axis in a three-dimensional coordinate system. The second rotating module, based on its second motor and second belt, enables the rotating body to rotate around the X-axis in a three-dimensional coordinate system. The third rotating module, driven by a third motor and a third drive wheel via a third belt, rotates the support frame around the Z-axis in a three-dimensional coordinate system. The overall motion is gyroscopic, allowing the camera module to generate opposing forces based on shaking in different directions. This mitigates shaking caused by hand movements, ambient airflow disturbances, and other multi-dimensional vibrations, ensuring the image acquisition module's anti-shake reliability during image capture. The shake control module is electrically connected to the first, second, and third rotating modules, enabling real-time capture of shaking and synchronous control of each module's operation, thus adapting to shaking interference during shooting. The image acquisition module is integrated within the first rotating module, with each module working compactly together, achieving miniaturization of the camera module. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of the structure of a camera stabilization module provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of a stabilized camera module along the positive X-axis of a three-dimensional coordinate system, provided in an embodiment of the present invention. Figure 3 This is a schematic diagram of the structure of the first rotating module provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of the first rotating module without the rotating wheel provided in an embodiment of the present invention; Figure 5 This is a schematic diagram of the structure of the third rotating module provided in an embodiment of the present invention; Figure 6 This is a functional block diagram of the jitter control module provided in an embodiment of the present invention.

[0018] In the diagram: 1. Frame structure column; 2. Support frame; 30. First tilting bearing; 31. First tilting linkage column; 40. Rotating body; 42. Support frame; 43. First belt; 44. First motor; 45. First drive wheel; 46. Positioning groove; 47. Rotating wheel; 48. Rotating ball; 50. Second motor; 51. Second tilting bearing; 52. Second tilting linkage column; 53. Second belt; 54. Second driven wheel; 55. Second drive wheel; 60. Rotating bearing; 61. Camera block; 62. Rotating shaft; 63. Third belt; 64. Third driven wheel; 65. Third drive wheel; 66. Third motor; 70. Support plate; 71. Camera unit. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0021] See Figure 1-5As shown in some embodiments of this application, a stabilized camera module includes: a frame structure module, including a frame structure column 1 and a support frame 2, the frame structure column 1 and the support frame 2 being fixedly connected; a first rotating module, including a rotating body 40, a rotating wheel 47, a first belt 43, a first motor 44 and a first drive wheel 45, the rotating body 40 and the first motor 44 being fixedly connected, the first motor 44 and the support frame 2 being fixedly connected, the first motor 44 being fixedly connected to the support frame 2 and the first drive wheel 45, the rotating body 40 being sleeved on the outside of the rotating wheel 47, the first belt 43 engaging the rotating wheel 47 and the first drive wheel 45; and a second rotating module, the second rotating module being fixedly connected to the first rotating module, the second rotating module including a second motor 50, a second belt 53, a second drive wheel 55 and a second driven wheel 54, the second motor 50 and the second drive wheel 55 being fixedly connected; and a second... A belt 53 engages with a second driving pulley 55 and a second driven pulley 54. A support frame 2 and a second motor 50 are fixedly connected. A third rotating module includes a camera block 61, a rotating shaft 62, a third belt 63, a third driven pulley 64, a third driving pulley 65, a third motor 66, and a rotating bearing 60. The camera block 61 and the rotating shaft 62 are fixedly connected. The third driven pulley 64 is fitted onto the rotating shaft 62. The rotating bearing 60 is fitted onto the rotating shaft 62. The third motor 66 is fixedly connected to the support frame 2 and the third driving pulley 65. The third belt 63 engages with the third driven pulley 64 and the third driving pulley 65. A camera acquisition module is located inside the first rotating module and is used for shooting. A shake control module is electrically connected to the first, second, and third rotating modules and is used to control the operation of the first, second, and third rotating modules.

[0022] Specifically, the frame structure module is the load-bearing structure of the camera module. The frame structure column 1 and the support frame 2 of the frame structure module are fixedly connected. The frame structure column 1 provides support for the overall structure, while the support frame 2 provides an installation environment for the first rotating module, the second rotating module, and the third rotating module. The four-legged support of the frame structure column 1 ensures the stability of the support and avoids the risk of shaking due to unstable support. The first rotating module includes a rotating body 40, a rotating wheel 47, a first belt 43, a first motor 44, and a first drive wheel 45. The rotating body 40 is a circular frame, which can better distribute the shaking force from the X-axis direction, thereby ensuring the stability of the camera. The rotating body 40 and the first motor 44 are fixedly connected to ensure the stability of the power provided by the first motor 44. The diameter of the rotating wheel 47 is smaller than the diameter of the rotating body 40, and the rotating wheel 47 can rotate inside the rotating body 40. The first belt 43 meshes with the rotating wheel 47 and the first drive wheel 45. When the rotating body 40 of the first rotating module is subjected to shaking force from the X-axis direction in the three-dimensional coordinate system, the first motor 44 drives the first drive wheel 45 to rotate, and the rotating wheel 47 rotates around its own axis through the synchronous transmission of the first belt 43, that is, the rotation inside the rotating body 40. Wheel 47 rotates around the Y-axis in the three-dimensional coordinate system, thereby eliminating the shaking force from the X-axis direction and ensuring the effect of video and photo capture. The second rotating module includes a second motor 50, a second belt 53, a second driving wheel 55, and a second driven wheel 54. The second motor 50 is fixed to the right side wall of the support frame 2 by bolts. The second motor 50 cooperates with the second driving wheel 55 and is fixedly connected to the second driving wheel 55. When the second motor 50 drives the second driving wheel 55 to rotate, the second driven wheel 54 rotates synchronously through the transmission of the second belt 53. The second rotating module and the first rotating module are fixedly connected, so that the second driven wheel 54 can drive the first rotating module to rotate. That is, the first rotating module rotates around the X-axis in the three-dimensional coordinate system, further improving the anti-shake capability during shooting.The third rotating module includes a camera block 61, a rotating shaft 62, a third belt 63, a third driven pulley 64, a third driving pulley 65, a third motor 66, and a rotating bearing 60. The third driven pulley 64 is fitted onto the rotating shaft 62. The rotating bearing 60 is a deep groove ball bearing. The rotating bearing 60, fitted onto the rotating shaft 62, allows the third driven pulley 64 to synchronously drive the rotating shaft 62 to rotate. The rotating bearing 60 is responsible for supporting the rotating shaft 62 to reduce friction generated during rotation, thereby improving the reliability of image stabilization. The third motor 66 is connected to the support frame 2 and the third driving pulley 65. A fixed connection is established, with the third belt 63 engaging the third driven wheel 64 and the third driving wheel 65. This allows the third motor 66 to drive the third driving wheel 65, which in turn drives the third driven wheel 64 to rotate via the third belt 63. Since the rotating shaft 62 and the camera block 61 are fixedly connected, the third driving wheel 65 can drive the support frame 2 to rotate around the Z-axis in the three-dimensional coordinate system. The camera block 61 has four through holes for fixing, ensuring the rotational reliability of the support frame 2 and reducing the risk of image blurring due to shaking. The image acquisition module is located inside the first rotating module. The image acquisition module is responsible for performing the shooting action, while the shake control module is the control core of the camera module. Electrically connected to the first, second, and third rotating modules, the shake control module comprehensively monitors shake in different dimensions and controls the operation of these modules, reducing external interference and improving image acquisition quality.

[0023] In some embodiments of this application, the rotating body 40 is provided with several sets of rotating units inside, and each set of rotating units includes several rotating balls 48. The first rotating module also includes a support frame 42, which is fixedly connected to the rotating wheel 47. The rotating body 40 is provided with several positioning grooves 46. The camera acquisition module includes a support plate 70 and a camera unit 71, which is fixedly connected to the support frame 42. The camera unit 71 is fixedly connected to the upper surface of the support plate 70.

[0024] Specifically, the rotating body 40 of the first rotating module has four sets of rotating units evenly arranged circumferentially inside, each set containing three rotating balls 48. This avoids affecting the overall anti-shake capability due to the lag in the rotation of the rotating wheel 47. The support frame 42 of the first rotating module is a C-shaped metal piece, which is tightly fixed to the rotating wheel 47. The rotating body 40 is provided with two positioning grooves 46, each providing installation space for other components to fix the rotating body 40. The support plate 70 and the support frame 42 are fixedly connected, and the camera unit 71 is fixedly connected to the upper surface of the support plate 70, ensuring the stability of the camera unit 71. By setting multiple rotating balls 48 to cooperate with the rotating wheel 47, the reliability of the rotation of the rotating wheel 47 inside the rotating body 40 is ensured, thereby offsetting the shaking from multiple directions, reducing the response time of the anti-shake system, and ensuring the clarity of the image captured by the camera unit 71.

[0025] In some embodiments of this application, the second rotating module further includes a second flip bearing 51 and a second flip linkage column 52. The second flip bearing 51 and the support frame 2 are fixedly connected. The second driven wheel 54 is sleeved on the second flip linkage column 52. The second flip bearing 51 is sleeved on the second flip linkage column 52, and the end of the second flip linkage column 52 away from the second driven wheel 54 is engaged with a positioning groove 46.

[0026] Specifically, the second rotating module's second flip bearing 51 is a deep groove ball bearing, fixed to the support frame 2. The second driven wheel 54 is fitted with the second flip linkage column 52, and the second flip bearing 51 is also fitted with the second flip linkage column 52. The second flip bearing 51 reduces the friction when the second flip linkage column 52 rotates, thereby reducing the anti-shake response time. The end of the second flip linkage column 52 away from the second driven wheel 54 is engaged with a positioning groove 46. The second flip linkage column 52 is inserted into the interior of the positioning groove 46. Furthermore, the second flip linkage column 52 is fixed to the rotating body 40 by bolts. During the process of the second driving wheel 55 driving the second driven wheel 54 through the second belt 53, the rotation of the second driven wheel 54 synchronously drives the rotation of the second flip linkage column 52, causing the rotating body 40 to flip around the X-axis in the three-dimensional coordinate system as the rotation center. The mutual cooperation between the second flip linkage column 52 and the positioning groove 46 restricts the radial movement of the rotating body 40 during flipping, thereby offsetting the jitter in the flipping direction and further ensuring the quality of image acquisition.

[0027] In some embodiments of this application, the image-stabilized camera module further includes an auxiliary second rotation module. The auxiliary second rotation module includes a first flip bearing 30 and a first flip linkage column 31. The first flip bearing 30 is fixedly connected to the support frame 2. The first flip bearing 30 is sleeved on the first flip linkage column 31, and the end of the first flip linkage column 31 away from the first flip bearing 30 is engaged with a positioning groove 46. The first flip linkage column 31 and the second flip linkage column 52 are coaxial.

[0028] Specifically, the second motor 50 in the second rotating module is fixed to the support frame 2, and its output end is connected to the second driving wheel 55. The second belt 53 meshes with the second driving wheel 55 and the second driven wheel 54. One end of the second flip linkage column 52 is fixed to the second driven wheel 54, and the other end is engaged with a positioning groove 46 of the rotating body 40. The auxiliary second rotating module is symmetrically arranged in space with the second rotating module. The auxiliary second rotating module includes a first flip bearing 30 and a first flip linkage column 31. The first flip bearing 30 is also fixed to the support frame 2, and the first flip bearing 30 is fitted with the first flip linkage column 31. The end of the first flip linkage column 31 away from the first flip bearing 30 is engaged with a positioning groove 46. Since the first flip linkage column 31 and the second flip linkage column 52 are coaxial, the force balance of the rotating body 40 during flipping is ensured, avoiding attitude deviation caused by force on one side, thereby improving the overall anti-shake stability. The first flip bearing 30 and the second flip bearing 51 respectively constrain the movement of the first flip linkage column 31 and the second flip linkage column 52, reducing radial sway during the flipping process, making the flipping of the rotating body 40 more stable. The first flip linkage column 31 and the second flip linkage column 52 are engaged with the corresponding positioning groove 46, further restricting the flipping trajectory of the rotating body 40, thereby preventing the rotating body 40 from over-flipping or lurching, so that the camera acquisition module can offset external shaking in all directions, ensuring the stability of the image acquired by the camera unit 71, and meeting the requirements of high-precision image acquisition.

[0029] The specific working principle of the camera module is as follows: When the shake control module senses external shaking force, it controls the first motor 44 of the first rotating module, the second motor 50 of the second rotating module, and the third motor 66 of the third rotating module. This causes the first motor 44 to drive the first drive wheel 45 to rotate, and through the synchronous transmission of the first belt 43, it drives the rotating wheel 47 to rotate around its own axis. This causes the rotating wheel 47 within the rotating body 40 to rotate around the Y-axis in the three-dimensional coordinate system. Meanwhile, the second motor 50 drives the second drive wheel 55 to rotate, and the second driven wheel 54 rotates synchronously through the transmission of the second belt 53. The rotation of the second driven wheel 54 synchronously drives the first drive wheel 45 to rotate. The rotation of the second flip linkage column 52 causes the rotating body 40 to flip around the X-axis in the three-dimensional coordinate system. The third motor 66 drives the third drive wheel 65 and drives the third driven wheel 64 to rotate through the third belt 63. The rotating shaft 62 is fixedly connected to the camera block 61. After the camera block 61 is fixed, the third drive wheel 65 can drive the support frame 2 to rotate around the Z-axis in the three-dimensional coordinate system. The overall motion process is a gyroscopic motion, which allows the camera module to generate opposite forces according to the shaking in different directions, thereby ensuring the anti-shake reliability of the camera unit 71 when acquiring images and thus ensuring image quality.

[0030] See Figure 6 As shown, in some embodiments of this application, the jitter control module includes a data acquisition and analysis module, a data processing module, and a jitter adjustment module. The data acquisition and analysis module is configured to acquire jitter data for each axis, determine the correlation relationship of all jitter data based on an association rule algorithm, and determine the driving time of the first motor 44, the second motor 50, and the third motor 66 based on the jitter camera model, all jitter data, and the correlation relationship. The data processing module is configured to construct a humidity monitoring area based on the camera module, determine the ambient humidity based on the humidity monitoring area, and determine the pass / fail status of the camera stabilization environment of the camera module based on the relationship between the ambient humidity and the first belt 43, the third belt 63, and the second belt 53. The jitter adjustment module is configured to determine the ambient jitter coefficient by comparing the ambient humidity with the historical jitter database when the camera stabilization environment of the camera module is not qualified, and adjust the driving time based on the ambient jitter coefficient to determine the target driving time. The first motor 44, the second motor 50, and the third motor 66 are started according to the target driving time.

[0031] Specifically, jitter data refers to the vibration amplitude and frequency of the camera module in multiple directions in a three-dimensional coordinate system. The acquisition and analysis module captures jitter data in real time through vibration sensors and calls the association rule algorithm to perform in-depth mining of this jitter data. The association rule algorithm will filter out the inherent relationship between jitter data in different axes, such as whether other directions are accompanied by vibrations of a specific frequency when the longitudinal jitter amplitude increases, or the interval pattern between longitudinal jitter and lateral jitter, etc., and then construct the correlation relationship between all jitter data. The data acquisition and analysis module inputs all jitter data and correlations into the jitter camera model. Based on the pattern of camera imaging deviation under different jitter states, the model inversely calculates the required start-up time for the first motor 44, second motor 50, and third motor 66 to ensure they can compensate for the current jitter. The data processing module constructs a humidity monitoring area based on the overall structural layout of the camera module. This area comprehensively includes the environmental space where the camera module is located, and a humidity sensor acquires the ambient humidity in real time. The coefficient of friction on the surfaces of the first belt 43, third belt 63, and second belt 53 changes under different humidity conditions. In high-humidity environments, to ensure the first motor 44, second motor 50, and third motor 66 reach the required belt transmission time, the motor's running time needs to be extended to compensate for slippage and reduced transmission efficiency. Therefore, based on... The camera module's image stabilization environment is assessed based on the relationship between ambient humidity and the first belt 43, third belt 63, and second belt 53. If the image stabilization environment is unqualified, a historical data retrieval mechanism is triggered to obtain a historical shake database and compare it with the ambient humidity. The historical shake database covers motor drive conditions under similar or identical shake scenarios in the past, thereby determining the environmental shake coefficient that reflects the impact of the current ambient humidity on shake compensation. The shake adjustment module uses the environmental shake coefficient as a correction basis to dynamically calibrate the drive time, ultimately determining a precise target drive time. The first motor 44, second motor 50, and third motor 66 start with the target drive time, enabling the camera module to not only respond quickly to shake changes but also adapt to different environmental conditions, thereby improving the camera module's image stabilization accuracy and environmental adaptability, ensuring stable and clear image output even in scenarios with hand shaking or airflow disturbances.

[0032] In some embodiments of this application, when determining the association relationship of all jitter data based on the association rule algorithm, the process includes: dividing all jitter data into several candidate itemsets based on the Eclat algorithm, determining frequent itemsets based on the support of the candidate itemsets, and determining the association result between each jitter data based on the frequent itemsets.

[0033] Specifically, the acquisition and analysis module uses all collected jitter data as a dataset. Based on the vertical data format characteristics of the Eclat algorithm, it divides the jitter data into several candidate itemsets. Each candidate itemset contains data combinations with similar jitter features. The Eclat algorithm calculates the support of each candidate itemset in the overall jitter data, i.e., the probability that the jitter data combinations contained in the candidate itemset occur simultaneously. Based on the support, frequent itemsets are selected. Frequent itemsets represent jitter data combinations that occur simultaneously in actual jitter scenarios. The correlation between each jitter data is determined based on the co-occurrence patterns of jitter data within frequent itemsets. For example, if a frequent itemset shows that large vertical jitter is often accompanied by specific horizontal frequency jitter, a correlation can be concluded. The vertical partitioning method of the Eclat algorithm ensures the reliability of the candidate itemsets, reduces redundant data processing, avoids interference from invalid associations, and provides an accurate data foundation for subsequent calculation of drive time using a jitter camera model, improving the reliability of drive time calculation and the targeting of image stabilization compensation.

[0034] In some embodiments of this application, when determining the driving time of the first motor 44, the second motor 50, and the third motor 66 based on the jitter camera model, all jitter data, and correlation relationships, the process includes: pre-acquiring an LSTM neural network model and a model dataset; dividing the model dataset into a model training set and a model test set; training the LSTM neural network model based on the training set; testing the trained LSTM neural network model based on the model test set; determining an error index; if the error index value of the currently trained LSTM neural network model is less than or equal to the error index value of the previously trained LSTM neural network model, then stopping training and determining the currently trained LSTM neural network model as the jitter camera model; if the error index value of the currently trained LSTM neural network model is greater than the error index value of the previously trained LSTM neural network model, then adjusting the learning rate of the currently trained LSTM neural network model and continuing training; and substituting all jitter data and all correlation results into the jitter camera model to determine the driving time of the first motor 44, the second motor 50, and the third motor 66.

[0035] Specifically, an LSTM neural network model and a model dataset are pre-acquired. The model dataset includes vibration information of the camera module along different axial directions, such as displacement, tilt angle, sway amplitude, and vibration period, as well as time-series data showing the changes of this vibration data over time. The model dataset also includes operating parameters of the first motor 44, the second motor 50, and the third motor 66, including motor speed, torque, start-up time, and start-stop acceleration. The model dataset is divided into a training set and a test set in a 3:2 ratio to ensure the model's generalization ability. The training set is used to train the model, while the test set is used to verify the model's performance. The LSTM neural network model is trained using a training set. LSTM neural networks excel at processing time-series data, learning and capturing patterns between data points from the training set. After training, the trained LSTM neural network model is tested using a test set to determine error metrics, such as mean squared error or mean absolute error. The error metric value of the current trained LSTM neural network model is compared with that of the model after the previous training. The error metric value is the numerical value of a specific metric chosen from among the error metrics. If the error metric value of the current trained LSTM neural network model is less than or equal to that of the model after the previous training, it indicates that the model performance is stabilizing or has reached its optimal state, and training can be stopped. The current trained LSTM neural network model is then designated as a jitter camera model. If the error metric value of the current trained LSTM neural network model is greater than that of the model after the previous training, it indicates that the model has deviated in the learning direction. In this case, the learning rate of the current trained LSTM neural network model is adjusted to correct the deviation in the learning direction, and training continues. After determining the shake camera model, all the collected shake data and all the correlation results obtained through the Eclat algorithm are substituted into the shake camera model to output the driving time of the first motor 44, the second motor 50 and the third motor 66. This realizes intelligent calculation of the driving time and improves the accuracy and reliability of the camera module's image stabilization response.

[0036] In some embodiments of this application, when constructing a humidity monitoring area based on a camera module and determining the ambient humidity based on the humidity monitoring area, the process includes: magnifying the camera module's size data in all directions and determining the magnification boundary based on the focus range of the camera unit 71 to construct a three-dimensional monitoring space; removing the camera module's module space from the three-dimensional monitoring space to determine the humidity monitoring area; dividing the humidity monitoring area into several humidity acquisition areas and acquiring humidity data for each humidity acquisition area; and determining the average of all humidity data as the ambient humidity.

[0037] Specifically, based on the dimensions (length, width, and height) of the camera module itself, the system expands outwards in all directions. The magnification boundary is determined by the focusing range of the camera unit 71, ensuring that the expanded space covers the environmental areas around the camera module that might affect belt performance, without being too large and resulting in ineffective monitoring. This constructs a complete three-dimensional monitoring space. Within this three-dimensional monitoring space, the space occupied by the camera module itself is removed, leaving the humidity monitoring area. This ensures that the collected humidity data comes from the actual working environment. The humidity monitoring area is equally divided into several humidity collection areas, the specific number of which can be dynamically adjusted according to the actual size of the humidity monitoring area. Humidity data is obtained from each humidity collection area, and the average of all humidity data is determined as the ambient humidity. This approach balances comprehensive monitoring with avoiding the influence of humidity deviations in a single area on the overall ambient humidity judgment, improving the accuracy of the ambient humidity readings. This provides a reliable basis for subsequent assessment of the stabilization environment's suitability, ensuring the stability of the camera module's stabilization performance under different humidity conditions.

[0038] In some embodiments of this application, when determining the pass / fail status of the camera module's image stabilization environment based on the relationship between ambient humidity and the first belt 43, the third belt 63, and the second belt 53, the following steps are taken: determining a standard ambient humidity based on the first belt 43, the third belt 63, and the second belt 53; comparing the ambient humidity with the standard ambient humidity; if the ambient humidity is greater than the standard ambient humidity, the camera module's image stabilization environment is deemed unqualified; if the ambient humidity is less than or equal to the standard ambient humidity, the camera module's image stabilization environment is deemed qualified; and the first motor 44, the second motor 50, and the third motor 66 are started according to the driving time.

[0039] Specifically, the standard ambient humidity is determined based on the material characteristics of the first belt 43, the third belt 63, and the second belt 53. Since the first belt 43, the third belt 63, and the second belt 53 are made of the same material, the standard ambient humidity can be determined by testing one of the first belt 43, the third belt 63, or the second belt 53. The standard ambient humidity can be determined by laboratory simulation or the belt manufacturer's instructions, based on the maximum ambient humidity corresponding to the absence of slippage in the first belt 43, the third belt 63, and the second belt 53. This standard ambient humidity is used as the standard ambient humidity. The standard ambient humidity allows the first belt 43, the third belt 63, and the second belt 53 to maintain transmission, ensuring that the power of the first motor 44, the second motor 50, and the third motor 66 can be accurately transmitted. Comparing the ambient humidity with the standard ambient humidity reveals that if the ambient humidity is higher than the standard, it indicates that the current environment will cause problems such as decreased belt elasticity and reduced friction coefficient, leading to transmission slippage or power transmission delay. For example, under normal humidity conditions, when the motor rotates once, the belt moves synchronously by one pulley circumference, causing the load end to rotate once. However, in a high humidity environment, when the motor rotates once, the belt slips off the pulley, and the pulley only moves the belt 0.8 times the pulley circumference, with the remaining 0.2 times being belt slippage relative to the pulley. The load end only rotates 0.8 times. To make the load end also rotate once, the operating... If the driving time is extended, and the motor speed remains constant during this process, the camera module's image stabilization environment is deemed unqualified. If the ambient humidity is less than or equal to the standard ambient humidity, it indicates that the belt can maintain good transmission performance, and the motor can convert into corresponding image stabilization adjustment actions when driven. In this case, the image stabilization environment is deemed qualified. At this time, the first motor 44, the second motor 50, and the third motor 66 are started according to the driving time. This avoids image stabilization failure caused by belt performance degradation when humidity exceeds the standard, and ensures the reliability of image stabilization when humidity is normal, thereby ensuring the stability of the camera module's image stabilization performance.

[0040] In some embodiments of this application, when comparing environmental humidity with a historical jitter database to determine the environmental jitter coefficient, and adjusting the drive time based on the environmental jitter coefficient to determine the target drive time, the following steps are taken: the historical jitter database includes several historical environmental jitter coefficients and several historical environmental humidities, and each historical environmental jitter coefficient corresponds to a historical environmental humidity. When there is a historical environmental humidity in the historical jitter database that is the same as the environmental humidity, the historical environmental jitter coefficient corresponding to that historical environmental humidity is determined as the environmental jitter coefficient. When there is no historical environmental humidity in the historical jitter database that is the same as the environmental humidity, the environmental jitter coefficient is determined based on the environmental humidity model and the environmental humidity. The target drive time is the product of the environmental jitter coefficient and the drive time.

[0041] Specifically, the historical shake database stores several historical environmental shake coefficients and several historical environmental humidity levels. Each historical environmental shake coefficient corresponds to a specific historical environmental humidity level. These data are derived from the actual image stabilization operation records of the camera module under different humidity conditions in the past, accurately reflecting the degree of influence of different humidity levels on shake compensation. The current environmental humidity determined by the humidity monitoring area is compared one by one with all historical environmental humidity levels in the historical shake database. If a historical environmental humidity level that is exactly the same as the current environmental humidity exists in the historical shake database, the historical environmental shake coefficient corresponding to that environmental humidity is directly determined as the environmental shake coefficient, thus ensuring the reliability and consistency of image stabilization compensation. If no historical environmental humidity level that is the same as the current environmental humidity exists in the historical shake database, it indicates that there is no directly reusable historical situation for the current humidity scenario. In this case, a pre-built environmental humidity model is invoked. The environmental humidity model is based on the relationship between a large number of environmental humidity levels and motor operation. Its establishment process is the same as that of the shake camera model, and will not be repeated here. Ambient humidity is input into an environmental humidity model to determine the environmental jitter coefficient, ensuring the intelligence of image stabilization compensation. Through data-driven automated adjustments, the reliability of image stabilization is continuously improved. The determined environmental jitter coefficient is multiplied by the drive time to determine the target drive time. If the camera stabilization environment is unsuitable, meaning a higher target drive time is required, the environmental jitter coefficient will increase accordingly to compensate for the risk of image stabilization failure due to belt performance degradation, thereby ensuring the stability of the camera module's image stabilization performance.

[0042] Compared with the prior art, the beneficial effect of the present invention in the above embodiments is that the first rotating module, the second rotating module and the third rotating module work together to cope with jitter interference from different directions. The first rotating module, based on its rotating body, rotating wheels, and first belt, enables the rotating body to rotate around the Y-axis in a three-dimensional coordinate system. The second rotating module, based on its second motor and second belt, enables the rotating body to rotate around the X-axis in a three-dimensional coordinate system. The third rotating module, driven by a third motor and a third drive wheel via a third belt, rotates the support frame around the Z-axis in a three-dimensional coordinate system. The overall motion is gyroscopic, allowing the camera module to generate opposing forces based on shaking in different directions. This mitigates shaking caused by hand movements, ambient airflow disturbances, and other multi-dimensional vibrations, ensuring the image acquisition module's anti-shake reliability during image capture. The shake control module is electrically connected to the first, second, and third rotating modules, enabling real-time capture of shaking and synchronous control of each module's operation, thus adapting to shaking interference during shooting. The image acquisition module is integrated within the first rotating module, with each module working compactly together, achieving miniaturization of the camera module.

[0043] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program goods according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0044] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A camera module with image stabilization, characterized in that, include: A frame structure module includes frame structure columns and supporting frames, wherein the frame structure columns and supporting frames are fixedly connected; The first rotating module includes a rotating body, a rotating wheel, a first belt, a first motor, and a first drive wheel. The rotating body and the first motor are fixedly connected. The first motor and the support frame are fixedly connected. The first motor, the support frame, and the first drive wheel are fixedly connected. The rotating body is sleeved on the outside of the rotating wheel. The first belt meshes with the rotating wheel and the first drive wheel. The second rotating module is fixedly connected to the first rotating module. The second rotating module includes a second motor, a second belt, a second driving pulley, and a second driven pulley. The second motor and the second driving pulley are fixedly connected. The second belt meshes with the second driving pulley and the second driven pulley. The support frame and the second motor are fixedly connected. The third rotating module includes a camera block, a rotating shaft, a third belt, a third driven wheel, a third driving wheel, a third motor, and a rotating bearing. The camera block and the rotating shaft are fixedly connected. The third driven wheel is sleeved on the rotating shaft. The rotating bearing is sleeved on the rotating shaft. The third motor is fixedly connected to the support frame and the third driving wheel. The third belt meshes with the third driven wheel and the third driving wheel. A camera acquisition module is disposed inside the first rotating module, and the camera acquisition module is used for taking pictures; A jitter control module is provided, wherein the first rotation module, the second rotation module, and the third rotation module are electrically connected to the jitter control module, and the jitter control module is used to control the operation of the first rotation module, the second rotation module, and the third rotation module.

2. The image-stabilized camera module according to claim 1, characterized in that, The rotating body has several sets of rotating units inside, and each set of rotating units includes several rotating balls. The first rotating module also includes a support frame, which is fixedly connected to the rotating wheel. The rotating body has several positioning grooves. The camera acquisition module includes a support plate and a camera unit, which is fixedly connected to the support frame. The camera unit is fixedly connected to the upper surface of the support plate.

3. The image-stabilized camera module according to claim 2, characterized in that, The second rotating module also includes a second flip bearing and a second flip linkage column. The second flip bearing and the support frame are fixedly connected. The second driven wheel is sleeved on the second flip linkage column. The second flip bearing is sleeved on the second flip linkage column, and the end of the second flip linkage column away from the second driven wheel is engaged with a positioning groove.

4. The image-stabilized camera module according to claim 3, characterized in that, It also includes an auxiliary second rotating module, which includes a first flip bearing and a first flip linkage column. The first flip bearing is fixedly connected to the support frame. The first flip bearing is sleeved on the first flip linkage column, and the end of the first flip linkage column away from the first flip bearing is engaged with a positioning groove. The first flip linkage column and the second flip linkage column are coaxial.

5. The image-stabilized camera module according to claim 4, characterized in that, The jitter control module includes an acquisition and analysis module, a data processing module, and a jitter adjustment module; The acquisition and analysis module is configured to acquire jitter data for each axis, determine the correlation relationship of all jitter data based on the association rule algorithm, and determine the drive time of the first motor, the second motor and the third motor based on the jitter camera model, all jitter data and the correlation relationship. The data processing module is configured to construct a humidity monitoring area based on the camera module, determine the ambient humidity based on the humidity monitoring area, and make a qualification judgment on the camera stabilization environment of the camera module based on the relationship between the ambient humidity and the first belt, the third belt and the second belt. The shake adjustment module is configured to, when the camera stabilization environment of the camera module is unqualified, compare the ambient humidity with the historical shake database to determine the ambient shake coefficient, adjust the drive time based on the ambient shake coefficient to determine the target drive time, and start the first motor, the second motor and the third motor according to the target drive time.

6. The image-stabilized camera module according to claim 5, characterized in that, When determining the association relationships of all jitter data based on association rule algorithms, the following are included: Based on the Eclat algorithm, all jitter data is divided into several candidate itemsets. Frequent itemsets are determined based on the support of the candidate itemsets, and the association results between each jitter data are determined based on the frequent itemsets.

7. The image-stabilized camera module according to claim 6, characterized in that, When determining the drive times of the first motor, the second motor, and the third motor based on the jitter camera model, all jitter data, and the aforementioned correlation, the following steps are included: The LSTM neural network model and model dataset are obtained in advance. The model dataset is divided into a model training set and a model test set. The LSTM neural network model is trained based on the training set and tested based on the model test set to determine the error index. If the error index value of the currently trained LSTM neural network model is less than or equal to the error index value of the previously trained LSTM neural network model, then training is stopped, and the currently trained LSTM neural network model is determined as the shaking camera model. If the error metric of the currently trained LSTM neural network model is greater than the error metric of the previously trained LSTM neural network model, then the learning rate of the currently trained LSTM neural network model is adjusted, and training continues. Substitute all the jitter data and all the correlation results into the jitter camera model to determine the driving time of the first motor, the second motor and the third motor.

8. The image-stabilized camera module according to claim 7, characterized in that, When constructing a humidity monitoring area based on a camera module and determining the ambient humidity based on the humidity monitoring area, the process includes: The size data of the camera module is magnified in all directions, and the magnification boundary is determined based on the focus range of the camera unit to construct a three-dimensional monitoring space. The module space of the camera module is removed from the three-dimensional monitoring space to determine the humidity monitoring area. The humidity monitoring area is divided into several humidity collection areas, and humidity data for each humidity collection area is obtained. The average value of all humidity data is then determined as the ambient humidity.

9. The image-stabilized camera module according to claim 8, characterized in that, When determining the pass / fail status of the camera stabilization environment for the camera module based on the relationship between the ambient humidity and the first belt, third belt, and second belt, the following are included: The standard ambient humidity is determined based on the first belt, the third belt, and the second belt, and then compared with the ambient humidity. If the ambient humidity is greater than the standard ambient humidity, the camera module's image stabilization environment is deemed unqualified. If the ambient humidity is less than or equal to the standard ambient humidity, the camera module is deemed to have a qualified image stabilization environment, and the first motor, the second motor, and the third motor are started according to the driving time.

10. The image-stabilized camera module according to claim 9, characterized in that, When comparing the ambient humidity data with a historical jitter database to determine the ambient jitter coefficient, and adjusting the drive time based on the ambient jitter coefficient to determine the target drive time, the process includes: The historical jitter database includes several historical environmental jitter coefficients and several historical environmental humidity values, with each historical environmental jitter coefficient corresponding to a historical environmental humidity value. When the historical ambient humidity is the same as the ambient humidity in the historical jitter database, the historical ambient jitter coefficient corresponding to the historical ambient humidity is determined as the ambient jitter coefficient. When there is no historical ambient humidity in the historical jitter database that is the same as the ambient humidity, the ambient jitter coefficient is determined based on the ambient humidity model and the ambient humidity. The target drive time is the product of the environmental jitter coefficient and the drive time.

Images