Search

CN-121985201-A - Working method and system of projection camera integrated with temperature control function

CN121985201ACN 121985201 ACN121985201 ACN 121985201ACN-121985201-A

Abstract

The invention discloses a working method and a working system of a projection camera integrating a temperature control function, and relates to the field of cameras. The method comprises the steps of synchronously collecting an infrared thermal field, dark current of an image sensor and a micro-mirror array duty state, calculating a multisource thermal disturbance characteristic value through an entropy algorithm and photo-thermal phase difference analysis, driving a thermal diffusion diagram neural network by the characteristic value, inverting a thermal strain field, projecting the thermal strain field to a Zernike basis function space to generate a thermal aberration risk index and a thermal map, calculating an active temperature control intervention grade and strategy parameters through an adaptive threshold and fuzzy reasoning based on the risk index and a convection intensity estimated value, constructing a multiscale non-convex optimization problem according to the intervention grade, solving to obtain a collaborative scheduling weight and an optimization instruction, extracting a maximum Lyapunov index, generating a thermal stability look-ahead confidence degree with the risk index, predicting future thermal evolution and carrying out adaptive regulation. The invention improves the imaging accuracy of the projection camera under the thermal disturbance through the multi-level cooperative temperature control.

Inventors

  • ZHU JIANGBING
  • LI JINHUA

Assignees

  • 北京博视像元科技有限公司

Dates

Publication Date
20260505
Application Date
20260123

Claims (8)

  1. 1. A method of operating a projection camera with integrated temperature control, comprising: S10, synchronously collecting dark current of an infrared thermal field and a complementary metal oxide semiconductor image sensor and a micro mirror array duty state of a digital micro mirror device, calculating disorder degree of the thermal field through an entropy algorithm, calculating photo-thermal phase mismatch degree based on photo-thermal phase difference, and calculating a multi-source thermal disturbance characteristic value according to the disorder degree and the mismatch degree; S20, taking a multisource thermal disturbance characteristic value as an initial excitation to drive a dynamic thermal diffusion diagram neural network with embedded thermal conduction topology, inverting a transient thermal strain field of a key optical interface, projecting the transient thermal strain field to a Zernike aberration basis function space to generate a thermal aberration risk index, and generating a risk thermodynamic diagram based on the index; S30, calculating an active temperature control intervention level through a self-adaptive threshold curved surface and a fuzzy reasoning engine based on a thermal aberration risk index and an environmental convection intensity estimated value based on a micro-electromechanical system anemometer array, and dynamically designating a combination strategy parameter of liquid cooling, thermoelectric refrigeration and light source shaping; S40, constructing a non-convex collaborative optimization problem under multi-objective constraint according to an active temperature control intervention level, taking the time-varying upper bound minimization of a Lyapunov thermal energy functional as an objective function, carrying out iterative solution by a semi-definite relaxation and alternating direction multiplier method to obtain collaborative scheduling weight, and generating an optimization instruction by combining a risk thermodynamic diagram; s50, extracting the maximum Lyapunov index of the thermal evolution track based on the multi-source thermal disturbance characteristic value sequence to represent the thermodynamic chaos degree, generating a thermal stability prospective confidence degree based on the Lyapunov index and the current thermal aberration risk index, predicting future thermal evolution, and performing self-adaptive regulation.
  2. 2. The method for operating a projection camera with integrated temperature control function according to claim 1, wherein the method is characterized by synchronously collecting the dark current of an infrared thermal field and a complementary metal oxide semiconductor image sensor and the duty cycle of a micromirror array of a digital micromirror device, calculating the disorder degree of the thermal field through an entropy algorithm, calculating the mismatch degree of the photo-thermal phase based on the photo-thermal phase difference, and calculating the characteristic value of multi-source thermal disturbance according to the disorder degree and the mismatch degree, and comprises the following steps: The infrared space temperature field, the dark current noise distribution of the complementary metal oxide semiconductor image sensor and the micro-mirror array duty state map of the digital micro-mirror device are synchronously triggered and collected in parallel through microsecond level, and alignment is carried out; Calculating the disorder degree of the thermal field based on the spatial gradient directional entropy of the infrared spatial temperature field, calculating the photo-thermal phase mismatch degree according to the cross-correlation time delay of the light source modulation signal and the thermal response, and respectively quantifying the disorder degree of the thermal distribution and the hysteresis effect of the photo-thermal energy transfer; And generating a multi-source thermal disturbance characteristic value based on the thermal field disorder degree and the photo-thermal phase mismatch degree and the temperature field Laplacian curvature information.
  3. 3. The method of claim 1, wherein the step of inverting the transient thermal strain field of the key optical interface and projecting the same to the zernike aberration basis function space to generate a thermal aberration risk index, and generating a risk thermodynamic diagram based on the index comprises the following steps: Constructing a dynamic thermal diffusion map neural network comprising a thermal conduction topology matrix and an anisotropic coefficient based on a computer-aided design model of the equipment three-dimensional structure and material thermal conductivity parameters; taking a multisource thermal disturbance characteristic value as excitation, inverting a transient thermal expansion strain field of an optical interface through a dynamic thermal diffusion diagram neural network, projecting the transient thermal expansion strain field to a Zernike aberration basis function space, and calculating a thermal aberration risk index by combining energy and change rate of a thermal sensitivity mode; the thermal aberration risk index is correlated with the spatial coordinates of the optical imaging plane, and a spatially-tagged risk thermodynamic diagram is generated to identify and label image regions where geometric distortion may occur.
  4. 4. The method for operating a projection camera with integrated temperature control function according to claim 1, wherein the method for calculating the active temperature control intervention level through the adaptive threshold curved surface and the fuzzy inference engine based on the thermal aberration risk index and the environmental convection intensity estimated value based on the micro-electromechanical system anemometer array dynamically designates the combination strategy parameters of liquid cooling, thermoelectric cooling and light source shaping, and specifically comprises the following sub-steps: Acquiring airflow data through a micro electromechanical system anemometer array, and quantifying the turbulence degree of the airflow by utilizing wavelet packet decomposition and shannon entropy to generate an environment convection intensity estimated value representing external heat dissipation capacity; constructing a self-adaptive threshold curved surface taking a thermal aberration risk index and environmental convection intensity as input, and dynamically adjusting a decision boundary through online learning of a historical intervention effect; the active temperature control intervention grade is generated by a fuzzy inference engine with online self-calibration capability through fusion of the thermal aberration risk index and the environment convection intensity, and is directly mapped into the cooperative control parameters of the liquid cooling flow, the thermoelectric refrigerator power and the laser modulation multi-actuator.
  5. 5. The method for operating a projection camera with integrated temperature control function according to claim 1, wherein a non-convex collaborative optimization problem under multi-objective constraint is constructed according to an active temperature control intervention level, a time-varying upper bound of a lyapunov thermal energy functional is minimized as an objective function, and a collaborative scheduling weight is obtained through iterative solution of a semi-definite relaxation and alternating direction multiplier method, and an optimization instruction is generated by combining a risk thermodynamic diagram, and the method specifically comprises the following sub-steps: Constructing a multi-target non-convex optimization problem based on an active temperature control intervention level, taking the time-varying upper bound of a minimized Lyapunov thermal energy functional as a target, and ensuring that a thermal control strategy does not damage imaging quality under the conditions of heat dissipation, signal-to-noise ratio, dynamic range of a complementary metal oxide semiconductor image sensor and total power consumption imaging and hardware constraint; Converting the non-convex thermo-optical optimization problem into a convex form through semi-definite relaxation, solving three sub-problems of thermal control, light source and imaging by adopting an alternate direction multiplication method, and obtaining a cooperative scheduling weight vector meeting normalization constraint; and mapping the cooperative scheduling weight vector into three execution instructions of liquid cooling flow distribution, light source pulse modulation and complementary metal oxide semiconductor image sensor partition gain compensation in real time, so as to realize space self-adaptive cooperative optimization of heat dissipation, illumination and imaging resources.
  6. 6. The method for operating a projection camera with integrated temperature control function according to claim 1, wherein the maximum lyapunov index of the thermal evolution track is extracted based on a multi-source thermal disturbance characteristic value sequence to represent the thermodynamic chaos degree, and the thermal stability look-ahead confidence is generated based on the lyapunov index and the current thermal aberration risk index to predict the future thermal evolution and perform adaptive regulation, specifically comprising the following sub-steps: Caching a historical multi-source thermal disturbance characteristic sequence through a sliding window, determining optimal delay time and embedding dimension based on a mutual information method and a false nearest neighbor method, and reconstructing a thermal evolution track in a high-dimensional phase space by utilizing delay embedding; Estimating the maximum Lyapunov exponent by calculating the exponent divergence rate of adjacent orbits in the reconstructed thermal evolution track, and judging that the thermal evolution of the system has chaos if the exponent is larger than zero, wherein the intervention is needed in advance; And constructing a thermal stability look-ahead confidence comprehensive reflection system future thermal behavior through the maximum Lyapunov exponent, the current thermal aberration risk exponent and the asymmetry of a recent thermal disturbance sequence, and directly driving a perception front-end to adaptively adjust.
  7. 7. The method for operating a projection camera with integrated temperature control function according to claim 6, wherein the historical multi-source thermal disturbance characteristic sequence is cached through a sliding window, the optimal delay time and the embedding dimension are determined based on a mutual information method and a false nearest neighbor method, and the thermal evolution track in the high-dimensional phase space is reconstructed by utilizing delay embedding, and the method specifically comprises the following sub-steps: the camera system maintains a sliding time window to continuously cache a historical multi-source heat disturbance characteristic value sequence; Determining optimal delay time by a mutual information method to ensure maximum independence among embedded coordinates, and adaptively selecting minimum embedded dimension by a false nearest neighbor method to ensure that a dynamic structure is not folded; The original sequence is mapped into a set of state vectors, thereby forming a continuous thermal evolution trace in the reconstructed phase space.
  8. 8. A system for operating a projection camera with integrated temperature control, comprising: the thermal disturbance module synchronously collects dark current of the infrared thermal field and the complementary metal oxide semiconductor image sensor and the duty cycle state of the micromirror array of the digital micromirror device, calculates disorder degree of the thermal field through an entropy algorithm, calculates photo-thermal phase mismatch degree based on photo-thermal phase difference, and calculates a multi-source thermal disturbance characteristic value according to disorder degree and mismatch degree; The thermodynamic diagram module takes the multisource thermal disturbance characteristic value as an initial excitation to drive a dynamic thermal diffusion diagram neural network embedded with a thermal conduction topology, inverts a transient thermal strain field of a key optical interface, projects the transient thermal strain field to a Zernike aberration basis function space to generate a thermal aberration risk index, and generates a risk thermodynamic diagram based on the index; The intervention grade module is used for calculating an active temperature control intervention grade through a self-adaptive threshold curved surface and a fuzzy reasoning engine based on the thermal aberration risk index and an environmental convection intensity estimated value based on the micro-electromechanical system anemometer array, and dynamically designating the combination strategy parameters of liquid cooling, thermoelectric refrigeration and light source shaping; The optimization instruction module is used for constructing a non-convex collaborative optimization problem under multi-objective constraint according to the active temperature control intervention level, taking the time-varying upper bound minimization of the Lyapunov thermal energy functional as an objective function, carrying out iterative solution through a semi-definite relaxation and alternating direction multiplier method to obtain collaborative scheduling weight and generating an optimization instruction by combining a risk thermodynamic diagram; and the self-adaptive module is used for extracting the maximum Lyapunov index of the thermal evolution track based on the multisource thermal disturbance characteristic value sequence to represent the thermodynamic chaos degree, generating thermal stability prospective confidence degree based on the Lyapunov index and the current thermal aberration risk index, predicting future thermal evolution and carrying out self-adaptive regulation.

Description

Working method and system of projection camera integrated with temperature control function Technical Field The present invention relates to the field of cameras, and in particular, to a method and a system for operating a projection camera with an integrated temperature control function. Background In high precision optical imaging and active projection systems, thermal perturbations have become a core bottleneck that limits the long-term stability and imaging accuracy of the system. With the development of equipment to high power, high frame rate and microminiaturization, the problems of complex multi-source thermal coupling, difficult prediction and compensation of thermal aberration, low efficiency of resource utilization caused by perceived and controlled splitting and serious thermal instability caused by environmental disturbance are increasingly highlighted. Based on the above, the invention provides a projection camera with integrated temperature control function and a working method thereof. Disclosure of Invention The invention provides a working method of a projection camera integrating a temperature control function, which comprises the following steps: S10, synchronously collecting dark current of an infrared thermal field and a complementary metal oxide semiconductor image sensor and a micro mirror array duty state of a digital micro mirror device, calculating disorder degree of the thermal field through an entropy algorithm, calculating photo-thermal phase mismatch degree based on photo-thermal phase difference, and calculating a multi-source thermal disturbance characteristic value according to the disorder degree and the mismatch degree; S20, taking a multisource thermal disturbance characteristic value as an initial excitation to drive a dynamic thermal diffusion diagram neural network with embedded thermal conduction topology, inverting a transient thermal strain field of a key optical interface, projecting the transient thermal strain field to a Zernike aberration basis function space to generate a thermal aberration risk index, and generating a risk thermodynamic diagram based on the index; S30, calculating an active temperature control intervention level through a self-adaptive threshold curved surface and a fuzzy reasoning engine based on a thermal aberration risk index and an environmental convection intensity estimated value based on a micro-electromechanical system anemometer array, and dynamically designating a combination strategy parameter of liquid cooling, thermoelectric refrigeration and light source shaping; S40, constructing a non-convex collaborative optimization problem under multi-objective constraint according to an active temperature control intervention level, taking the time-varying upper bound minimization of a Lyapunov thermal energy functional as an objective function, carrying out iterative solution by a semi-definite relaxation and alternating direction multiplier method to obtain collaborative scheduling weight, and generating an optimization instruction by combining a risk thermodynamic diagram; s50, extracting the maximum Lyapunov index of the thermal evolution track based on the multi-source thermal disturbance characteristic value sequence to represent the thermodynamic chaos degree, generating a thermal stability prospective confidence degree based on the Lyapunov index and the current thermal aberration risk index, predicting future thermal evolution, and performing self-adaptive regulation. The working method of the projection camera integrating the temperature control function comprises the following steps of synchronously collecting dark current of an infrared thermal field and a complementary metal oxide semiconductor image sensor and a micro mirror array duty state of a digital micro mirror device, calculating disorder degree of the thermal field through an entropy algorithm, calculating mismatch degree of photo-thermal phase based on photo-thermal phase difference, and calculating a multi-source thermal disturbance characteristic value according to the disorder degree and the mismatch degree, wherein the working method comprises the following steps of: The infrared space temperature field, the dark current noise distribution of the complementary metal oxide semiconductor image sensor and the micro-mirror array duty state map of the digital micro-mirror device are synchronously triggered and collected in parallel through microsecond level, and alignment is carried out; Calculating the disorder degree of the thermal field based on the spatial gradient directional entropy of the infrared spatial temperature field, calculating the photo-thermal phase mismatch degree according to the cross-correlation time delay of the light source modulation signal and the thermal response, and respectively quantifying the disorder degree of the thermal distribution and the hysteresis effect of the photo-thermal energy transfer; And generating a multi-source thermal di