Reference device for thermal imaging image data normalization and application method

Abstract

The invention discloses a reference device for thermal imaging image data normalization and an application method, and belongs to the technical field of thermal imaging. The method comprises the following steps: calculating the layout number and positions of reference devices to obtain space coordinates of each reference device; arranging the reference devices according to the space coordinates of the reference devices; shooting a thermal imaging image of the reference device; and according to the thermal imaging image, image data normalization is carried out. The invention is based on a reference device for thermal imaging image normalization, and the thermal conductivity of a reference plate material can basically cover the thermal conductivity of different objects in a measurement range. Weighted average is carried out according to the heat conductivity of the four materials, a material weighted temperature deviation value is obtained through calculation, the object temperature at a certain time is further corrected to a temperature value at an initial time, and the space-time consistency of the object temperature measurement value during thermal imaging work is guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of thermal imaging technology, and in particular to a reference device and an application method for normalizing thermal imaging image data. Background Art

[0002] Thermal imaging technology uses infrared radiation to detect surface temperature distribution. A thermal imager converts temperature differences into visual images, providing a direct visual indication of the surface temperature. Thermal imaging technology is widely used in industrial inspection, security monitoring, cultural relic protection, medical diagnosis, and other fields.

[0003] In the field of cultural relics protection, thermal imaging technology is based on the difference in thermal properties between relics and the surrounding environment. It obtains thermal imaging images non-contactly through drone mounting, handheld shooting, etc., and identifies the characteristics and diseases of relics. It has the characteristics of strong flexibility and non-destructiveness to relics.

[0004] In the actual application of current thermal imaging technology, the ambient temperature fluctuates dynamically over time. The objects in the measured area change with the ambient temperature and have different thermal conductivities based on their own different properties. They appear as different temperature values ​​at different ambient temperatures, which are significantly different from the initial temperature state. The problem of temperature and space-time consistency cannot be maintained, which affects the accuracy of recognition.

[0005] Existing technologies focus on relative temperature calibration of the image side (the image capture end) at the current ambient temperature. Calibration methods primarily rely on traditional black bodies and motorized shutters. Traditional black bodies require a long time to equilibrate with the ambient temperature, have slow thermal conductivity, and are difficult to adapt to scenarios with rapid temperature fluctuations. Motorized shutters and standard black bodies can only correct for the device's own dark current and gain drift; they cannot eliminate the impact of temperature changes on thermal radiation from the object side (the object being measured), resulting in data that cannot be aligned with the initial reference temperature.

[0006] Existing technologies can only establish a relative temperature benchmark for the image space in the current environment, and are unable to uniformly calibrate the temperature values ​​of the object space over a certain period of time, multiple measurements, or historical data to the initial reference temperature, resulting in poor comparability of time series data. Summary of the Invention

[0007] The purpose of the present invention is to provide a reference device and application method for normalizing thermal imaging image data, which normalizes all image data that change in time and space and unifies them to the temperature reference of the initial image, thereby achieving temporally and spatially consistent comparison of the temperatures reflected by each image data, laying the foundation for the correct interpretation of thermal imaging images.

[0008] To solve the above technical problems, the present invention provides a reference device for normalizing thermal imaging image data, which is characterized by comprising:

[0009] Reference device base plate;

[0010] A thermal imaging image normalization reference module includes four reference plates mounted on a base plate of a reference device; the four reference plates are made of four materials with different thermal conductivities;

[0011] A vial, comprising a vertical vial and a horizontal vial; both the vertical vial and the horizontal vial are mounted on a base plate of the reference device;

[0012] A temperature recorder is mounted on the reference device base plate.

[0013] Preferably, the four reference boards are a first material thermal imaging temperature reference module, a second material thermal imaging temperature reference module, a third material thermal imaging temperature reference module and a fourth material thermal imaging temperature reference module.

[0014] Preferably, the first material thermal imaging temperature reference module, the second material thermal imaging temperature reference module, the third material thermal imaging temperature reference module and the fourth material thermal imaging temperature reference module are aluminum alloy plate, stainless steel plate, ceramic tile plate and PVC plate respectively.

[0015] Preferably, a cross-shaped thermal insulation strip is installed between the first material thermal imaging temperature reference module, the second material thermal imaging temperature reference module, the third material thermal imaging temperature reference module and the fourth material thermal imaging temperature reference module.

[0016] Preferably, the base plate of the reference device is provided with threaded holes around its periphery;

[0017] A screw rod is installed in the threaded hole.

[0018] The present invention also provides an application method of a reference device for normalizing thermal imaging image data, comprising the following steps:

[0019] Calculate the number and position of reference devices and obtain the spatial coordinates of each reference device;

[0020] Arrange the reference devices according to the spatial coordinates of each reference device;

[0021] Take thermal images of the reference device;

[0022] Based on thermal imaging images, image data is normalized.

[0023] Preferably, calculating the number and location of reference devices comprises the following steps:

[0024] (1) Calculating the size of a reference device according to work requirements to obtain the reference device dimensions; the work requirements include pixel size, shooting height, and focal length; the reference device dimensions include the minimum side length of a normalized reference module for thermal imaging images;

[0025] D i =5·GSD

[0026]

[0027] Where: D i is the minimum side length of the normalized reference module of the thermal imaging image; GSD is the spatial resolution, which refers to the spatial distance represented by one pixel; S is the pixel size; H is the shooting height; f is the focal length;

[0028] Select the reference device based on its size;

[0029] (2) Calculate the ambient temperature change rate α based on the temperature change forecast for the time period in the weather forecast;

[0030] (3) Calculate the number of thermal imaging images to be taken in the shooting area based on the shooting area size, shooting equipment parameters, and shooting distance:

[0031] Assume the project area is x, then the total number of images taken in the project area is

[0032] (4) Calculate the shooting time based on the image shooting rate:

[0033] Assume that the shooting speed is ν, and the total shooting time of the project area is

[0034] Where: f is the focal length; x is the project area; H is the shooting height; L is the length of the long side of the sensor; W is the length of the short side of the sensor; Rx is the heading overlap rate; Ry is the heading overlap rate;

[0035] (5) Set the ambient temperature change value between adjacent reference devices and calculate the number of reference devices to be installed based on the ambient temperature change rate;

[0036] (6) Place the reference device in the first image, use the center coordinates of the first image as the placement coordinates of the first reference device, and record the reference device sequence number as 1. A second reference device is placed in the image. Indicates rounding to the next larger integer, and obtains the The actual space coordinates corresponding to the center of the image are used as the layout coordinates of the second reference device, and the reference device serial number is 2. The actual space coordinates corresponding to the center of the image are used as the layout coordinates of the nth reference device, and the serial number of the reference device is recorded as n.

[0037] Preferably, setting the ambient temperature variation value between adjacent reference devices and calculating the number of reference devices to be installed according to the ambient temperature variation rate specifically includes the following steps:

[0038] The reference device layout distance first determines the temperature accuracy requirement, and sets the maximum allowable ambient temperature change between adjacent reference devices as ΔTallow; the ambient temperature change rate as α; the shooting speed as ν; and the time interval between single photos as Δt.

[0039]

[0040] Consider the limit interval N of temperature change, where N is an integer:

[0041]

[0042] Adjacent flight strips cannot share temperature reference points;

[0043] Heading reference device layout spacing D x for:

[0044]

[0045] The lateral layout spacing is determined by the short side of the sensor and the lateral overlap rate, and has nothing to do with the shooting speed. The lateral reference device layout spacing D y :

[0046]

[0047] The reference device layout density ρ per unit area is:

[0048]

[0049] Where: D x Place the distance for the heading reference device; D y is the distance from the lateral reference device; H is the shooting height; L is the length of the long side of the sensor; f is the focal length; W is the length of the short side of the sensor; T is the maximum allowable temperature deviation; α is the ambient temperature change rate; Δt is the time interval between single photos; ν is the shooting speed; R x is the heading overlap ratio; R y is the heading overlap ratio.

[0050] Preferably, image data normalization is performed based on the thermal imaging image, specifically comprising the following steps:

[0051] (1) Obtain the ambient temperature T env,t ;

[0052] The temperature data T with time attributes is obtained through the temperature recorder of each reference device env,t , the acquisition frequency is the same as the shooting frequency, the temperature record data is exported, and the time-ambient temperature curve is fitted: T env,t =f(t);

[0053] (2) Obtain the reference point temperature Tref,i,t (i=1,2,3,4);

[0054] Extract the thermal imaging image containing the reference device and obtain the temperature data T of different materials of each reference device ref,i,t (i=1,2,3,4);

[0055] (3) Construct a two-dimensional array of reference points according to the time sequence; construct a two-dimensional array of images according to the time sequence;

[0056] (4) aligning and merging the reference point two-dimensional array and the image two-dimensional array according to the timestamp to obtain a merged data table;

[0057] (5) Read the merged data table line by line, load the original thermal imaging image, convert its color value into temperature value, apply temperature compensation for alignment time, obtain normalized temperature value, convert the normalized temperature value into pixel color value, and re-output the thermal imaging image;

[0058] For images without a reference device, by establishing t-ΔT weight,t Function, calculates the temperature offset value between known time points; performs temperature offset extrapolation calculation for time points outside the known time range; and performs temperature calibration on the original image at any time point to eliminate the impact of temperature offset.

[0059] Preferably, constructing a two-dimensional array of reference points in chronological order; constructing a two-dimensional array of images in chronological order specifically includes the following steps:

[0060] Step (3-1): Input the reference plate temperature time series [t, T ref,1,t , T ref,2,t , T ref,3,t , T ref,4,t ];

[0061] Step (3-2): Define the thermal conductivity matrix of the four materials as [k1, k2, k3, k4] and get the weight coefficients

[0062] Step (3-3): Extract the first frame of the image with the reference device, obtain the initial temperature value of the reference device, and determine the reference temperature T ref,i,0 ;

[0063] Step (3-4): Calculate the reference board temperature offset ΔT ref,i,t =T ref,i,t -T ref,i,0 , we get [t, ΔT ref,1,t , ΔT ref,2,t , ΔT ref,3,t , ΔT ref,4,t ];

[0064] Steps (3-5): Weighted Temperature Offset Construct a two-dimensional array of reference points;

[0065] Step (3-6): Construct a two-dimensional array of images.

[0066] Compared with the prior art, the present invention has the following beneficial effects:

[0067] 1. Use reference plates made of various materials to form a standardized reference module for thermal imaging images, integrate a vial and a temperature recorder to further form an overall device, and further calculate the formula for the dynamic layout distance and density of the reference device.

[0068] 2. Based on a reference device used for normalizing thermal imaging images, the thermal conductivity of the reference plate material essentially covers the thermal conductivity of different objects within the measurement range. A weighted average of the thermal conductivities of the four materials is calculated to determine the material-weighted temperature offset. This is then used to correct the object temperature at a specific time to the initial value, ensuring temporal and spatial consistency of the object temperature measurements during thermal imaging. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] The specific embodiments of the present invention are further described in detail below with reference to the accompanying drawings.

[0070] Figure 1 The plan view and elevation view of a reference device for normalizing thermal imaging images according to the present invention;

[0071] Figure 2 A flowchart of an application method for normalizing thermal imaging images according to the present invention;

[0072] Figure 3 To construct the reference point "t, ΔT weight,t ” and “t, image name” two-dimensional array flow chart. DETAILED DESCRIPTION

[0073] The following description sets forth numerous specific details to facilitate a thorough understanding of the present invention. However, the present invention can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without departing from the scope of the present invention. Therefore, the present invention is not limited to the specific implementations disclosed below.

[0074] The terms used in one or more embodiments of this specification are for the purpose of describing specific embodiments only and are not intended to limit one or more embodiments of this specification. The singular forms "a," "the," and "the" used in one or more embodiments of this specification and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items.

[0075] It should be understood that although the terms first, second, etc. may be used to describe various information in one or more embodiments of this specification, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of this specification, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0076] The present invention is described in further detail below with reference to the accompanying drawings:

[0077] In order to better illustrate the technical effects of the present invention, the present invention provides the following specific examples to illustrate the above technical process:

[0078] Example 1: A reference device and application method for normalizing thermal imaging image data, which normalizes all image data that change in time and space to the temperature reference of the initial image, thereby achieving temporally and spatially consistent comparison of the temperature reflected by each image data, laying the foundation for the correct interpretation of thermal imaging images.

[0079] Thermal imaging data reflects the temperature of the subject. If thermal imaging takes time to capture, the ambient temperature of the subject will change during this time, resulting in poor comparability of the temperatures reflected in each image. Before interpreting thermal imaging data, it is necessary to normalize the image data to a unified benchmark, ensuring consistent temporal and spatial comparability of the temperatures reflected in each image.

[0080] The present invention utilizes four materials with varying thermal conductivities on the same reference plate to form a reference module for normalizing thermal imaging image data. This module, combined with a two-way vial and a temperature recorder (6), forms a reference device. The module calculates the number of thermal images captured in the capture area, the duration of the capture, and the number of reference devices deployed and their positions in the proposed image sequence. The reference devices are deployed on-site, and thermal images are captured. Starting with the first reference device, the difference between the weighted average thermal conductivity temperatures of the four materials and the ambient temperature of adjacent reference devices is calculated to obtain a weighted material temperature offset within the capture time period. This offset then corrects the temperature of the images in the adjacent reference device image sequence to the same baseline temperature as the first image, achieving temporal and spatial consistency in the captured object's temperature within the thermal imaging capture time period.

[0081] 1. A reference device for normalizing thermal imaging image data, such as Figure 1 As shown, specifically:

[0082] The reference device consists of a base plate 1, a threaded hole 2, a screw 3, a vertical vial 4, a horizontal vial 5; a temperature recorder 6; a first material thermal imaging temperature reference module 7, a second material thermal imaging temperature reference module 8, a third material thermal imaging temperature reference module 9, a fourth material thermal imaging temperature reference module 10, a cross-shaped insulation strip 11, a warning label 12, and a reference device number 13 (ID).

[0083] The reference device base plate 1 is used as a carrier for all modules, instruments, and logos. It is made of engineering plastic material with low thermal conductivity, light weight, and good outdoor working ability.

[0084] The threaded holes 2 are used to install screws 3 when the device is arranged horizontally. By adjusting the height of the screws at the four corners of the base plate, the reference device can be used in a horizontal state. When the device is arranged vertically, it can serve as a hanging hole;

[0085] The vial includes a vertical vial 4 and a horizontal vial 5. The vertical vial 4 and the horizontal vial 5 are used to calibrate the placement angle of the device. When arranged horizontally, adjust the screw until both vials are within the designated center range, indicating that the device is now horizontally placed. When arranged vertically, gravity makes the vertical position vertical, and it is only necessary to adjust the two sides of the device up and down so that the horizontal vial is within the designated center range.

[0086] The temperature recorder 6 obtains the time series change value of the ambient temperature by recording the ambient temperature data at a set frequency;

[0087] The first material thermal imaging temperature reference module 7, the second material thermal imaging temperature reference module 8, the third material thermal imaging temperature reference module 9, and the fourth material thermal imaging temperature reference module 10 are square plates. The four materials with the highest thermal conductivity are aluminum alloy plate, stainless steel plate, ceramic tile plate, and PVC plate. Based on the material temperature identified by thermal imaging and the ambient temperature at the same time, a time stamp is used to construct an object time-temperature offset relationship model to assist in reverse compensation to the object temperature at the initial temperature T0 when the shooting is started.

[0088] Cross-shaped thermal insulation strip 11, in I-shape, can be made of PA66 nylon, which blocks the energy exchange between the four reference boards and avoids thermal interference between the reference boards when the temperature changes;

[0089] The warning label 12 is used to warn people around to prevent the device from being disturbed during operation;

[0090] The reference device number 13 is the ID of the reference device and is unique.

[0091] The size of the reference device is designed according to the requirements of thermal imaging work and has a variety of different sizes. The size meets the following basic conditions:

[0092] ① Thermal radiation signals are identifiable: Ensure that the thermal imager can clearly distinguish the temperature differences of different reference materials (aluminum alloy, stainless steel, ceramic tiles, PVC, etc.).

[0093] ② Aerial survey scale adaptation: The size should match the flight altitude and thermal imager resolution to avoid the reference module occupying an insufficient proportion in the image due to its small size.

[0094] ③ Minimize environmental interference: Reduce edge heat diffusion effects to ensure the accuracy of temperature measurement in the center area.

[0095] The core size of the device is the size of the normalized thermal imaging reference module, which is determined by the pixel size, lens focal length, sensor size, and flight altitude. Furthermore, to ensure that the thermal radiation signal is recognizable, the module must cover at least 5×5 pixels in the image to eliminate the influence of noise. Based on this, the minimum side length D of the normalized thermal imaging reference module is determined. i .

[0096] D i =5·GSD

[0097]

[0098] in:

[0099] D i ——The minimum side length of the normalized reference module of the thermal imaging image, in meters;

[0100] GSD - spatial resolution, refers to the spatial distance represented by a pixel, unit is m;

[0101] S——pixel size, unit: mm;

[0102] H——shooting height, unit: m;

[0103] f——focal length, unit: mm.

[0104] 2. Application Method of Reference Device for Normalization of Thermal Imaging Image Data

[0105] (1) Principles of application methods

[0106] The reference plate materials are aluminum alloy, stainless steel, ceramic tile, and PVC, with varying thermal conductivities. These materials generally cover the thermal conductivities of various objects within the measurement range. A weighted average of the thermal conductivities of the four materials is calculated to determine the material-weighted temperature offset. Assuming the thermal conductivity of the object is similar to that of the reference plate, the reference plate weighted temperature offset is identical to the temporal temperature offset of objects in the entire scene. The object temperature at a specific time is then corrected to the temperature at the initial time, ensuring spatiotemporal consistency of the object temperature measurements during thermal imaging.

[0107] The thermal conductivity of the four reference plates are k i , k1, k2, k3, k4, initial ambient temperature T env,0 , at this time the apparent temperature of the four reference plates is T ref,i,0 (i=1,2,3,4), the apparent temperature of the object is T obj,0 .

[0108] At time t, the ambient temperature T env,t , at this time the apparent temperature of the four reference plates is T ref,i,t (i=1,2,3,4), the apparent temperature of the object is T obj,t .

[0109] After t time, the reference board temperature offsets by ΔT ref,i,t =T ref,i,t -T ref,i,0 .

[0110] Since different materials have different thermal conductivity, the temperature offset values ​​that occur at the same time are different, and it is necessary to fit the weighted temperature offset ΔT weight,t

[0111]

[0112] Assuming that the temporal temperature offset of the reference plate and the objects in the entire scene is the same, the object correction temperature T obj,corr for:

[0113] Tobj,corr =T obj,t -ΔT weight,t

[0114] However, since the reference device does not exist in every frame of the image, in order to correct the temperature of all images to the initial state, it is necessary to adjust the temperature of the image according to the interval ΔT weight,t The relationship curve between the temperature offset value and time is fitted, and the temperature offset value of each frame image is obtained according to the shooting time of each frame.

[0115] (2) Steps of application method

[0116] The method implementation includes the following steps:

[0117] 1. Calculation of the number and location of reference devices

[0118] (1) Calculate the size of the reference device based on work requirements.

[0119] D i =5·GSD

[0120]

[0121] in:

[0122] D i ——The minimum side length of the normalized reference module of the thermal imaging image, in meters;

[0123] GSD - spatial resolution, refers to the spatial distance represented by a pixel, unit is m;

[0124] S——pixel size, unit: mm;

[0125] H——shooting height, unit: m;

[0126] f——focal length, unit: mm.

[0127] After obtaining the calculated size of the reference device, select the reference device whose actual size is larger than the calculated size and whose actual size is closest to the calculated size among devices of different sizes to carry out subsequent work.

[0128] (2) Calculate the ambient temperature change rate α (°C / s) based on the temperature change forecast for the time period in the weather forecast;

[0129] (3) Calculate the number of thermal imaging images taken in the shooting area based on the size of the shooting area, shooting equipment parameters, shooting distance, etc.: Assume that the total area of ​​the project area is x km 2 , then the total number of images taken in the project area is

[0130] in:

[0131] f——focal length, unit: mm;

[0132] x——project area, unit: km 2 ;

[0133] H——shooting height, unit: m;

[0134] L——the length of the long side of the sensor, in mm;

[0135] W——the length of the short side of the sensor, in mm;

[0136] Rx——heading overlap rate, %;

[0137] Ry——heading overlap rate, %;

[0138] (4) Calculate the shooting time based on the image shooting rate: Assume the shooting speed ν (m / s), the total shooting time of the project area is

[0139] (5) Set the ambient temperature variation between adjacent reference devices and calculate the number of reference devices based on the ambient temperature variation rate. The reference device layout distance first needs to determine the temperature accuracy requirements, and set the maximum allowable ambient temperature variation between adjacent reference devices as ΔTallow (°C); the ambient temperature variation rate as α (°C / s); the shooting speed as ν (m / s); and the time interval between single photos as Δt (s).

[0140]

[0141] Consider the limit interval N of temperature change (N is an integer):

[0142]

[0143] Furthermore, unlike image space control points, adjacent flight strips cannot share temperature reference points due to the temporal variability of ambient temperature.

[0144] Furthermore, the heading reference device layout spacing D x :

[0145]

[0146] The lateral layout spacing is determined by the short side of the sensor and the lateral overlap rate, and has nothing to do with the shooting speed. The lateral reference device layout spacing D y :

[0147]

[0148] Furthermore, the reference device deployment density per unit area is ρ (units / km 2 )for:

[0149]

[0150] in:

[0151] D x ——The placement distance of the heading reference device, in meters;

[0152] D y ——lateral reference device placement distance, in m;

[0153] H——shooting height, unit: m;

[0154] L——the length of the long side of the sensor, in mm;

[0155] f——focal length, unit: mm;

[0156] W——the length of the short side of the sensor, in mm;

[0157] Tallowed - maximum allowable temperature deviation, unit: °C;

[0158] α——ambient temperature change rate, unit: °C / s;

[0159] Δt——time interval between single photos, unit: s;

[0160] ν——shooting speed, m / s;

[0161] R x ——heading overlap rate, %;

[0162] R y ——heading overlap rate, %;

[0163] According to the reference device layout density ρ, the project area and the route design, the number of reference devices is obtained;

[0164] (6) After obtaining the number of thermal imaging images of the shooting area, design the route and obtain the actual spatial coordinates corresponding to the center of each image. According to the project area and route design, the actual number of reference devices required is calculated through the reference device layout density formula to obtain Xkm 2 The project area requires k reference devices. The reference devices are laid out in the first image, and the center coordinates of the first image are used as the layout coordinates of the first reference device, and the sequence number of the reference device is recorded as 1. The second reference device is deployed in the image ( Indicates rounding to the largest integer), get the The actual space coordinates corresponding to the center of the image are used as the layout coordinates of the second reference device, and the reference device serial number is 2. The actual space coordinates corresponding to the center of the image are used as the layout coordinates of the nth reference device, and the serial number of the reference device is recorded as n.

[0165] 2. Layout of reference device

[0166] Based on the calculated spatial coordinates of each reference device, each reference device is placed within the site. When placed horizontally, the screws of each reference device are adjusted so that both bubbles are within the designated center. When shooting vertically, after hanging the device using the threaded holes, gravity makes the vertical axis vertical, and the horizontal bubbles of the device are adjusted up and down to keep them within the designated center.

[0167] 3. Thermal imaging

[0168] Thermal imaging images are taken using the flight strip method. Please note that if the battery needs to be replaced during operation, when the battery replacement is completed and work is resumed, work must be continued from the last reference point where the work was interrupted.

[0169] 4. Image data normalization

[0170] (1) Obtain the ambient temperature T env,t The temperature recorder 6 of each reference device automatically obtains the temperature data T with time attribute env,t , the acquisition frequency is the same as the shooting frequency, the temperature record data is exported, and the time-ambient temperature curve is fitted: T env,t =f(t).

[0171] (2) Obtain the reference point temperature T ref,i,t (i=1,2,3,4). Extract the thermal imaging image containing the reference device and obtain the temperature data T of different materials of each reference device ref,i,t (i=1,2,3,4).

[0172] (3) Construct reference points "t, ΔT" in chronological order weight,t " two-dimensional array; construct the image "t, image name" two-dimensional array in chronological order. Further, the above step (3) includes the following steps:

[0173] Step (3-1): Input the reference plate temperature time series [t, T ref,1,t , T ref,2,t , T ref,3,t , T ref,4,t ].

[0174] Step (3-2): Define the thermal conductivity matrix of the four materials as [k1, k2, k3, k4] and get the weight coefficients

[0175] Step (3-3): Extract the first frame of the image with the reference device, obtain the initial temperature value of the reference device, and determine the reference temperature T ref,i,0 .

[0176] Step (3-4): Calculate the reference board temperature offset ΔT ref,i,t =T ref,i,t -T ref,i,0 , we get [t, ΔT ref,1,t , ΔT ref,2,t , ΔT ref,3,t , ΔT ref,4,t ].

[0177] Steps (3-5): Weighted Temperature Offset Construct reference point "t, ΔT weight,t ” Two-dimensional array.

[0178] Step (3-6): Construct a two-dimensional array of image "t, image name".

[0179] (4) The two time series matrices (reference point two-dimensional array "t, ΔT weight,t ” and image 2D array “t, image name”) are aligned and merged by timestamp.

[0180] (5) Read the merged data table line by line, load the original thermal imaging data, convert the color value (RGB or Lab value) of the unprocessed image into a temperature value, apply the temperature compensation of the alignment time, obtain the normalized temperature value, and then convert the normalized temperature value into a pixel color value, and re-output the thermal imaging image. For images without a reference device, by establishing t-ΔT weight,t This function calculates the temperature offset between known time points. It also extrapolates the temperature offset for time points outside the known time range. It also performs temperature calibration on the original image at any time point to eliminate the effects of temperature offset.

[0181] In the several embodiments provided herein, it should be understood that the disclosed devices and methods may be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of modules, subassemblies, or units is merely a logical functional division. In actual implementation, other divisions may be used, such as combining or integrating multiple units, subassemblies, or components into another device, or some features may be omitted or not implemented.

[0182] The units may or may not be physically separate, and the components shown as units may be one physical unit or multiple physical units, that is, they may be located in one place or distributed in multiple places. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0183] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0184] In particular, according to an embodiment disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part, and / or installed from a removable medium. When the computer program is executed by the central processing unit (CPU), the above-mentioned functions defined in the method of the present invention are executed. It should be noted that the above-mentioned computer-readable medium of the present invention can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, a system, device or device of an electrical, magnetic, optical, electromagnetic, infrared segment, or semiconductor, or any combination of the above.

[0185] The flow charts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, program segment, or a part of code, and the module, program segment, or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0186] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions within the technical scope disclosed by the present invention shall be covered by the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be subject to the scope of protection of the claims.

Claims

1. A reference device for normalizing thermal imaging image data, characterized in that: include: Reference device base plate; a thermal imaging image normalization reference module comprising four reference plates mounted on a reference device base plate; The four reference plates are made of four materials with different thermal conductivities; A vial, comprising a vertical vial and a horizontal vial; both the vertical vial and the horizontal vial are mounted on a base plate of the reference device; A temperature recorder is mounted on the reference device base plate.

2. The reference device for normalizing thermal imaging image data according to claim 1, characterized in that: The four reference boards are a first material thermal imaging temperature reference module, a second material thermal imaging temperature reference module, a third material thermal imaging temperature reference module and a fourth material thermal imaging temperature reference module.

3. The reference device for normalizing thermal imaging image data according to claim 2, characterized in that: The first material thermal imaging temperature reference module, the second material thermal imaging temperature reference module, the third material thermal imaging temperature reference module and the fourth material thermal imaging temperature reference module are respectively an aluminum alloy plate, a stainless steel plate, a ceramic tile plate and a PVC plate.

4. The reference device for normalizing thermal imaging image data according to claim 3, characterized in that: A cross-shaped thermal insulation strip is installed between the first material thermal imaging temperature reference module, the second material thermal imaging temperature reference module, the third material thermal imaging temperature reference module and the fourth material thermal imaging temperature reference module.

5. The reference device for normalizing thermal imaging image data according to claim 4, characterized in that: The base plate of the reference device is provided with threaded holes around its periphery; A screw rod is installed in the threaded hole.

6. An application method of a reference device for normalizing thermal imaging image data, wherein the reference device is the reference device for normalizing thermal imaging image data according to any one of claims 1 to 5, characterized in that: The following steps are involved: Calculate the number and position of reference devices and obtain the spatial coordinates of each reference device; Arrange the reference devices according to the spatial coordinates of each reference device; Take thermal images of the reference device; Based on thermal imaging images, image data is normalized.

7. The method for applying the reference device for normalizing thermal imaging image data according to claim 6, characterized in that: Calculating the number and location of reference devices includes the following steps: (1) Calculating the size of a reference device according to work requirements to obtain the reference device dimensions; the work requirements include pixel size, shooting height, and focal length; the reference device dimensions include the minimum side length of a normalized reference module for thermal imaging images; D i =5·GSD Where: D i is the minimum side length of the normalized reference module of the thermal imaging image; GSD is the spatial resolution, which refers to the spatial distance represented by one pixel; S is the pixel size; H is the shooting height; f is the focal length; Select the reference device based on its size; (2) Calculate the ambient temperature change rate α based on the temperature change forecast for the time period in the weather forecast; (3) Calculate the number of thermal imaging images to be taken in the shooting area based on the shooting area size, shooting equipment parameters, and shooting distance: Assume the project area is x, then the total number of images taken in the project area is (4) Calculate the shooting time based on the image shooting rate: Assume that the shooting speed is ν, and the total shooting time of the project area is Where: f is the focal length; x is the project area; H is the shooting height; L is the length of the long side of the sensor; W is the length of the short side of the sensor; Rx is the heading overlap rate; Ry is the heading overlap rate; (5) Set the ambient temperature change value between adjacent reference devices and calculate the number of reference devices to be installed based on the ambient temperature change rate; (6) Place the reference device in the first image, use the center coordinates of the first image as the placement coordinates of the first reference device, and record the reference device sequence number as 1; then A second reference device is placed in the image. Indicates rounding to the next larger integer, and obtains the The actual space coordinates corresponding to the center of the image are used as the layout coordinates of the second reference device, and the reference device serial number is 2; and so on. The actual space coordinates corresponding to the center of the image are used as the layout coordinates of the nth reference device, and the serial number of the reference device is recorded as n.

8. The method for applying the reference device for normalizing thermal imaging image data according to claim 7, characterized in that: Setting the ambient temperature variation between adjacent reference devices and calculating the number of reference devices to be installed based on the ambient temperature variation rate specifically includes the following steps: The reference device layout distance first determines the temperature accuracy requirement, and sets the maximum allowable ambient temperature change between adjacent reference devices as ΔTallow; the ambient temperature change rate as α; the shooting speed as ν; and the time interval between single photos as Δt. Consider the limit interval N of temperature change, where N is an integer: Adjacent flight strips cannot share temperature reference points; Heading reference device layout spacing D x for: The lateral layout spacing is determined by the short side of the sensor and the lateral overlap rate, and has nothing to do with the shooting speed. The lateral reference device layout spacing D y : The reference device layout density ρ per unit area is: Where: D x Place the distance for the heading reference device; D y is the distance from the lateral reference device; H is the shooting height; L is the length of the long side of the sensor; f is the focal length; W is the length of the short side of the sensor; T is the maximum allowable temperature deviation; α is the ambient temperature change rate; Δt is the time interval between single photos; ν is the shooting speed; R x is the heading overlap ratio; R y is the heading overlap ratio.

9. The method for applying the reference device for normalizing thermal imaging image data according to claim 8, characterized in that: Based on the thermal imaging image, image data normalization is carried out, which specifically includes the following steps: (1) Obtain the ambient temperature T env,t ; The temperature data T with time attributes is obtained through the temperature recorder of each reference device env,t , the acquisition frequency is the same as the shooting frequency, the temperature record data is exported, and the time-ambient temperature curve is fitted: T env,t =f(t); (2) Obtain the reference point temperature T ref,i,t (i=1,2,3,4); Extract the thermal imaging image containing the reference device and obtain the temperature data T of different materials of each reference device ref,i,t (i=1,2,3,4); (3) Construct a two-dimensional array of reference points according to the time sequence; construct a two-dimensional array of images according to the time sequence; (4) aligning and merging the reference point two-dimensional array and the image two-dimensional array according to the timestamp to obtain a merged data table; (5) Read the merged data table line by line, load the original thermal imaging image, convert its color value into temperature value, apply temperature compensation for alignment time, obtain normalized temperature value, convert the normalized temperature value into pixel color value, and re-output the thermal imaging image; For images without a reference device, by establishing t-ΔT weight,t Function, calculates the temperature offset value between known time points; performs temperature offset extrapolation calculation for time points outside the known time range; and performs temperature calibration on the original image at any time point to eliminate the impact of temperature offset.

10. The method for applying the reference device for normalizing thermal imaging image data according to claim 9, characterized in that: Constructing a two-dimensional array of reference points in chronological order; constructing a two-dimensional array of images in chronological order, specifically including the following steps: Step (3-1): Input the reference plate temperature time series [t, T ref,1,t , T ref,2,t , T ref,3,t , T ref,4,t ]; Step (3-2): Define the thermal conductivity matrix of the four materials as [k1, k2, k3, k4] and get the weight coefficients Step (3-3): Extract the first frame of the image with the reference device, obtain the initial temperature value of the reference device, and determine the reference temperature T ref,i,0 ; Step (3-4): Calculate the reference board temperature offset ΔT ref,i,t =T ref,i,t -T ref,i,0 , we get [t, ΔT ref,1,t , ΔT ref,2,t , ΔT ref,3,t , ΔT ref,4,t ]; Steps (3-5): Weighted Temperature Offset Construct a two-dimensional array of reference points; Step (3-6): Construct a two-dimensional array of images.

Images