System and method for detecting hidden camera by infrared and photoelectric combination

Abstract

This invention proposes a system and method for detecting hidden cameras using a combination of infrared and photoelectric sensors. The method includes: acquiring ambient temperature and light intensity distribution benchmark data using an infrared thermal imaging sensor and a photoelectric sensor, and establishing a feature benchmark library; emitting excitation light of different wavelengths and adjusting the supplementary light intensity according to the benchmark library; acquiring temperature distribution images at each supplementary light time point, calculating the temperature change rate matrix, and identifying abnormal regions of response time constant; acquiring the reflected light intensity matrix, calculating the reflected light intensity change rate matrix, and identifying abnormal reflectivity regions; generating a fusion feature matrix based on the temperature and reflected light intensity change rate matrices; and evaluating the similarity between the thermal response and photoelectric reflection distribution patterns in the neighborhood by constructing a feature fusion judgment function and calculating spatial correlation coefficients. When a preset threshold condition is met, the location is determined to be suspected of containing a hidden camera, and the candidate location coordinate set P is recorded. This invention can improve detection accuracy and reduce false alarm rate.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and specifically to a system and method for detecting hidden cameras using a combination of infrared and photoelectric technologies. Background Technology

[0002] When people stay or work in hotels, guesthouses, offices, conference rooms, and other similar venues, they often worry about the presence of hidden cameras due to concerns about personal privacy and information security. Currently, the search for hidden cameras mainly relies on human intervention or simple searches using devices such as mobile phones; however, existing methods are prone to false alarms and have low detection accuracy. Therefore, more effective methods are needed to address these issues and ensure people's privacy and information security. Summary of the Invention

[0003] Based on the above-mentioned problems, this invention proposes a system and method for detecting hidden cameras using a combination of infrared and photoelectric technologies. Through this invention, multi-dimensional feature recognition of hidden cameras is achieved, improving detection accuracy and reducing false alarm rate.

[0004] In view of this, one aspect of the present invention proposes a method for detecting hidden cameras using a combination of infrared and photoelectric technologies, comprising:

[0005] An infrared thermal imaging sensor is used to scan the area to be detected and obtain a baseline image of the ambient temperature distribution.

[0006] A baseline database of environmental photoelectric characteristics is established by collecting ambient light intensity distribution benchmark data through a photoelectric sensor array.

[0007] The multi-LED supplementary lighting system is controlled to emit excitation light of different wavelengths sequentially according to a preset time sequence;

[0008] The intensity of the supplementary light at each time point is adaptively adjusted according to the environmental photoelectric feature reference library to ensure that the excitation light can effectively penetrate the preset type of camouflage material;

[0009] At each supplemental lighting time point, the infrared thermal imaging sensor acquires temperature distribution images in real time;

[0010] Calculate the temperature change rate matrix;

[0011] Based on the heat conduction theory model, regions with abnormal response time constants were identified;

[0012] The photoelectric sensor array synchronously collects the reflected light intensity matrix at each supplementary lighting time node;

[0013] Calculate the matrix of the rate of change of reflected light intensity;

[0014] Extract the unique Fresnel reflection coefficient characteristics of the lens surface to identify abnormal areas with reflectivity in the range of 0.04-0.08;

[0015] Spatial location matching is performed between the temperature change rate matrix and the reflected light intensity change rate matrix to establish the feature correspondence of the same physical coordinate point and generate a fused feature matrix.

[0016] For each coordinate point Extract thermal response data sequences and photoelectric reflection data sequences within a neighborhood window centered at the point with a radius of δ, and calculate the spatial correlation coefficient. This is used to assess the similarity between the thermal response and photoelectric reflection distribution patterns in the neighborhood of that location;

[0017] Construct a feature fusion decision function ;

[0018] when >First preset thresholdTh and When the second preset threshold Tc is reached, it is determined that there is a hidden camera at that location, and the set of candidate location coordinates P is recorded.

[0019] Optionally, after the step of determining that there is a suspected hidden camera at the location, the method further includes:

[0020] For each suspected camera location in the candidate location coordinate set P The system controls the multi-LED supplementary lighting system to emit a preset modulated light signal of a specific frequency at that position.

[0021] Using photoelectric sensors at corresponding locations Detect the presence of feedback response caused by the camera's automatic gain control circuit and record the response strength. ;

[0022] Simultaneously, use infrared thermal imaging sensors to monitor the location. Changes in the thermal characteristics of the circuit at the location Changes in power consumption of the detection circuit;

[0023] Calculate the verification confidence level ,when When verifying the threshold Tv, the presence of a hidden camera at that location is finally confirmed, and the coordinates of the hidden camera and the overall confidence level are output.

[0024] Optionally, the feature fusion decision function uses an adaptive weight optimization algorithm to dynamically adjust the weight coefficients, specifically including:

[0025] Extract the feature vector of the current detection environment ;

[0026] Calculate the texture complexity index of the detection region ;

[0027] Establish a historical testing database ,in For historical environmental feature vectors, For the corresponding optimal weight combination, To improve detection accuracy;

[0028] Calculate the Euclidean distance between the current environmental feature E and the historical records. Select the K historical records with the smallest distance as the reference set. ;

[0029] Construct a three-layer feedforward neural network model: The input layer receives the enhanced environmental feature vector. ,in Minimum historical distance;

[0030] The hidden layer uses the ReLU activation function, and the output layer uses the Softmax normalization function to generate the predicted weights. ;

[0031] The formula for weighted prediction is: ;

[0032] Calculate the reference set Medium-weighted average Among them, weight ;

[0033] Establish an adaptive fusion function The fusion coefficient λx is determined through confidence level assessment: ;

[0034] The feature fusion decision function is updated using the final weight coefficients: ;

[0035] After completing the test and obtaining the verification results, calculate the accuracy index of this test. ;

[0036] if Preset threshold Then the current environmental feature E and the final weight will be... and accuracy Add it as a new record to the historical database H;

[0037] Simultaneously, the neural network parameters are updated using stochastic gradient descent. .

[0038] Optionally, the step of adaptively adjusting the supplementary light intensity at each time point based on an environmental photoelectric feature reference library to ensure that the excitation light can effectively penetrate a preset type of camouflage material includes:

[0039] Based on environmental photoelectric feature benchmark library Analyze the statistical characteristics of light intensity distribution and calculate the light intensity variance distribution map. ,in It is a local mean;

[0040] Extracting the spectral reflectance feature vector of the material ;

[0041] Establish a camouflage material classifier Identify camouflage material categories of a preset type by matching spectral features. ∈{fabric, plastic, metal, coating, composite material};

[0042] For each of the n1 wavelengths and identified material categories Query the material optical parameter database to obtain the transmission coefficient and absorption coefficient μ;

[0043] Calculate the required penetration power density ,in The minimum power density required for camera excitation, The estimated thickness of the camouflage material is preset;

[0044] Establish safe power constraints ,in The power limit for the corresponding wavelength is the safe power limit for the human eye.

[0045] Constructing a multi-objective optimization function ,in This is the wavelength weighting coefficient. This is the power consumption penalty factor;

[0046] The Lagrange multiplier method is used to solve the constrained optimization problem, and the optimal power allocation vector is obtained. ;

[0047] Calculate the driving current of each LED array. ;

[0048] At each time node Real-time monitoring of changes in ambient light intensity Calculate the ambient light intensity disturbance coefficient ;

[0049] Establish an adaptive supplementary light intensity adjustment function ;

[0050] Pulse width modulation (PWM) technology is used to control LED arrays of various wavelengths, with PWM duty cycle... This enables precise control of light intensity;

[0051] Based on the test results from the aforementioned steps, the penetration effect evaluation index is calculated. ,in The detected signal strength, The baseline signal strength when there is no camouflage;

[0052] like <threshold Then the power enhancement mode will be activated: Where γx is the enhancement coefficient;

[0053] Update the transmission coefficient and absorption coefficient in the material optical parameter database, establish a self-learning optimization mechanism, and improve the penetration effect of subsequent detection.

[0054] Optionally, the step of identifying regions with abnormal response time constants based on the heat conduction theory model includes:

[0055] At each illumination time point, the infrared thermal imaging sensor continuously acquires temperature distribution images, establishing a time-temperature response sequence for each pixel. ;

[0056] For the position of each pixel within the detection area Extract the complete thermal response curve from the start of supplemental lighting to the preset steady state, and record the changes during the temperature rise and temperature fall phases;

[0057] Identify and mark key time points in the response curve: the start time of supplemental lighting. Peak temperature End of supplemental lighting and temperature recovery time ;

[0058] The thermal response time constant of each pixel is calculated based on the first-order exponential response model. This parameter reflects the characteristic time required for a material to reach a steady-state temperature after receiving heat.

[0059] Extracting thermal response amplitude parameters This represents the temperature rise from the reference temperature to the peak temperature, reflecting the material's heat absorption capacity.

[0060] Calculate the thermal decay coefficient This describes the rate at which the temperature returns to the reference value after the supplemental lighting stops, reflecting the heat dissipation characteristics of the material.

[0061] Establish the heat conduction feature vector for each pixel. ;

[0062] Establish a benchmark library of thermal conductivity characteristics for common surface materials, including standard thermal response parameter ranges for material types such as wood, metal, plastic, fabric, glass, and ceramics;

[0063] For each material type, a normal range for the thermal response time constant is defined. arrive And the normal range of the corresponding thermal response amplitude and attenuation coefficient;

[0064] Based on the unique thermal characteristics of camera devices, thermal response feature templates are established for components such as camera chips, lens assemblies, and circuit boards; the typical time constant range for camera chips is 0.5 seconds to 2.0 seconds.

[0065] The heat conduction feature vector of each pixel The similarity of the features is calculated by matching and comparing them with the material features in the benchmark library.

[0066] Identify the thermal response time constant Pixels that are significantly deviated from the normal range of the surrounding materials;

[0067] Specially marked Pixels that fall within the feature range of the camera device are clustered into candidate camera regions.

[0068] Candidate regions are filtered by shape and size to exclude abnormal thermal response regions that clearly do not conform to the camera's geometric characteristics;

[0069] For each candidate camera region, calculate the thermal feature consistency index to evaluate the uniformity of thermal response parameters of each pixel within the region;

[0070] Analyze the contrast of thermal characteristics between the candidate region and the surrounding background material, and calculate the significance of the difference in thermal response;

[0071] Establish a confidence scoring mechanism for thermal features, taking into account factors such as time constant matching degree, response amplitude rationality, and regional shape regularity;

[0072] The system outputs the location coordinates, area, and thermal feature confidence score of the abnormal thermal response region, providing accurate spatial positioning information for subsequent photoelectric feature extraction.

[0073] Optionally, the step of extracting the Fresnel reflection coefficient characteristics unique to the lens surface and identifying abnormal areas with reflectivity in the range of 0.04-0.08 includes:

[0074] Ambient light intensity distribution benchmark data collected by photoelectric sensor array Preprocessing is performed to remove environmental noise interference and establish environmental reflectivity benchmark values ​​for each coordinate point. ;

[0075] Controlling each LED unit in a multi-LED supplemental lighting system according to a preset spatial angle sequence Excitation light is emitted toward the area to be inspected to ensure that the incident beam can illuminate the potential lens surface from different angles;

[0076] The photoelectric sensor array synchronously acquires data at each incident angle. Corresponding reflected light intensity distribution data This forms an angle-reflected light intensity data matrix;

[0077] For each coordinate point Calculate its reflectance sequence at different incident angles. ,in The incident light intensity;

[0078] Analyze the degree of agreement between the reflectivity sequence and Fresnel's law of reflection, and extract the Fresnel reflection coefficient that reflects the optical properties of the lens surface;

[0079] The calculated Fresnel reflection coefficients at each coordinate point are compared and matched with the preset lens material reflectivity range of 0.04-0.08;

[0080] Identify the coordinates of points whose Fresnel reflection coefficient falls within this range and have typical lens surface reflection characteristics, and mark these coordinates as optical anomalous areas;

[0081] To eliminate interference signals caused by reflective objects such as metal surfaces and mirror materials, ensure that the identified abnormal areas have the optical reflection characteristics of the camera lens.

[0082] Optionally, the temperature change rate matrix and the matrix of the rate of change of reflected light intensity Spatial location matching is performed to establish feature correspondences for the same physical coordinate points, and a fused feature matrix is ​​generated. The steps include:

[0083] The spatial coordinate system of the infrared thermal imaging sensor and the photoelectric sensor array is calibrated to determine the relative positional relationship and field of view of the two sensor systems.

[0084] Establish a unified physical coordinate system and map the pixel coordinate system of the infrared thermal imaging sensor and the detection unit coordinate system of the photoelectric sensor array to the same spatial reference coordinate system;

[0085] Analysis of the temperature change rate matrix and the matrix of the rate of change of reflected light intensity Spatial resolution differences;

[0086] Interpolation resampling technology is used to unify the spatial resolution of the two matrices to the same grid density, ensuring that each physical coordinate point has a corresponding data value in both matrices;

[0087] Based on the time series of the multi-LED supplementary lighting system, the temperature change rate data and the reflected light intensity change rate data are timestamped and corrected.

[0088] Compensate for the response time difference between the infrared thermal imaging sensor and the photoelectric sensor to ensure that the thermal response data and photoelectric reflection data generated at the same supplementary lighting moment can correspond accurately;

[0089] Based on the established coordinate system mapping relationship, the temperature change rate matrix is... and the matrix of the rate of change of reflected light intensity Perform geometric distortion correction;

[0090] Eliminate spatial position deviations caused by sensor installation angle and lens distortion to achieve precise alignment of the same physical positions in the two matrices;

[0091] Traversing the corrected temperature change rate matrix For each coordinate point in the matrix, the corresponding rate of change of reflected light intensity is... Find data values ​​with the same coordinates in the middle;

[0092] Establish a one-to-one correspondence between the thermal response characteristic value and the photoelectric reflection characteristic value of each physical coordinate point to form a feature data pair at the coordinate point level;

[0093] The established feature data pairs are reorganized according to their spatial coordinates to generate a fused feature matrix containing both thermal response and photoelectric reflection information. ;

[0094] In the fusion feature matrix In the middle, each coordinate point It includes thermal response characteristic components from the temperature change rate matrix and photoelectric reflection characteristic components from the reflected light intensity change rate matrix, providing a data foundation for subsequent correlation analysis and determination function calculation.

[0095] Optionally, for each suspected camera location in the candidate location coordinate set P The steps of controlling the multi-LED supplementary lighting system to emit a preset modulated light signal of a specific frequency at the designated location include:

[0096] Based on the feature fusion judgment function value of each position in the candidate position coordinate set P Spatial correlation coefficient Candidate positions are sorted in descending order of detection confidence.

[0097] Establish a verification queue and prioritize processing suspected camera locations with the highest confidence level to improve verification efficiency;

[0098] Based on the current candidate location coordinates to be verified The spatial distribution of each LED unit in the multi-LED supplementary lighting system is determined, and the spatial distance and illumination angle of each LED unit to the target position are calculated.

[0099] Select the group of LED units that are closest to the target location and have the best illumination angle as the main illumination source, and select adjacent auxiliary LED units for supplementary illumination.

[0100] Based on the ambient light intensity reference value of the target location in the environmental optoelectronic feature reference library, the basic intensity of the modulated light signal is set to ensure that the modulated signal can effectively penetrate potential camouflage and obstruction.

[0101] The specific frequency of the modulated light signal is set as the preset sensitive frequency band of the camera's automatic gain control circuit, including a combination sequence of low-frequency modulated signals and high-frequency pulse signals;

[0102] By controlling the emission angle and beam spread of the main and auxiliary LED units corresponding to the selected primary illumination source, the beams of multiple LEDs are aligned with the target position. An area of ​​overlapping light spots is formed at this location;

[0103] By adjusting the luminous intensity distribution of each LED unit, a uniform and stable modulated light illumination area is formed at the target position, avoiding detection errors caused by uneven light intensity distribution.

[0104] According to the preset modulation signal timing, control the focus to the target position. The LED chip units are combined to emit modulated light signals of different frequencies in sequence;

[0105] The transmission duration for each modulation frequency is set according to the typical response time of the camera's automatic gain control circuit to ensure sufficient signal excitation time;

[0106] During the transmission of modulated optical signals, the target position is monitored in real time using an array of photoelectric sensors. The change in light intensity confirms that the modulated light signal has accurately reached the target area;

[0107] Simultaneously, infrared thermal imaging sensors are used to monitor temperature changes at the target location to verify whether light energy is effectively applied to the suspected camera location, providing a reference benchmark for subsequent response signal detection.

[0108] Optionally, the simultaneous use of an infrared thermal imaging sensor to monitor the position... Changes in the thermal characteristics of the circuit at the location The steps for detecting changes in circuit power consumption include:

[0109] Before the modulated light signal is emitted, the target position is determined using an infrared thermal imaging sensor. High-precision temperature measurements are performed on the circuit and its surrounding area to obtain the reference value of the circuit's static operating temperature at that location. ;

[0110] Record the thermal radiation distribution pattern at this location without external light excitation, and establish a reference library of static thermal characteristics of the circuit.

[0111] During the transmission of the modulated signal, the infrared thermal imaging sensor continuously monitors the target position in a high-frequency sampling mode. Temperature changes;

[0112] According to the time nodes of the modulated optical signal, the first real-time temperature data sequence under each modulation frequency excitation is collected synchronously. Establish temperature-time variation curves;

[0113] The first real-time temperature data sequence obtained Compared with the circuit's static operating temperature reference value Perform difference calculations to extract the temperature change sequence at the target location;

[0114] Analyze the correspondence between temperature change and the timing of the modulated optical signal, identify temperature fluctuation characteristics that are synchronized with the frequency change of the optical signal, and eliminate irrelevant factors;

[0115] Based on the typical power consumption variation characteristics of the camera image sensor chip and automatic gain control circuit under optical signal excitation, the temperature threshold range for circuit operating state changes is set.

[0116] Detect target location Pay close attention to whether there is a localized temperature rise that corresponds to the increased power consumption of the camera circuit, and especially the temperature gradient changes in the heat-generating areas of the chip.

[0117] The response time and attenuation characteristics of the temperature fluctuations were analyzed and compared with the thermal time constant of a typical camera circuit.

[0118] It identifies temperature change patterns with unique thermal response speeds of camera circuits, eliminating temperature change interference caused by the physical effects of material thermal expansion and contraction.

[0119] Based on the aforementioned analysis results, the target location is quantified. The intensity of changes in the thermal characteristics of the circuit ;

[0120] Taking into account multiple dimensions such as temperature change amplitude, response time characteristics, and synchronization with the excitation signal, thermal characteristic parameters reflecting the degree of circuit activity at this location are generated.

[0121] The thermal characteristic parameter is used as an important input data for the final verification confidence calculation and is fused with the photoelectric reflection response data for judgment.

[0122] Another aspect of the present invention provides a system for detecting hidden cameras using a combination of infrared and photoelectric sensors, for performing a method for detecting hidden cameras using a combination of infrared and photoelectric sensors, comprising: an infrared thermal imaging sensor, a photoelectric sensor, and a server;

[0123] The server is configured as follows:

[0124] An infrared thermal imaging sensor is used to scan the area to be detected and obtain a baseline image of the ambient temperature distribution.

[0125] A baseline database of environmental photoelectric characteristics is established by collecting ambient light intensity distribution benchmark data through a photoelectric sensor array.

[0126] The multi-LED supplementary lighting system is controlled to emit excitation light of different wavelengths sequentially according to a preset time sequence;

[0127] The intensity of the supplementary light at each time point is adaptively adjusted according to the environmental photoelectric feature reference library to ensure that the excitation light can effectively penetrate the preset type of camouflage material;

[0128] At each supplemental lighting time point, the infrared thermal imaging sensor acquires temperature distribution images in real time;

[0129] Calculate the temperature change rate matrix;

[0130] Based on the heat conduction theory model, regions with abnormal response time constants were identified;

[0131] The photoelectric sensor array synchronously collects the reflected light intensity matrix at each supplementary lighting time node;

[0132] Calculate the matrix of the rate of change of reflected light intensity;

[0133] Extract the unique Fresnel reflection coefficient characteristics of the lens surface to identify abnormal areas with reflectivity in the range of 0.04-0.08;

[0134] Spatial location matching is performed between the temperature change rate matrix and the reflected light intensity change rate matrix to establish the feature correspondence of the same physical coordinate point and generate a fused feature matrix.

[0135] For each coordinate point Extract thermal response data sequences and photoelectric reflection data sequences within a neighborhood window centered at the point with a radius of δ, and calculate the spatial correlation coefficient. This is used to assess the similarity between the thermal response and photoelectric reflection distribution patterns in the neighborhood of that location;

[0136] Construct a feature fusion decision function ;

[0137] when >First preset thresholdTh and When the second preset threshold Tc is reached, it is determined that there is a hidden camera at that location, and the set of candidate location coordinates P is recorded.

[0138] The system and method for detecting hidden cameras using infrared and photoelectric sensors, employing the technical solution of this invention, achieves multi-dimensional feature recognition of hidden cameras through the coordinated operation of infrared thermal imaging and photoelectric sensors, improving detection accuracy and reducing false alarm rate. Multi-spectral supplementary lighting technology can effectively penetrate common camouflage materials, overcoming the problem of single detection methods being susceptible to environmental interference, and maintaining high detection accuracy even in complex environments. It supports the detection of various types of camera devices, including pinhole cameras, wireless cameras, and night vision cameras, and is suitable for various scenarios such as hotels, offices, and conference rooms. Attached Figure Description

[0139] Figure 1 This is a flowchart of a method for detecting hidden cameras using a combination of infrared and photoelectric technologies according to an embodiment of the present invention;

[0140] Figure 2 This is a schematic block diagram of a system for detecting hidden cameras using a combination of infrared and photoelectric methods, provided in one embodiment of the present invention. Detailed Implementation

[0141] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0142] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0143] The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0144] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0145] The following reference Figures 1 to 2 This invention describes a system and method for detecting hidden cameras using a combination of infrared and photoelectric technologies, provided by some embodiments of the present invention.

[0146] like Figure 1 As shown, one embodiment of the present invention provides a method for detecting hidden cameras using a combination of infrared and photoelectric methods, comprising:

[0147] An infrared thermal imaging sensor is used to scan the area to be detected and acquire a baseline image of the ambient temperature distribution. ;

[0148] In this step, the infrared thermal imaging sensor detects the infrared radiation energy emitted by the object's surface, converts the thermal radiation signal into an electrical signal, and then generates a visualized temperature distribution image through signal processing. Objects at different temperatures emit infrared radiation of different intensities, and the sensor can detect these minute differences in radiation and quantify them into temperature values. This step performs systematic temperature measurements on the entire area to be detected, covering all possible locations where cameras may be hidden. The scanning process includes point-by-point or area-by-area temperature acquisition in both horizontal and vertical directions. This means in coordinates The reference temperature value at the location is a two-dimensional temperature matrix that records the initial temperature state of each spatial location. The reference image serves as the "thermal background" of the environment, providing a comparative reference for subsequent temperature change detection and recording the natural temperature distribution at each location under conditions without external excitation.

[0149] Ambient light intensity distribution baseline data is collected using a photoelectric sensor array. Establish a reference library of environmental photoelectric characteristics;

[0150] In this step, the photoelectric sensor converts the light signal into an electrical signal and quantifies the light intensity by measuring the photocurrent or photovoltage. The array configuration means that multiple photoelectric sensors are arranged in a certain spatial distribution to form a sensor network. Each sensor unit is responsible for detecting the light intensity information at its corresponding spatial location. Different positions, angles and surface materials have different reflection and absorption characteristics of ambient light. The sensor array can simultaneously collect light intensity data from multiple spatial locations to form a data set that reflects the light distribution of the entire detection area. This means in coordinates The reference light intensity value at the location is a two-dimensional light intensity matrix that records the initial illumination state at each spatial location. The reference data reflects the natural ambient light distribution under conditions without artificial lighting, including the natural reflection characteristics of various material surfaces in the environment, providing a comparative benchmark for subsequent light intensity change detection. The reference library contains not only simple numerical light intensity data. It also includes photoelectric characteristic parameters for various locations, such as reflectivity and scattering characteristics, and may contain multiple sets of benchmark data under different time periods and ambient light conditions. The feature library records the photoelectric response characteristics of various common materials in the detection environment, providing reference templates for identifying abnormal photoelectric characteristics and supporting adaptive light intensity adjustment and anomaly judgment. The environmental photoelectric characteristic benchmark library is essentially the "optical fingerprint library" of the entire detection system, recording the optical "identity card" of various materials and locations in the detection environment. When subsequent supplementary lighting excitation occurs, the system can quickly identify locations with abnormal optical response characteristics by comparing with the benchmark library, thereby locking down areas where cameras may be hidden.

[0151] Control the multi-LED supplemental lighting system according to a preset time sequence Emit excitation light of different wavelengths in sequence. The excitation light includes visible light, near-infrared light, and mid-infrared light bands.

[0152] In this step, an array lighting system composed of multiple LED beads of different types is used. Each bead or group of beads is responsible for emitting light of a specific wavelength and has an independent control circuit, capable of individually controlling the on / off state and intensity of each bead. Time series This represents a preset time sequence, with each time point corresponding to the emission time of a specific wavelength of light. The time intervals can be equal or unequal, and are optimized according to detection requirements. The total time sequence length and interval design need to balance detection accuracy and detection speed. Sequential emission avoids spectral aliasing and signal interference caused by simultaneous emission of multiple wavelengths, ensures that the excitation effect of each wavelength can be accurately recorded and analyzed, and provides independent time windows for the responses of different wavelengths. Wavelength The selection principle is based on the optimized selection of the spectral response characteristics of the camera sensor; considering the transmission and reflection characteristics of common camouflage materials; and covering the sensitive bands of the camera's automatic gain control circuit. This multi-wavelength timing excitation method enables "multi-dimensional detection" of hidden cameras, much like using different "keys" to "unlock" different response mechanisms of the camera, greatly improving the success rate and accuracy of detection. Even if the camera is partially obscured or disguised, there will always be some wavelength of excitation that can generate a detectable response signal.

[0153] The intensity of the supplementary light at each time point is adaptively adjusted according to the environmental photoelectric feature reference library to ensure that the excitation light can effectively penetrate the preset type of camouflage material;

[0154] At each supplemental lighting time point, the infrared thermal imaging sensor acquires temperature distribution images in real time. ;

[0155] In this step, each time node For a specific supplemental lighting moment, when the multi-LED system emits a wavelength When the excitation light is applied, the infrared thermal imaging sensor is simultaneously triggered to collect temperature data, achieving precise time synchronization between light excitation and temperature detection. The sensor measures the temperature immediately upon or immediately after the supplementary light excitation, capturing the instantaneous or short-term temperature changes caused by the light excitation and avoiding the loss of temperature change information due to time delay. This means in time ,coordinate The temperature values ​​at the location form a three-dimensional data set of time-space-temperature. Each time point generates a complete temperature distribution image, where each pixel represents the temperature value at the corresponding spatial location. Images from different time points reflect the temperature response under different wavelength excitations. Camera devices exhibit different temperature response patterns under light excitation compared to their surroundings. Functional modules such as the image sensor chip and automatic gain control circuitry change their operating states due to light excitation, resulting in changes in power consumption and heat generation. By acquiring temperature images from multiple time points, a time series of temperature changes is established. Different wavelength excitations may cause different degrees of temperature response, providing a complete data foundation for subsequent temperature change rate calculations. This real-time temperature acquisition mechanism enables "thermal imaging tracking" of hidden cameras, much like installing a "thermometer" at each suspicious location, continuously monitoring its thermal response under different light excitations. In this way, even cameras that are completely hidden externally will leave a "thermal fingerprint" in the temperature image if their internal circuitry responds to light excitation, thus being detected by the detection system.

[0156] Calculate the temperature change rate matrix :

[0157] ;

[0158] Based on the thermal conduction theory model, regions with abnormal response time constants were identified (where the thermal response characteristics of the camera chip are τ = 0.5-2.0 seconds).

[0159] The photoelectric sensor array synchronously acquires the reflected light intensity matrix at each supplementary lighting time point. ;

[0160] In this step, the photoelectric sensor array maintains strict time synchronization with the multi-LED supplementary lighting system at each supplementary lighting time node. When wavelength When the excitation light is emitted, the photoelectric sensor simultaneously starts data acquisition to ensure that the acquired reflected light intensity data is precisely matched with the corresponding excitation light wavelength. Synchronous acquisition can avoid the mismatch between light intensity data and excitation source caused by time deviation, capture the instantaneous reflection response after excitation light irradiation, and provide an accurate time-light intensity correspondence for subsequent reflection characteristic analysis. Indicates time ,coordinate The detected reflected light intensity values ​​at the location form a three-dimensional data set of time-space-intensity. Each time node generates a complete reflected light intensity distribution matrix, where each element represents the reflected light intensity at the corresponding spatial location. The matrices at different time nodes reflect the reflection response patterns under different wavelength excitations. This synchronous reflected light intensity acquisition mechanism enables "optical fingerprint recognition" of hidden cameras, much like using an "optical microscope" to observe the reflection characteristics of each suspicious location. In this way, even tiny lens surfaces can be identified due to their unique optical reflection characteristics, providing important optical dimensional evidence for camera detection.

[0161] Calculate the matrix of the rate of change of reflected light intensity :

[0162] ;

[0163] Extract the unique Fresnel reflection coefficient characteristics of the lens surface to identify abnormal areas with reflectivity in the range of 0.04-0.08;

[0164] temperature change rate matrix and the matrix of the rate of change of reflected light intensity Spatial location matching is performed to establish feature correspondences for the same physical coordinate points, and a fused feature matrix is ​​generated. ;

[0165] For each coordinate point Extract the thermal response data sequence within a neighborhood window centered at the point with a radius of δ. and photoelectric reflection data sequence Calculate the spatial correlation coefficient This is used to assess the similarity between the thermal response and photoelectric reflection distribution patterns in the neighborhood of that location;

[0166] In this step, Corr represents the Pearson correlation coefficient calculation function; and This means that by extracting data from all pixels within a circular neighborhood centered at (x,y) with radius δ, two data sequences are formed: a thermal response sequence and a photoelectric reflection sequence. Calculate the Pearson correlation coefficient between the two sequences, with the result ranging from [value missing]. ;when When the thermal response and photoelectric reflection are highly consistent in spatial distribution, it is very likely a camera; when When the two responses are uncorrelated, it indicates that there may be noise or interference; when When the time is negative, it indicates a false correlation, which usually indicates a false signal.

[0167] Construct a feature fusion decision function , where α, β, and γ are weighting coefficients;

[0168] In this step, because single features are prone to misjudgment, multi-dimensional fusion is used to improve detection accuracy. Thermal feature recognition identifies the working status, photoelectric feature recognition identifies the physical structure, and correlation verification verifies spatial consistency. These three aspects mutually verify each other to reduce the false judgment rate. The weight coefficients α, β, and γ can be dynamically adjusted according to environmental conditions and material properties. Under different camouflage scenarios, the weights can be adjusted to optimize the detection effect. The weight coefficients can obtain the optimal weight combination through machine learning training, can be determined statistically based on the feature distribution of known camera samples, or can be optimized by combining expert experience and experimental data.

[0169] when >First preset thresholdTh and When the second preset threshold Tc is reached, it is determined that there is a hidden camera at that location, and the set of candidate location coordinates is recorded. .

[0170] In this step, the overall feature intensity must reach a set threshold and the spatial correlation must be high enough; these two conditions ensure the reliability of the detection.

[0171] The technical solution adopted in this embodiment achieves multi-dimensional feature recognition of hidden cameras through the coordinated work of infrared thermal imaging and photoelectric sensors, thereby improving detection accuracy and reducing false alarm rate. The multi-spectral supplementary lighting technology can effectively penetrate common camouflage materials, overcome the problem of single detection methods being susceptible to environmental interference, and maintain high detection accuracy even in complex environments. It supports the detection of various types of camera devices, including pinhole cameras, wireless cameras, and night vision cameras, and is suitable for various scenarios such as hotels, offices, and conference rooms.

[0172] In some possible embodiments of the present invention, after the step of determining that there is a suspected hidden camera at the location, the method further includes:

[0173] For each suspected camera location in the candidate location coordinate set P The system controls the multi-LED supplementary lighting system to emit a preset modulated light signal of a specific frequency at that position.

[0174] Using photoelectric sensors at corresponding locations Detect the presence of feedback response caused by the camera's automatic gain control (AGC) circuit and record the response strength. ;

[0175] Simultaneously, use infrared thermal imaging sensors to monitor the location. Changes in the thermal characteristics of the circuit at the location Changes in power consumption of the detection circuit;

[0176] Calculate the verification confidence level ,when When verifying the threshold Tv, the presence of a hidden camera at that location is finally confirmed, and the coordinates of the hidden camera and the overall confidence level are output.

[0177] In this embodiment, AGC (Automatic Gain Control) is the core circuit function of the camera. When the external lighting changes, the camera automatically adjusts its gain to maintain image quality. This solution activates this circuit response by emitting a modulated light signal of a specific frequency. The multi-LED supplementary lighting system precisely targets the suspected location and emits modulated light. If a camera is indeed located at that location, its AGC circuit will respond to the change in lighting. The photoelectric sensor detects the change in reflected light intensity caused by the AGC adjustment. A higher value indicates a more pronounced AGC response, suggesting a higher likelihood of a camera being present. When a camera is operating, especially during AGC circuit response, it generates additional power consumption. These power consumption changes directly lead to minute variations in chip temperature, exhibiting specific temporal and amplitude characteristics. The infrared thermal imaging sensor monitors this temperature change in real time; when modulated light excites the AGC circuit, it causes a momentary increase in power consumption. This reflects the thermal characteristics caused by changes in the circuit's operating state; the larger the value, the more pronounced the circuit activity. w1 controls the importance of the AGC response characteristics, while w2 controls the contribution of thermal characteristic changes; their weights can be adjusted according to the characteristics of different camera types. The solution in this embodiment can adapt to different environmental conditions and achieve optimal detection results without manual adjustment. Through a dual confirmation mechanism of modulated optical signal verification and circuit thermal characteristic analysis, misjudgments are effectively avoided, ensuring the reliability of the detection results.

[0178] In some possible embodiments of the present invention, the feature fusion determination function employs an adaptive weight optimization algorithm to dynamically adjust the weight coefficients, specifically including:

[0179] Extract the feature vector of the current detection environment ,in For the average ambient temperature, For ambient light intensity, For relative humidity, The main material type of the detection area is coded;

[0180] Calculate the texture complexity index of the detection region ,in The grayscale value of the visible light image. For the Laplacian operator, N×M is the image size, where N represents the number of pixels in the vertical direction and M represents the number of pixels in the horizontal direction.

[0181] Establish a historical testing database ,in For historical environmental feature vectors, For the corresponding optimal weight combination, To improve detection accuracy;

[0182] Calculate the Euclidean distance between the current environmental feature E and the historical records. Select the K historical records with the smallest distance as the reference set. ;in, This represents the i-th component of the current environment feature vector; This represents the i-th component of the historical environment feature vector;

[0183] In this step, i is an index variable, ranging from 1 to 4, corresponding to four environmental feature components: when i=1, it corresponds to the average environmental temperature. When i=2, the corresponding ambient light intensity The relative humidity when i=3 Material type code when i=4 This is the standard mathematical expression for calculating the Euclidean distance between two four-dimensional feature vectors, used to measure the similarity between the current detection environment and the historical detection environment.

[0184] Construct a three-layer feedforward neural network model: The input layer receives the enhanced environmental feature vector. ,in Minimum historical distance;

[0185] The hidden layer uses the ReLU activation function, and the output layer uses the Softmax normalization function to generate the predicted weights. ;

[0186] The formula for weighted prediction is: , where W1 and W2 are weight matrices, and b1 and b2 are bias vectors;

[0187] Calculate the reference set H ref Medium-weighted average ,in, For the optimal weight combination in the j-th historical record, the weights are... , σ represents the distance between the current environment and the j-th historical environment, and σ is the temperature parameter.

[0188] Establish an adaptive fusion function The fusion coefficient λx is determined through confidence level assessment: D1, D2, and D3 are the learning parameters;

[0189] The feature fusion decision function is updated using the final weight coefficients: ;

[0190] After completing the test and obtaining the verification results, calculate the accuracy index of this test. ;

[0191] if Preset threshold Then the current environmental feature E and the final weight will be... and accuracy Add it as a new record to the historical database H;

[0192] Simultaneously, the neural network parameters are updated using stochastic gradient descent. Where η is the learning rate and L is the loss function. The loss function L represents the parameters. The gradient enables continuous learning and optimization of the system.

[0193] In this embodiment, by constructing a multidimensional environmental feature vector and a texture complexity index (TCM), the system can accurately identify the characteristics of different detection environments and automatically adjust the detection strategy, maintaining high detection accuracy even under conditions of temperature and illumination changes. The established historical detection database H can store and utilize past detection experience. Through Euclidean distance matching and weighted fusion, the system's detection capability continuously improves with the number of uses. The dual-weight fusion mechanism combining the neural network model with historical data can quickly complete the real-time adjustment of weight coefficients, improving the detection accuracy in complex environments compared to a fixed-weight scheme. The online learning and update mechanism enables the system to continuously adapt to new camera types and hiding methods without retraining the entire model, adapting to new environments or new device types. The adaptive fusion coefficient λx controlled by the sigmoid function ensures the smoothness and stability of weight adjustment, avoiding parameter oscillations and making the detection results more reliable.

[0194] In some possible embodiments of the present invention, the step of adaptively adjusting the supplementary light intensity at each time point according to an environmental photoelectric feature reference library to ensure that the excitation light can effectively penetrate a preset type of camouflage material includes:

[0195] Based on environmental photoelectric feature benchmark library Analyze the statistical characteristics of light intensity distribution and calculate the light intensity variance distribution map. ,in It is a local mean;

[0196] Extracting the spectral reflectance feature vector of the material ,in , , These are the reflectance values ​​for the visible, near-infrared, and mid-infrared bands, respectively. The optical attenuation coefficient;

[0197] Establish a camouflage material classifier Identify camouflage material categories of a preset type by matching spectral features. ∈{fabric, plastic, metal, coating, composite material};

[0198] For each of the n1 wavelengths and identified material categories Query the material optical parameter database to obtain the transmission coefficient and absorption coefficient μ;

[0199] Calculate the required penetration power density ,in The minimum power density required for camera excitation, The estimated thickness of the camouflage material is preset;

[0200] Establish safe power constraints ,in The power limit for the corresponding wavelength is the safe power limit for the human eye.

[0201] Constructing a multi-objective optimization function ,in This is the wavelength weighting coefficient. This is the power consumption penalty factor;

[0202] The Lagrange multiplier method is used to solve the constrained optimization problem, and the optimal power allocation vector is obtained. ;

[0203] Calculate the driving current of each LED array. ,in For LED efficiency, Thermal resistance;

[0204] At each time node Real-time monitoring of changes in ambient light intensity Calculate the ambient light intensity disturbance coefficient ;

[0205] Establish an adaptive supplementary light intensity adjustment function ,in This is the environmental disturbance compensation coefficient. For modulation depth, The modulation frequency;

[0206] Pulse Width Modulation (PWM) technology is used to control LED arrays of various wavelengths, with PWM duty cycle... This enables precise control of light intensity;

[0207] Based on the test results from the aforementioned steps, the penetration effect evaluation index is calculated. ,in The detected signal strength, The baseline signal strength when there is no camouflage;

[0208] like <threshold Then the power enhancement mode will be activated: Where γx is the enhancement coefficient;

[0209] Update the transmission coefficient and absorption coefficient in the material optical parameter database, establish a self-learning optimization mechanism, and improve the penetration effect of subsequent detection.

[0210] In this embodiment, spectral feature recognition and power calculation effectively penetrate common camouflage materials (fabric, plastic film, coating, etc.), improving penetration efficiency. Multi-objective optimization using the Lagrange multiplier method ensures optimal power allocation across bands, improving detection rate in camouflaged environments compared to single-band detection. A dynamic supplementary lighting intensity adjustment mechanism maintains stable detection performance under varying ambient light intensity, shortening response time. Constraints and precise PWM control ensure that the supplementary light power is always within the safe range for the human eye, while maximizing the penetration effect; the continuous updating mechanism of the material optical parameter database enables the system to adapt to new camouflage materials, and the learning efficiency increases with the number of uses.

[0211] In some possible embodiments of the present invention, the step of identifying regions with abnormal response time constants based on the heat conduction theory model includes:

[0212] At each illumination time point, the infrared thermal imaging sensor continuously acquires temperature distribution images, establishing a time-temperature response sequence for each pixel. ;

[0213] For the position of each pixel within the detection area Extract the complete thermal response curve from the start of supplemental lighting to the preset steady state, and record the changes during the temperature rise and temperature fall phases;

[0214] Identify and mark key time points in the response curve: the start time of supplemental lighting. Peak temperature End of supplemental lighting and temperature recovery time ;

[0215] The thermal response time constant of each pixel is calculated based on the first-order exponential response model. This parameter reflects the characteristic time required for a material to reach a steady-state temperature after receiving heat.

[0216] Extracting thermal response amplitude parameters This represents the temperature rise from the reference temperature to the peak temperature, reflecting the material's heat absorption capacity.

[0217] Calculate the thermal decay coefficient This describes the rate at which the temperature returns to the reference value after the supplemental lighting stops, reflecting the heat dissipation characteristics of the material.

[0218] Establish the heat conduction feature vector for each pixel. ;

[0219] Establish a benchmark library of thermal conductivity characteristics for common surface materials, including standard thermal response parameter ranges for material types such as wood, metal, plastic, fabric, glass, and ceramics;

[0220] For each material type, a normal range for the thermal response time constant is defined. arrive And the normal range of the corresponding thermal response amplitude and attenuation coefficient;

[0221] Based on the unique thermal characteristics of camera devices, thermal response feature templates are established for components such as camera chips, lens assemblies, and circuit boards; the typical time constant range for camera chips is 0.5 seconds to 2.0 seconds.

[0222] The heat conduction feature vector of each pixel The similarity of the features is calculated by matching and comparing them with the material features in the benchmark library.

[0223] Identify the thermal response time constant Pixels that are significantly deviated from the normal range of the surrounding material (the areas corresponding to these pixels may contain objects that are different from the surrounding material).

[0224] Specially marked Pixels that fall within the feature range of the camera device are clustered into candidate camera regions.

[0225] Candidate regions are filtered by shape and size to exclude abnormal thermal response regions that clearly do not conform to the camera's geometric characteristics;

[0226] For each candidate camera region, calculate the thermal feature consistency index to evaluate the uniformity of thermal response parameters of each pixel within the region;

[0227] Analyze the contrast of thermal characteristics between the candidate region and the surrounding background material, and calculate the significance of the difference in thermal response;

[0228] Establish a confidence scoring mechanism for thermal features, taking into account factors such as time constant matching degree, response amplitude rationality, and regional shape regularity;

[0229] The system outputs the location coordinates, area, and thermal feature confidence score of the abnormal thermal response region, providing accurate spatial positioning information for subsequent photoelectric feature extraction.

[0230] In this embodiment, by establishing a complete time-temperature response sequence and identifying key time nodes, subtle differences in thermal conductivity between the camera device and the background material can be accurately captured, improving recognition accuracy. The established thermal conductivity feature benchmark library covers standard thermal response parameters of common surface materials, effectively distinguishing camera chips and lens components from background materials such as wood, metal, and plastic, avoiding misjudgments caused by material confusion. Simultaneously, the feature parameters of three dimensions—thermal response time constant τ, thermal response amplitude A, and thermal attenuation coefficient λ—are analyzed, improving detection accuracy compared to single-parameter judgment. Through pixel-level thermal feature analysis and region clustering, the camera position can be accurately located, controlling spatial positioning errors and providing an accurate spatial benchmark for subsequent photoelectric feature matching. Through thermal feature contrast analysis and confidence assessment mechanisms, interference from other heat sources in the environment (such as electrical equipment, radiators, etc.) is effectively eliminated, reducing the false alarm rate. Continuous monitoring of the complete thermal response process from the start of supplementary lighting to temperature recovery can identify dynamic thermal feature changes of the camera device in the working state, improving the detection capability of concealed cameras.

[0231] In some possible embodiments of the present invention, the step of extracting the Fresnel reflection coefficient characteristics unique to the lens surface and identifying abnormal regions with reflectivity in the range of 0.04-0.08 includes:

[0232] Ambient light intensity distribution benchmark data collected by photoelectric sensor array Preprocessing is performed to remove environmental noise interference and establish environmental reflectivity benchmark values ​​for each coordinate point. ;

[0233] Controlling each LED unit in a multi-LED supplemental lighting system according to a preset spatial angle sequence Excitation light is emitted toward the area to be inspected to ensure that the incident beam can illuminate the potential lens surface from different angles;

[0234] The photoelectric sensor array synchronously collects the reflected light intensity distribution data corresponding to each incident angle θᵢ. This forms an angle-reflected light intensity data matrix;

[0235] For each coordinate point Calculate its reflectance sequence at different incident angles. ,in The incident light intensity;

[0236] Analyze the degree of agreement between the reflectivity sequence and Fresnel's law of reflection, and extract the Fresnel reflection coefficient that reflects the optical properties of the lens surface;

[0237] The calculated Fresnel reflection coefficients at each coordinate point are compared and matched with the preset lens material reflectivity range of 0.04-0.08;

[0238] Identify the coordinates of points whose Fresnel reflection coefficient falls within this range and have typical lens surface reflection characteristics, and mark these coordinates as optical anomalous areas;

[0239] To eliminate interference signals caused by reflective objects such as metal surfaces and mirror materials, ensure that the identified abnormal areas have the optical reflection characteristics of the camera lens.

[0240] The solution in this embodiment, through multi-angle incident beam and Fresnel reflection coefficient analysis, can accurately distinguish camera lenses from other highly reflective surfaces, significantly reducing the false alarm rate. Utilizing the inherent optical properties of the lens surface, even if the camera is partially obscured or disguised, it can still be identified through its unique Fresnel reflection characteristics. By setting a specific reflectivity range (0.04-0.08) and matching Fresnel reflection characteristics, interference from materials such as metal and glass can be effectively eliminated, improving the reliability of detection. Qualitative optical phenomena are transformed into quantifiable Fresnel reflection coefficients, providing a precise numerical basis for subsequent multi-dimensional feature fusion and judgment. The preset reflectivity range screening mechanism reduces candidate regions requiring further analysis, improving the overall efficiency of the detection algorithm.

[0241] In some possible embodiments of the present invention, the temperature change rate matrix ΔT(x,y) and the reflected light intensity change rate matrix are... Spatial location matching is performed to establish feature correspondences for the same physical coordinate points, and a fused feature matrix is ​​generated. The steps include:

[0242] The spatial coordinate system of the infrared thermal imaging sensor and the photoelectric sensor array is calibrated to determine the relative positional relationship and field of view of the two sensor systems.

[0243] Establish a unified physical coordinate system and map the pixel coordinate system of the infrared thermal imaging sensor and the detection unit coordinate system of the photoelectric sensor array to the same spatial reference coordinate system;

[0244] Analysis of the temperature change rate matrix and the matrix of the rate of change of reflected light intensity Spatial resolution differences;

[0245] Interpolation resampling technology is used to unify the spatial resolution of the two matrices to the same grid density, ensuring that each physical coordinate point has a corresponding data value in both matrices;

[0246] Based on the time series of the multi-LED supplementary lighting system, the temperature change rate data and the reflected light intensity change rate data are timestamped and corrected.

[0247] Compensate for the response time difference between the infrared thermal imaging sensor and the photoelectric sensor to ensure that the thermal response data and photoelectric reflection data generated at the same supplementary lighting moment can correspond accurately;

[0248] Based on the established coordinate system mapping relationship, the temperature change rate matrix is... and the matrix of the rate of change of reflected light intensity Perform geometric distortion correction;

[0249] Eliminate spatial position deviations caused by sensor installation angle and lens distortion to achieve precise alignment of the same physical positions in the two matrices;

[0250] Traversing the corrected temperature change rate matrix For each coordinate point in the matrix, the corresponding rate of change of reflected light intensity is... Find data values ​​with the same coordinates in the middle;

[0251] Establish a one-to-one correspondence between the thermal response characteristic value and the photoelectric reflection characteristic value of each physical coordinate point to form a feature data pair at the coordinate point level;

[0252] The established feature data pairs are reorganized according to their spatial coordinates to generate a fused feature matrix containing both thermal response and photoelectric reflection information. ;

[0253] In the fusion feature matrix In the middle, each coordinate point It includes thermal response characteristic components from the temperature change rate matrix and photoelectric reflection characteristic components from the reflected light intensity change rate matrix, providing a data foundation for subsequent correlation analysis and determination function calculation.

[0254] The scheme in this embodiment ensures precise spatial correspondence between thermal imaging data and photoelectric data through rigorous coordinate system calibration and geometric correction, avoiding feature mismatch caused by sensor position deviation. Multimodal feature fusion can comprehensively utilize two different physical properties, thermal response and optical reflection, to mutually verify and complement each other, significantly reducing the false judgment rate of single sensor detection. Through time synchronization correction and spatial resolution normalization, it ensures good spatiotemporal consistency of data in the fused feature matrix, providing a reliable foundation for subsequent correlation analysis. The unified fused feature matrix avoids repetitive coordinate transformation and data lookup operations, improving the computational efficiency of subsequent feature analysis and judgment processes. The spatial correspondence of dual-modal features can effectively identify and eliminate accidental interference signals in the environment, improving the accuracy and stability of hidden camera detection. The standardized fused feature matrix provides a standardized data input format for subsequent correlation calculation, judgment function construction, and threshold comparison.

[0255] In some possible embodiments of the present invention, the step of targeting each suspected camera location in the candidate location coordinate set P... The steps of controlling the multi-LED supplementary lighting system to emit a preset modulated light signal of a specific frequency at the designated location include:

[0256] Based on the feature fusion judgment function value of each position in the candidate position coordinate set P Spatial correlation coefficient Candidate positions are sorted in descending order of detection confidence.

[0257] Establish a verification queue and prioritize processing suspected camera locations with the highest confidence level to improve verification efficiency;

[0258] Based on the current candidate location coordinates to be verified The spatial distribution of each LED unit in the multi-LED supplementary lighting system is determined, and the spatial distance and illumination angle of each LED unit to the target position are calculated.

[0259] Select the group of LED units that are closest to the target location and have the best illumination angle as the main illumination source, and select adjacent auxiliary LED units for supplementary illumination.

[0260] Based on the ambient light intensity reference value of the target location in the environmental optoelectronic feature reference library, the basic intensity of the modulated light signal is set to ensure that the modulated signal can effectively penetrate potential camouflage and obstruction.

[0261] The specific frequency of the modulated light signal is set as the preset sensitive frequency band of the camera's automatic gain control circuit, including a combination sequence of low-frequency modulated signals and high-frequency pulse signals;

[0262] By controlling the emission angle and beam spread of the main and auxiliary LED units corresponding to the selected primary illumination source, the beams of multiple LEDs are aligned with the target position. An area of ​​overlapping light spots is formed at this location;

[0263] By adjusting the luminous intensity distribution of each LED unit, a uniform and stable modulated light illumination area is formed at the target position, avoiding detection errors caused by uneven light intensity distribution.

[0264] According to the preset modulation signal timing, control the focus to the target position. The LED chip units are combined to emit modulated light signals of different frequencies in sequence;

[0265] The transmission duration for each modulation frequency is set according to the typical response time of the camera's automatic gain control circuit to ensure sufficient signal excitation time;

[0266] During the transmission of modulated optical signals, the target position is monitored in real time using an array of photoelectric sensors. The change in light intensity confirms that the modulated light signal has accurately reached the target area;

[0267] Simultaneously, infrared thermal imaging sensors are used to monitor temperature changes at the target location to verify whether light energy is effectively applied to the suspected camera location, providing a reference benchmark for subsequent response signal detection.

[0268] In this embodiment, multi-LED collaborative focusing control enables precise projection of modulated light signals onto the specific location of a suspected camera, avoiding energy waste and signal interference caused by large-area illumination. The modulation signal is designed for a specific response frequency band of the camera's automatic gain control circuit, significantly improving the success rate of exciting the camera circuit response and the specificity of detection. Adaptive intensity adjustment based on environmental baseline data ensures that the modulated light signal can effectively penetrate common camouflage materials and reach the hidden camera surface. Prioritization and serialization of candidate positions reduce unnecessary verification time and improve the overall efficiency of the detection system. The uniform light field distribution formed by multi-LED collaborative illumination avoids blind spots and uneven intensity caused by single-point light sources, improving the stability of response signal detection. Real-time monitoring of the illumination effect and the photoelectric and thermal response of the target area ensures that each candidate position receives sufficient and effective verification stimulation, providing a reliable data foundation for final camera confirmation.

[0269] In some possible embodiments of the present invention, the simultaneous use of an infrared thermal imaging sensor to monitor the position... Changes in the thermal characteristics of the circuit at the location The steps for detecting changes in circuit power consumption include:

[0270] Before the modulated light signal is emitted, the target position is determined using an infrared thermal imaging sensor. High-precision temperature measurements are performed on the circuit and its surrounding area to obtain the reference value of the circuit's static operating temperature at that location. ;

[0271] Record the thermal radiation distribution pattern at this location without external light excitation, and establish a reference library of static thermal characteristics of the circuit.

[0272] During the transmission of the modulated signal, the infrared thermal imaging sensor continuously monitors the target position in a high-frequency sampling mode. Temperature changes;

[0273] According to the time nodes of the modulated optical signal, the first real-time temperature data sequence under each modulation frequency excitation is collected synchronously. Establish temperature-time variation curves;

[0274] The first real-time temperature data sequence obtained Compared with the circuit's static operating temperature reference value Perform difference calculations to extract the temperature change sequence at the target location;

[0275] Analyze the correspondence between temperature change and the timing of the modulated optical signal, identify the temperature fluctuation characteristics that are synchronized with the frequency change of the optical signal, and eliminate irrelevant factors (such as ambient temperature drift).

[0276] Based on the typical power consumption variation characteristics of the camera image sensor chip and automatic gain control circuit under optical signal excitation, the temperature threshold range for circuit operating state changes is set.

[0277] Detect target location Pay close attention to whether there is a localized temperature rise that corresponds to the increased power consumption of the camera circuit, and especially the temperature gradient changes in the heat-generating areas of the chip.

[0278] The response time and attenuation characteristics of the temperature fluctuations were analyzed and compared with the thermal time constant of a typical camera circuit.

[0279] It identifies temperature change patterns with unique thermal response speeds of camera circuits, eliminating temperature change interference caused by the physical effects of material thermal expansion and contraction.

[0280] Based on the aforementioned analysis results, the target location is quantified. The intensity of changes in the thermal characteristics of the circuit ;

[0281] Taking into account multiple dimensions such as temperature change amplitude, response time characteristics, and synchronization with the excitation signal, thermal characteristic parameters reflecting the degree of circuit activity at this location are generated.

[0282] The thermal characteristic parameter is used as an important input data for the final verification confidence calculation and is fused with the photoelectric reflection response data for judgment.

[0283] This embodiment utilizes infrared thermal imaging technology to monitor thermal characteristics caused by changes in circuit power consumption, enabling the determination of whether a hidden camera is operational without physical contact. The thermal characteristics generated by the circuit's operation are unique physical identifiers of the camera device. Compared to simple optical detection, adding a thermal detection dimension significantly improves the accuracy and reliability of identification. Even if the camera is physically obscured or disguised, the heat generated during circuit operation will still be manifested through thermal conduction and radiation. Thermal imaging can penetrate some camouflage materials to detect hidden devices. By monitoring changes in the circuit's thermal response under modulated light excitation, it is possible to verify whether a suspected camera possesses typical circuit functions such as automatic gain control, eliminating interference from non-functional optical components. Quantifying changes in circuit thermal characteristics provides an objective numerical reference for the final comprehensive judgment, avoiding the uncertainty of subjective judgment. The rapid response of the thermal effect caused by changes in circuit power consumption allows for quick verification of the circuit activity status at suspected locations, improving the overall response speed of the detection system. Thermal feature detection and photoelectric reflection detection complement each other; multi-dimensional feature fusion can significantly reduce false alarm and false negative rates, improving the overall performance of the detection system.

[0284] Please see Figure 2 Another embodiment of the present invention provides a system for detecting hidden cameras using infrared and photoelectric combined methods, comprising: an infrared thermal imaging sensor, a photoelectric sensor, and a server;

[0285] The server is configured as follows:

[0286] An infrared thermal imaging sensor is used to scan the area to be detected and acquire a baseline image of the ambient temperature distribution. ;

[0287] Ambient light intensity distribution baseline data is collected using a photoelectric sensor array. Establish a reference library of environmental photoelectric characteristics;

[0288] Control the multi-LED supplemental lighting system according to a preset time sequence Emit excitation light of different wavelengths in sequence. The excitation light includes visible light, near-infrared light, and mid-infrared light bands.

[0289] The intensity of the supplementary light at each time point is adaptively adjusted according to the environmental photoelectric feature reference library to ensure that the excitation light can effectively penetrate the preset type of camouflage material;

[0290] At each supplemental lighting time point, the infrared thermal imaging sensor acquires temperature distribution images in real time. ;

[0291] Calculate the temperature change rate matrix :

[0292] ;

[0293] Based on the heat conduction theory model, regions with abnormal response time constants were identified;

[0294] The photoelectric sensor array synchronously acquires the reflected light intensity matrix at each supplementary lighting time point. ;

[0295] Calculate the matrix of the rate of change of reflected light intensity :

[0296] ;

[0297] Extract the unique Fresnel reflection coefficient characteristics of the lens surface to identify abnormal areas with reflectivity in the range of 0.04-0.08;

[0298] temperature change rate matrix and the matrix of the rate of change of reflected light intensity Spatial location matching is performed to establish feature correspondences for the same physical coordinate points, and a fused feature matrix is ​​generated. ;

[0299] For each coordinate point Extract the thermal response data sequence within a neighborhood window centered at the point with a radius of δ. and photoelectric reflection data sequence Calculate the spatial correlation coefficient This is used to assess the similarity between the thermal response and photoelectric reflection distribution patterns in the neighborhood of that location;

[0300] Construct a feature fusion decision function , where α, β, and γ are weighting coefficients;

[0301] when >First preset thresholdTh and When the second preset threshold Tc is reached, it is determined that there is a hidden camera at that location, and the set of candidate location coordinates is recorded. .

[0302] It should be known that, Figure 2 The block diagram of the infrared and photoelectric combined detection system for hidden cameras shown is for illustrative purposes only, and the number of modules shown does not limit the scope of protection of this invention. The infrared and photoelectric combined detection system for hidden cameras provided in this embodiment can be used to execute various embodiments of the corresponding infrared and photoelectric combined detection methods for hidden cameras. For specific implementation details, please refer to the descriptions of the respective method embodiments, which will not be repeated here.

[0303] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0304] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0305] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical or other forms.

[0306] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0307] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0308] If the integrated units described above are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0309] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0310] The embodiments of this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

[0311] While the present invention has been disclosed above, it is not limited thereto. Any person skilled in the art can easily conceive of variations or substitutions without departing from the spirit and scope of the present invention, and various modifications and alterations can be made, including combinations of the different functions and implementation steps described above, as well as software and hardware implementation methods, all of which are within the protection scope of the present invention.

Claims

1. A method for detecting hidden cameras using a combination of infrared and photoelectric technologies, characterized in that, include: An infrared thermal imaging sensor is used to scan the area to be detected and obtain a baseline image of the ambient temperature distribution. A baseline database of environmental photoelectric characteristics is established by collecting ambient light intensity distribution benchmark data through a photoelectric sensor array. The multi-LED supplementary lighting system is controlled to emit excitation light of different wavelengths sequentially according to a preset time sequence; The intensity of the supplementary light at each time point is adaptively adjusted according to the environmental photoelectric feature reference library to ensure that the excitation light can effectively penetrate the preset type of camouflage material; At each supplemental lighting time point, the infrared thermal imaging sensor acquires temperature distribution images in real time; Calculate the temperature change rate matrix; Based on the heat conduction theory model, regions with abnormal response time constants were identified; The photoelectric sensor array synchronously collects the reflected light intensity matrix at each supplementary lighting time node; Calculate the matrix of the rate of change of reflected light intensity; Extract the unique Fresnel reflection coefficient characteristics of the lens surface to identify abnormal areas with reflectivity in the range of 0.04-0.08; Spatial location matching is performed between the temperature change rate matrix and the reflected light intensity change rate matrix to establish the feature correspondence of the same physical coordinate point and generate a fused feature matrix. For each coordinate point Extract thermal response data sequences and photoelectric reflection data sequences within a neighborhood window centered at the point with a radius of δ, and calculate the spatial correlation coefficient. This is used to assess the similarity between the thermal response and photoelectric reflection distribution patterns in the neighborhood of that point. Construct a feature fusion decision function ; when >First preset thresholdTh and When the second preset threshold Tc is reached, it is determined that there is a hidden camera at that location, and the set of candidate location coordinates P is recorded.

2. The method for detecting hidden cameras using a combination of infrared and photoelectric methods according to claim 1, characterized in that, Following the step of determining that there is a suspected hidden camera at the location, the following is also included: For each suspected camera location in the candidate location coordinate set P The system controls the multi-LED supplementary lighting system to emit a preset modulated light signal of a specific frequency at that position. Using photoelectric sensors at corresponding locations Detect the presence of feedback response caused by the camera's automatic gain control circuit and record the response strength. ; Simultaneously, use infrared thermal imaging sensors to monitor the location. Changes in the thermal characteristics of the circuit at the location Changes in power consumption of the detection circuit; Calculate the verification confidence level ,when When verifying the threshold Tv, the existence of a hidden camera at that location is finally confirmed, and the coordinates of the hidden camera and the overall confidence level are output; among them, , 2 represents the weighting coefficient. The importance of controlling AGC response characteristics The contribution of controlling changes in thermal characteristics.

3. The method for detecting hidden cameras using infrared and photoelectric combined methods according to claim 2, characterized in that, The feature fusion decision function uses an adaptive weight optimization algorithm to dynamically adjust the weight coefficients, specifically including: Extract the feature vector of the current detection environment ,in, For the average ambient temperature, For ambient light intensity, For relative humidity, The main material type of the detection area is coded; Calculate the texture complexity index of the detection region ,in The grayscale value of the visible light image. For the Laplacian operator, N×M is the image size, where N represents the number of pixels in the vertical direction and M represents the number of pixels in the horizontal direction. Establish a historical testing database ,in For historical environmental feature vectors, For the corresponding optimal weight combination, To improve detection accuracy; Calculate the Euclidean distance between the current environmental feature E and the historical records. Select the K historical records with the smallest distance as the reference set. ,in, This represents the i-th component of the current environment feature vector; This represents the i-th component of the historical environment feature vector; Construct a three-layer feedforward neural network model: The input layer receives the enhanced environmental feature vector. ,in Minimum historical distance; The hidden layer uses the ReLU activation function, and the output layer uses the Softmax normalization function to generate the predicted weights. ; The formula for weighted prediction is: ,in , This is the weight matrix. , It is the bias vector; Calculate the reference set Medium-weighted average Among them, weight ;in, For the j-th historical record, the optimal weight combination is... σ represents the distance between the current environment and the j-th historical environment, and σ is the temperature parameter. Establish an adaptive fusion function The fusion coefficient λx is determined through confidence level assessment: Where D1, D2, and D3 are the learning parameters; The feature fusion decision function is updated using the final weight coefficients: ; After completing the test and obtaining the verification results, calculate the accuracy index of this test. ; if Preset threshold Then the current environmental feature E and the final weight will be... and accuracy Add it as a new record to the historical database H; Simultaneously, the neural network parameters are updated using stochastic gradient descent. Where η is the learning rate and L is the loss function. The loss function L represents the parameters. The gradient.

4. The method for detecting hidden cameras using infrared and photoelectric combined methods according to claim 3, characterized in that, The step of adaptively adjusting the supplementary light intensity at each time point based on an environmental photoelectric feature reference library to ensure that the excitation light can effectively penetrate a preset type of camouflage material includes: Based on environmental photoelectric feature benchmark library Analyze the statistical characteristics of light intensity distribution and calculate the light intensity variance distribution map. ,in It is a local mean; Extracting the spectral reflectance feature vector of the material ,in , , These are the reflectance values ​​for the visible, near-infrared, and mid-infrared bands, respectively. The optical attenuation coefficient; Establish a camouflage material classifier Identify camouflage material categories of a preset type by matching spectral features. ∈{fabric, plastic, metal, coating, composite material}; For each of the n1 wavelengths and identified material categories Query the material optical parameter database to obtain the transmission coefficient and absorption coefficient μ; Calculate the required penetration power density ,in The minimum power density required for camera excitation, The estimated thickness of the camouflage material is preset; Establish safe power constraints ,in The power limit for the corresponding wavelength is the safe power limit for the human eye. Constructing a multi-objective optimization function ,in This is the wavelength weighting coefficient. This is the power consumption penalty factor; The Lagrange multiplier method is used to solve the constrained optimization problem, and the optimal power allocation vector is obtained. ; Calculate the driving current of each LED array. ,in For LED efficiency, Thermal resistance; At each time node Real-time monitoring of changes in ambient light intensity Calculate the ambient light intensity disturbance coefficient ; Establish an adaptive supplementary light intensity adjustment function ,in This is the environmental disturbance compensation coefficient. For modulation depth, The modulation frequency; Pulse width modulation (PWM) technology is used to control LED arrays of various wavelengths, with PWM duty cycle... This enables precise control of light intensity; Calculate penetration effect evaluation index ,in The detected signal strength, The baseline signal strength when there is no camouflage; like <threshold Then the power enhancement mode will be activated: Where γx is the enhancement coefficient; Update the transmission coefficient and absorption coefficient in the material optical parameter database, establish a self-learning optimization mechanism, and improve the penetration effect of subsequent detection.

5. The method for detecting hidden cameras using infrared and photoelectric combined methods according to claim 4, characterized in that, The step of identifying regions with abnormal response time constants based on the heat conduction theory model includes: At each illumination time point, the infrared thermal imaging sensor continuously acquires temperature distribution images, establishing a time-temperature response sequence for each pixel. ; For the position of each pixel within the detection area Extract the complete thermal response curve from the start of supplemental lighting to the preset steady state, and record the changes during the temperature rise and temperature fall phases; Identify and mark key time points in the response curve: the start time of supplemental lighting. Peak temperature End of supplemental lighting and temperature recovery time ; The thermal response time constant of each pixel is calculated based on the first-order exponential response model. This parameter reflects the characteristic time required for a material to reach a steady-state temperature after receiving heat. Extracting thermal response amplitude parameters This represents the temperature rise from the reference temperature to the peak temperature, reflecting the material's heat absorption capacity. Calculate the thermal decay coefficient This describes the rate at which the temperature returns to the reference value after the supplemental lighting stops, reflecting the heat dissipation characteristics of the material. Establish the heat conduction feature vector for each pixel. ; Establish a benchmark library of thermal conductivity characteristics for common surface materials, including standard thermal response parameter ranges for material types such as wood, metal, plastic, fabric, glass, and ceramics; For each material type, a normal range for the thermal response time constant is defined. arrive And the normal range of the corresponding thermal response amplitude and attenuation coefficient; Based on the unique thermal characteristics of camera devices, thermal response feature templates are established for components such as camera chips, lens assemblies, and circuit boards; the typical time constant range for camera chips is 0.5 seconds to 2.0 seconds. The heat conduction feature vector of each pixel The similarity of the features is calculated by matching and comparing them with the material features in the benchmark library. Identify the thermal response time constant Pixels that are significantly deviated from the normal range of the surrounding materials; Specially marked Pixels that fall within the feature range of the camera device are clustered into candidate camera regions. Candidate regions are filtered by shape and size to exclude abnormal thermal response regions that clearly do not conform to the camera's geometric characteristics; For each candidate camera region, calculate the thermal feature consistency index to evaluate the uniformity of thermal response parameters of each pixel within the region; Analyze the contrast of thermal characteristics between the candidate region and the surrounding background material, and calculate the significance of the difference in thermal response; Establish a confidence scoring mechanism for thermal features, taking into account factors such as time constant matching degree, response amplitude rationality, and regional shape regularity; The system outputs the location coordinates, area, and thermal feature confidence score of the abnormal thermal response region, providing accurate spatial positioning information for subsequent photoelectric feature extraction.

6. The method for detecting hidden cameras using infrared and photoelectric combined methods according to claim 5, characterized in that, The step of extracting the Fresnel reflection coefficient characteristics unique to the lens surface and identifying abnormal areas with reflectivity in the range of 0.04-0.08 includes: Ambient light intensity distribution benchmark data collected by photoelectric sensor array Preprocessing is performed to remove environmental noise interference and establish environmental reflectivity benchmark values ​​for each coordinate point. ; Controlling each LED unit in a multi-LED supplemental lighting system according to a preset spatial angle sequence Excitation light is emitted toward the area to be inspected to ensure that the incident beam can illuminate the potential lens surface from different angles; The photoelectric sensor array synchronously acquires data at each incident angle. Corresponding reflected light intensity distribution data This forms an angle-reflected light intensity data matrix; For each coordinate point Calculate its reflectance sequence at different incident angles. ,in The incident light intensity; Analyze the degree of agreement between the reflectivity sequence and Fresnel's law of reflection, and extract the Fresnel reflection coefficient that reflects the optical properties of the lens surface; The calculated Fresnel reflection coefficients at each coordinate point are compared and matched with the preset lens material reflectivity range of 0.04-0.08; Identify the coordinates of points whose Fresnel reflection coefficient falls within this range and have typical lens surface reflection characteristics, and mark these coordinates as optical anomalous areas; To eliminate interference signals caused by reflective objects such as metal surfaces and mirror materials, ensure that the identified abnormal areas have the optical reflection characteristics of the camera lens.

7. The method for detecting hidden cameras using infrared and photoelectric combined methods according to claim 6, characterized in that, The temperature change rate matrix and the matrix of the rate of change of reflected light intensity Spatial location matching is performed to establish feature correspondences for the same physical coordinate points, and a fused feature matrix is ​​generated. The steps include: The spatial coordinate system of the infrared thermal imaging sensor and the photoelectric sensor array is calibrated to determine the relative positional relationship and field of view of the two sensor systems. Establish a unified physical coordinate system and map the pixel coordinate system of the infrared thermal imaging sensor and the detection unit coordinate system of the photoelectric sensor array to the same spatial reference coordinate system; Analysis of the temperature change rate matrix and the matrix of the rate of change of reflected light intensity Spatial resolution differences; Interpolation resampling technology is used to unify the spatial resolution of the two matrices to the same grid density, ensuring that each physical coordinate point has a corresponding data value in both matrices; Based on the time series of the multi-LED supplementary lighting system, the temperature change rate data and the reflected light intensity change rate data are timestamped and corrected. Compensate for the response time difference between the infrared thermal imaging sensor and the photoelectric sensor to ensure that the thermal response data and photoelectric reflection data generated at the same supplementary lighting moment can correspond accurately; Based on the established coordinate system mapping relationship, the temperature change rate matrix is... and the matrix of the rate of change of reflected light intensity Perform geometric distortion correction; Eliminate spatial position deviations caused by sensor installation angle and lens distortion to achieve precise alignment of the same physical positions in the two matrices; Traversing the corrected temperature change rate matrix For each coordinate point in the matrix, the corresponding rate of change of reflected light intensity is... Find data values ​​with the same coordinates in the middle; Establish a one-to-one correspondence between the thermal response characteristic value and the photoelectric reflection characteristic value of each physical coordinate point to form a feature data pair at the coordinate point level; The established feature data pairs are reorganized according to their spatial coordinates to generate a fused feature matrix containing both thermal response and photoelectric reflection information. ; In the fusion feature matrix In the middle, each coordinate point It includes thermal response characteristic components from the temperature change rate matrix and photoelectric reflection characteristic components from the reflected light intensity change rate matrix, providing a data foundation for subsequent correlation analysis and determination function calculation.

8. The method for detecting hidden cameras using infrared and photoelectric combined methods according to claim 7, characterized in that, The method for each suspected camera location in the candidate location coordinate set P The steps of controlling the multi-LED supplementary lighting system to emit a preset modulated light signal of a specific frequency at the designated location include: Based on the feature fusion judgment function value of each position in the candidate position coordinate set P Spatial correlation coefficient Candidate positions are sorted in descending order of detection confidence. Establish a verification queue and prioritize processing suspected camera locations with the highest confidence level to improve verification efficiency; Based on the current candidate location coordinates to be verified The spatial distribution of each LED unit in the multi-LED supplementary lighting system is determined, and the spatial distance and illumination angle of each LED unit to the target position are calculated. Select the group of LED units that are closest to the target location and have the best illumination angle as the main illumination source, and select adjacent auxiliary LED units for supplementary illumination. Based on the ambient light intensity reference value of the target location in the environmental optoelectronic feature reference library, the basic intensity of the modulated light signal is set to ensure that the modulated signal can effectively penetrate potential camouflage and obstruction. The specific frequency of the modulated light signal is set as the preset sensitive frequency band of the camera's automatic gain control circuit, including a combination sequence of low-frequency modulated signals and high-frequency pulse signals; By controlling the emission angle and beam spread of the main and auxiliary LED units corresponding to the selected primary illumination source, the beams of multiple LEDs are aligned with the target position. An area of ​​overlapping light spots is formed at this location; By adjusting the luminous intensity distribution of each LED unit, a uniform and stable modulated light illumination area is formed at the target position, avoiding detection errors caused by uneven light intensity distribution. According to the preset modulation signal timing, control the focus to the target position. The LED chip units are combined to emit modulated light signals of different frequencies in sequence; The transmission duration for each modulation frequency is set according to the typical response time of the camera's automatic gain control circuit to ensure sufficient signal excitation time; During the transmission of modulated optical signals, the target position is monitored in real time using an array of photoelectric sensors. The change in light intensity confirms that the modulated light signal has accurately reached the target area; Simultaneously, infrared thermal imaging sensors are used to monitor temperature changes at the target location to verify whether light energy is effectively applied to the suspected camera location, providing a reference benchmark for subsequent response signal detection.

9. The method for detecting hidden cameras using infrared and photoelectric combined methods according to claim 8, characterized in that, Simultaneously, the position is monitored using an infrared thermal imaging sensor. Changes in the thermal characteristics of the circuit at the location The steps for detecting changes in circuit power consumption include: Before the modulated light signal is emitted, the target position is determined using an infrared thermal imaging sensor. High-precision temperature measurements are performed on the circuit and its surrounding area to obtain the reference value of the circuit's static operating temperature at that location. ; Record the thermal radiation distribution pattern at this location without external light excitation, and establish a reference library of static thermal characteristics of the circuit. During the transmission of the modulated signal, the infrared thermal imaging sensor continuously monitors the target position in a high-frequency sampling mode. Temperature changes; According to the time nodes of the modulated optical signal, the first real-time temperature data sequence under each modulation frequency excitation is collected synchronously. Establish temperature-time variation curves; The first real-time temperature data sequence obtained Compared with the circuit's static operating temperature reference value Perform difference calculations to extract the temperature change sequence at the target location; Analyze the correspondence between temperature change and the timing of the modulated optical signal, identify temperature fluctuation characteristics that are synchronized with the frequency change of the optical signal, and eliminate irrelevant factors; Based on the typical power consumption variation characteristics of the camera image sensor chip and automatic gain control circuit under optical signal excitation, the temperature threshold range for circuit operating state changes is set. Detect target location Pay close attention to whether there is a localized temperature rise that corresponds to the increased power consumption of the camera circuit, and especially the temperature gradient changes in the heat-generating areas of the chip. The response time and attenuation characteristics of the temperature fluctuations were analyzed and compared with the thermal time constant of a typical camera circuit. It identifies temperature change patterns with unique thermal response speeds of camera circuits, eliminating temperature change interference caused by the physical effects of material thermal expansion and contraction. Quantify target location The intensity of changes in the thermal characteristics of the circuit ; Taking into account multiple dimensions such as temperature change amplitude, response time characteristics, and synchronization with the excitation signal, thermal characteristic parameters reflecting the degree of circuit activity at this location are generated. The thermal characteristic parameter is used as an important input data for the final verification confidence calculation and is fused with the photoelectric reflection response data for judgment.

10. A system for detecting hidden cameras using infrared and photoelectric combined methods, for performing the method for detecting hidden cameras using infrared and photoelectric combined methods as described in any one of claims 1 to 9, characterized in that, include: Infrared thermal imaging sensors, photoelectric sensors, and servers; The server is configured as follows: An infrared thermal imaging sensor is used to scan the area to be detected and obtain a baseline image of the ambient temperature distribution. A baseline database of environmental photoelectric characteristics is established by collecting ambient light intensity distribution benchmark data through a photoelectric sensor array. The multi-LED supplementary lighting system is controlled to emit excitation light of different wavelengths sequentially according to a preset time sequence; The intensity of the supplementary light at each time point is adaptively adjusted according to the environmental photoelectric feature reference library to ensure that the excitation light can effectively penetrate the preset type of camouflage material; At each supplemental lighting time point, the infrared thermal imaging sensor acquires temperature distribution images in real time; Calculate the temperature change rate matrix; Based on the heat conduction theory model, regions with abnormal response time constants were identified; The photoelectric sensor array synchronously collects the reflected light intensity matrix at each supplementary lighting time node; Calculate the matrix of the rate of change of reflected light intensity; Extract the unique Fresnel reflection coefficient characteristics of the lens surface to identify abnormal areas with reflectivity in the range of 0.04-0.08; Spatial location matching is performed between the temperature change rate matrix and the reflected light intensity change rate matrix to establish the feature correspondence of the same physical coordinate point and generate a fused feature matrix. For each coordinate point Extract thermal response data sequences and photoelectric reflection data sequences within a neighborhood window centered at the point with a radius of δ, and calculate the spatial correlation coefficient. This is used to assess the similarity between the thermal response and photoelectric reflection distribution patterns in the neighborhood of that point. Construct a feature fusion decision function ; when >First preset thresholdTh and When the second preset threshold Tc is reached, it is determined that there is a hidden camera at that location, and the set of candidate location coordinates P is recorded.

Images