Method for analyzing a sequence of infrared thermographic images and related device
By applying mean filtering and density clustering to the infrared thermal image sample set, a minimum neighborhood circle function is constructed, which solves the subjectivity problem in existing infrared thermal image analysis and improves the accuracy and objectivity of the analysis.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING UNIV OF POSTS & TELECOMM
- Filing Date
- 2023-10-16
- Publication Date
- 2026-07-24
Smart Images
Figure CN117523243B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of infrared image analysis technology, and in particular to a method and related equipment for analyzing the sequence of infrared thermal images. Background Technology
[0002] Due to the characteristics of human physiological structure and the different uses of organs, the temperature distribution characteristics of different parts of the human body are different and follow certain patterns. Generally, the skin temperature is lowest at the extremities and higher closer to the center of the torso. These different distribution characteristics reflect the sequential nature of human body temperature distribution. In some cases, the temperature distribution of the human body can become disordered, thus disrupting this sequential nature. Current research on the sequential nature of human body temperature distribution mainly focuses on the sequential analysis of infrared thermal imaging. Most studies rely on visual observation of infrared thermal images, using subjective analysis of the ordered or disordered state of the images to aid in the analysis. With the development of computer technology, researchers have begun to use image processing techniques to visualize infrared thermal images. Based on this, new human-shaped visual structural diagrams are constructed, and each region is filled using a five-color method according to the temperature characteristics of elements. Finally, the human eye observes the human body temperature distribution in the visualized image to aid in the analysis, thereby achieving the visualization and analysis of the sequential characteristics of human body infrared thermal imaging. However, since most existing sequential analysis methods rely on visual analysis of visualized images, they are subject to a certain degree of subjectivity, resulting in relatively low accuracy of the final results. Summary of the Invention
[0003] In view of this, the purpose of this application is to propose a method and related equipment for analyzing the sequence of infrared thermograms.
[0004] To achieve the above objectives, this application provides a method for analyzing the sequence of infrared thermograms, comprising:
[0005] The mean filter is applied to each sample in the acquired infrared thermal image sample set to obtain the filtered temperature value; the infrared thermal image sample set includes infrared thermal image sample sets of normal people and infrared thermal image sample sets of abnormal people;
[0006] Based on the filtered temperature values, a sequential feature vector is constructed for each sample in the sample set;
[0007] The sequential feature vectors are dimensionality reduced to obtain a set of sample points;
[0008] Density clustering is performed on the sample point set, and a minimum neighborhood circle function is constructed based on the sample points after density clustering to obtain the center and radius of the minimum neighborhood circle function;
[0009] The newly acquired infrared thermal image is analyzed based on the center and radius of the circle.
[0010] In one possible implementation, the filtered temperature value is calculated using the following formula:
[0011]
[0012] Where, t′ i,j t represents the temperature value in the i-th row and j-th column after filtering. i+m,j+n This represents the temperature value in the (i+m)th row and (j+n)th column before filtering, with a window size of a×b.
[0013] In one possible implementation, constructing the sequential feature vector for each sample in the sample set based on the filtered temperature value includes:
[0014] Based on the filtered temperature values, a lateral sequential feature vector is constructed for each sample in the sample set, and
[0015] Based on the filtered temperature values, a longitudinal sequential feature vector is constructed for each sample in the sample set.
[0016] In one possible implementation, constructing the lateral sequential feature vector for each sample in the sample set based on the filtered temperature value includes:
[0017] Calculate the mean temperature of each column for each sample in the sample set;
[0018] The lateral sequential feature vector of each sample in the sample set is constructed based on the mean of the temperature in each column.
[0019] In one possible implementation, constructing the longitudinal sequential feature vector for each sample in the sample set based on the filtered temperature value includes:
[0020] Calculate the mean temperature of each row for each sample in the sample set;
[0021] The longitudinal sequential feature vector of each sample in the sample set is constructed based on the mean temperature of each row.
[0022] In one possible implementation, the sequential feature vector includes a horizontal sequential feature vector and a vertical sequential feature vector;
[0023] The step of reducing the dimensionality of the sequential feature vector to obtain the sample point set includes:
[0024] Calculate the magnitude of the horizontal sequential feature vector;
[0025] Calculate the magnitude of the longitudinal sequential feature vector;
[0026] The sample point set is obtained based on the magnitude of the horizontal sequential feature vector and the magnitude of the vertical sequential feature vector.
[0027] In one possible implementation, the step of performing density clustering on the sample point set and constructing a minimum neighborhood circle function based on the density-clustered sample points to obtain the center and radius of the minimum neighborhood circle function includes:
[0028] Set neighborhood parameters;
[0029] Based on the neighborhood parameters and the sample point set, the core object is obtained;
[0030] Traverse all the core objects to obtain a set of category clusters;
[0031] The largest category cluster is selected from the set of category clusters to obtain the category cluster of normal human samples;
[0032] Construct a minimum neighborhood circle function that contains all sample points in the normal human sample category cluster;
[0033] Based on the minimum neighborhood circle function, the center and radius of the minimum neighborhood circle function are obtained.
[0034] In one possible implementation, the analysis of the newly acquired infrared thermal image based on the center and radius of the circle includes:
[0035] The newly acquired infrared thermal image is processed to obtain new sample points;
[0036] If the Euclidean distance from the new sample point to the center of the minimum neighborhood circle function is greater than the radius, the newly acquired infrared thermal image is an abnormal infrared thermal image.
[0037] In response to the fact that the Euclidean distance from the new sample point to the center of the minimum neighborhood circle function is less than or equal to the radius, the newly acquired infrared thermal image is a normal person's infrared thermal image.
[0038] Based on the same inventive concept, embodiments of this application also provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the infrared thermal image sequence analysis method as described in any of the above claims.
[0039] Based on the same inventive concept, embodiments of this application also provide a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute any of the infrared thermogram sequence analysis methods described above.
[0040] As can be seen from the above, the infrared thermogram sequence analysis method and related equipment provided in this application obtain filtered temperature values by performing mean filtering on each sample in the acquired infrared thermogram sample set; the infrared thermogram sample set includes infrared thermogram sample sets of normal individuals and infrared thermogram sample sets of abnormal individuals; based on the filtered temperature values, a sequence feature vector is constructed for each sample in the sample set; the sequence feature vector is subjected to dimensionality reduction processing to obtain a sample point set; density clustering is performed on the sample point set, and a minimum neighborhood circle function is constructed based on the sample points after density clustering to obtain the center and radius of the minimum neighborhood circle function; based on the center and radius, the newly acquired infrared thermogram is analyzed. This application embodiment uses a pre-set sample set to perform density clustering to obtain the center and radius of the minimum neighborhood circle function for normal individual infrared thermogram samples. This application embodiment objectifies the infrared thermogram analysis process, effectively eliminating the influence of subjective factors between different individuals and effectively improving the accuracy of infrared thermogram analysis. Attached Figure Description
[0041] To more clearly illustrate the technical solutions in this application or related technologies, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0042] Figure 1 This is a schematic flowchart of the infrared thermogram sequence analysis method according to an embodiment of this application;
[0043] Figure 2 This is a schematic diagram of the electronic device structure according to an embodiment of this application. Detailed Implementation
[0044] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.
[0045] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of this application should have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms "first," "second," and similar terms used in the embodiments of this application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are only used to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.
[0046] As described in the background section, due to the characteristics of human physiological structure and the different uses of organs, the temperature distribution characteristics of different parts of the human body are different and follow certain patterns. Generally, the skin temperature is lowest at the extremities and higher closer to the center of the torso. These different distribution characteristics reflect the sequential nature of human body temperature distribution. In some cases, the temperature distribution of the human body can become disordered, thus disrupting this sequential nature. Current research on the sequential nature of human body temperature distribution mainly focuses on the sequential analysis of infrared thermal imaging. Most studies rely on visual observation of infrared thermal images, using subjective analysis of the ordered or disordered state of the images to aid in the analysis. With the development of computer technology, researchers have begun to use image processing technology to visualize infrared thermal images. Based on this, a new human-shaped visual structural diagram is constructed, and each region is filled using a five-color method according to the temperature characteristics of elements. Finally, the human body temperature distribution in the visualized image is observed by the human eye to aid in the analysis, thereby realizing the visualization and analysis of the sequential characteristics of human infrared thermal imaging. However, since most existing sequential analysis methods rely on visual analysis of visualized images, they are subject to a certain degree of subjectivity, resulting in relatively low accuracy of the final results.
[0047] In summary, this application proposes a method for analyzing the sequence of infrared thermograms. This method involves applying mean filtering to each sample in an acquired infrared thermogram sample set to obtain filtered temperature values. The infrared thermogram sample set includes infrared thermogram samples from normal individuals and those from abnormal individuals. Based on the filtered temperature values, a sequence feature vector is constructed for each sample in the sample set. The sequence feature vector is then dimensionality-reduced to obtain a set of sample points. Density clustering is performed on the sample point set, and a minimum neighborhood circle function is constructed based on the clustered sample points to obtain the center and radius of the minimum neighborhood circle function. Based on the center and radius, newly acquired infrared thermograms are analyzed. This application utilizes a pre-set sample set for density clustering to obtain the center and radius of the minimum neighborhood circle function for normal human infrared thermogram samples. Furthermore, by objectifying the infrared thermogram analysis process, it effectively eliminates the influence of subjective factors present in existing technologies involving visual analysis and analysis by different individuals, making the infrared thermogram analysis process more objective and thus effectively improving the accuracy of the infrared thermogram analysis results.
[0048] The technical solutions of the embodiments of this application will be described in detail below through specific examples.
[0049] refer to Figure 1 The infrared thermogram sequence analysis method of this application includes the following steps:
[0050] Step S101: Perform mean filtering on each sample in the acquired infrared thermal image sample set to obtain the filtered temperature value; the infrared thermal image sample set includes infrared thermal image sample sets of normal people and infrared thermal image sample sets of abnormal people.
[0051] Step S102: Based on the filtered temperature value, construct the sequential feature vector of each sample in the sample set;
[0052] Step S103: Dimensionality reduction is performed on the sequential feature vector to obtain a set of sample points;
[0053] Step S104: Perform density clustering on the sample point set, and construct a minimum neighborhood circle function based on the sample points after density clustering to obtain the center and radius of the minimum neighborhood circle function;
[0054] Step S105: Analyze the newly acquired infrared thermal image based on the center and radius of the circle.
[0055] For step S101, it is first necessary to obtain an infrared thermal image sample set, which includes infrared thermal image sample sets of normal people and infrared thermal image sample sets of abnormal people. This is because the entire sample set needs to be clustered in the subsequent process. If there is only one sample set, the subsequent clustering cannot be performed.
[0056] It is important to note that the number of samples in the normal human infrared thermal image sample set should be roughly the same as the number of samples in the abnormal human infrared thermal image sample set; otherwise, the accuracy of the final results will be affected.
[0057] After constructing the infrared thermal image sample set, mean filtering is performed on each sample to obtain the filtered temperature value.
[0058] Mean filtering is a typical linear filtering algorithm that replaces the current pixel value with the average of the values of the r×r surrounding pixels. By iterating through every pixel in the image, this method can be used to perform mean filtering on the entire image. Typically, we take the average of all pixels within a region centered on the current pixel, with an equal number of rows and columns.
[0059] In some embodiments, the filtered temperature value is calculated using the following formula:
[0060]
[0061] Where, t′ i,j t represents the temperature value in the i-th row and j-th column after filtering. i+m,j+n This represents the temperature value in the (i+m)th row and (j+n)th column before filtering, with a window size of a×b.
[0062] In this embodiment of the application, normal human samples and abnormal human samples constitute a sample set R = {r1, r2, ..., r...} k ,……,r K}, where r k This represents the infrared thermogram of subject k, where K represents the number of samples in the sample set. It is important to note that the values of a and b mentioned above should both be odd numbers and of moderate magnitude.
[0063] After calculating the filtered temperature value, for step S102, it is necessary to construct the sequential feature vector of each sample in the sample set based on the filtered temperature value.
[0064] In some embodiments, constructing a sequential feature vector for each sample in the sample set based on the filtered temperature value includes: constructing a horizontal sequential feature vector for each sample in the sample set based on the filtered temperature value, and constructing a vertical sequential feature vector for each sample in the sample set based on the filtered temperature value.
[0065] In some embodiments, constructing a lateral sequential feature vector for each sample in the sample set based on the filtered temperature values includes: calculating the mean of the temperature of each column for each sample in the sample set; and constructing the lateral sequential feature vector for each sample in the sample set based on the mean of the temperature of each column.
[0066] In some embodiments, constructing a longitudinal sequential feature vector for each sample in the sample set based on the filtered temperature values includes: calculating the mean of the temperature of each row of each sample in the sample set; and constructing the longitudinal sequential feature vector for each sample in the sample set based on the mean of the temperature of each row.
[0067] In this embodiment, after mean filtering, sequential feature vectors for all samples are constructed.
[0068] For sample r k Calculate the mean temperature of each row in the sample.
[0069] The mean temperature of the i-th row is calculated using the following formula:
[0070]
[0071] Among them, l i a represents the starting column of the human body temperature values in the i-th row. i t′ represents the number of human body temperature values in the i-th row. i,j This represents the temperature value in the i-th row and j-th column of the filtered sample. Construct the vertical sequential feature vector of the sample. Where I represents the total number of rows in the sample.
[0072] For sample r k Calculate the mean temperature of each column in the sample.
[0073] The mean temperature of the j-th column is calculated using the following formula:
[0074]
[0075] Among them, h j This indicates the starting row of the j-th column of human body temperature values, b j This represents the number of human body temperature values in the j-th column, and it is used to construct the horizontal sequential feature vector of the samples. Where J represents the total number of columns in the sample.
[0076] The longitudinal and transverse sequential feature vectors of each sample constitute the sequential feature vector of each sample in the sample set.
[0077] For step S103, after obtaining the sequential feature vector of the sample set, it is necessary to perform dimensionality reduction on the sequential feature vector to obtain the sample point set.
[0078] In some embodiments, the sequential feature vector includes a horizontal sequential feature vector and a vertical sequential feature vector; the step of dimensionality reduction of the sequential feature vector to obtain a sample point set includes: calculating the modulus of the horizontal sequential feature vector; calculating the modulus of the vertical sequential feature vector; and obtaining the sample point set based on the modulus of the horizontal sequential feature vector and the modulus of the vertical sequential feature vector.
[0079] In this embodiment, the obtained feature vector is reduced to a two-dimensional scatter plot. For sample r k Calculate the lateral sequential feature vector of the sample. and longitudinal sequential feature vectors The modulus value is used to obtain the sample r. k Coordinates in the scatter plot
[0080] Specifically, the magnitudes of the horizontal sequential feature vector and the vertical sequential feature vector are calculated using the following formulas:
[0081]
[0082]
[0083] Continuing, a scatter plot is drawn based on the sample coordinates, resulting in the sample point set G = {g1, g2, ..., g...} k ,……,g K}
[0084] For step S105, after obtaining the sample point set, density clustering is performed on the sample point set, and a minimum neighborhood circle function is constructed based on the sample points after density clustering to obtain the center and radius of the minimum neighborhood circle function.
[0085] In some embodiments, performing density clustering on the sample point set and constructing a minimum neighborhood circle function based on the density-clustered sample points to obtain the center and radius of the minimum neighborhood circle function includes: setting neighborhood parameters; obtaining core objects based on the neighborhood parameters and the sample point set; traversing all the core objects to obtain a set of category clusters; selecting the largest category cluster from the set of category clusters to obtain the normal human sample category cluster; constructing a minimum neighborhood circle function containing all sample points in the normal human sample category cluster; and obtaining the center and radius of the minimum neighborhood circle function based on the minimum neighborhood circle function.
[0086] In this embodiment, density clustering is performed on the sample point set G in the scatter plot. Taking the density-based clustering of applications with noise (DBSCAN) method as an example, a neighborhood parameter (∈, MinPts) is set, where ∈ is the neighborhood radius and MinPts is the threshold defining the core point. For the sample point set G = {g1, g2, ..., g...}, ... k ,……,g K}, find sample point g using distance metric k The set of sample points N within the neighborhood of ∈ ∈ (g k )as follows:
[0087] N ∈ (g k )={p∈G|d(g k ,p)<∈}
[0088] Where d(g) k p) represents sample point g k To set N ∈ (g k The Euclidean distance of sample point p in g. k The number of samples in the sub-sample point set within the neighborhood of ∈ satisfies |N ∈ (g k If |≥MinPts, then the sample point g k As the core object, g k Any sample point p in the neighborhood of ∈ is given by g k Density direct access, if a core object g exists in set G. e From the core object g k If the density is directly reached, then g e The sample points in the neighborhood of ∈ are determined by g k Density reachability is transitive. Any core object and all its density-reachable sample points constitute a category cluster. Traversing all core objects yields the final set of category clusters. The largest category cluster in this set is selected as the normal human sample category cluster C = {g′1, g′2, ..., g′}. m ,……,g′ M}, where g′ m This represents the m-th sample point in the normal human sample cluster, where M represents the number of sample points in the cluster.
[0089] Then, a cluster C = {g′1, g′2, ..., g′} containing normal human sample categories is constructed. m ,……,g′ M The minimum neighborhood circle function for all sample points in cluster C. First, randomly select three points g′ in cluster C.α , g′ β , g′ γ .
[0090] The second step is to construct a structure containing g′ α , g′ β , g′ γ The smallest circle.
[0091] The specific process is as follows: in g′ α , g′ β , g′ γ Choose any two points, draw a circle with the line connecting the two points as its diameter, and iterate through g′. α , g′ β , g′ γ For any combination of two points in a circle, if the third point lies inside the circle, then the circle contains g'. α , g′ β , g′ γ The smallest circle containing g′ is defined by its center O1. If the third point is not inside the circle, the method of determining the circle by three points is used to determine the circle containing g′. α , g′ β , g′ γ The smallest circle with center O1.
[0092] The third step is to find the point g′ in C that is farthest from O1. i (1≤i≤M), if g′ i If g′ lies inside or on the circumference of a circle, then that circle is the smallest neighborhood circle we are looking for; i Outside the circle, then at g′ α , g′ β , g′ γ Any three points in g′ can be selected to generate a circle containing g′ using the above two-point or three-point method. α , g′ β , g′ γ , g′ i The smallest circle, these three points become the new g′ α , g′ β , g′ γ Then repeat steps two and three.
[0093] Finally, the smallest neighborhood circle function (x-o1) is obtained. 2 +(y-o2) 2 =ρ 2 The center of the circle is O(o1,o2), and the radius is ρ.
[0094] Then, in step S105, the newly acquired infrared thermal image is analyzed based on the center and radius of the circle.
[0095] In some embodiments, analyzing the newly acquired infrared thermal image based on the center and radius includes: processing the newly acquired infrared thermal image to obtain new sample points; in response to the Euclidean distance from the new sample point to the center of the minimum neighborhood circle function being greater than the radius, the newly acquired infrared thermal image is an infrared thermal image of an abnormal person; in response to the Euclidean distance from the new sample point to the center of the minimum neighborhood circle function being less than or equal to the radius, the newly acquired infrared thermal image is an infrared thermal image of a normal person.
[0096] In this embodiment, sample sequence detection is performed based on the positional relationship between the test sample point and the neighborhood circle. The sample point g is represented by the following formula: k Positional relationship s with the domain circle:
[0097]
[0098] Where d(g) k O) represents sample point g k The Euclidean distance to the center O of the smallest neighborhood circle. When s > 1, it indicates that the sample point g... k Outside the neighborhood circle, the analysis result at this point is poor sequence consistency, meaning that the sample r... k This represents the abnormal human sample category; when s≤1, it indicates that sample point g k Within the neighborhood circle, the analysis results at this point show good sequentiality, i.e., sample r... k This represents the normal human sample category.
[0099] As can be seen from the above embodiments, the infrared thermogram sequence analysis method described in this application analyzes the infrared thermogram sequence by performing mean filtering on each sample in the acquired infrared thermogram sample set to obtain a filtered temperature value. The infrared thermogram sample set includes infrared thermogram samples from normal individuals and infrared thermogram samples from abnormal individuals. Based on the filtered temperature value, a sequence feature vector is constructed for each sample in the sample set. The sequence feature vector is then subjected to dimensionality reduction processing to obtain a sample point set. Density clustering is performed on the sample point set, and a minimum neighborhood circle function is constructed based on the density-clustered sample points to obtain the center and radius of the minimum neighborhood circle function. Based on the center and radius, the newly acquired infrared thermogram is analyzed. This application embodiment uses a pre-set sample set to perform density clustering to obtain the center and radius of the minimum neighborhood circle function for normal human infrared thermogram samples. Furthermore, by objectifying the infrared thermogram analysis process, the influence of subjective factors present in the prior art, such as analysis by the naked eye and analysis by different people, is effectively eliminated, making the infrared thermogram analysis process more objective and thus effectively improving the accuracy of the infrared thermogram analysis results.
[0100] It should be noted that the method in this embodiment can be executed by a single device, such as a computer or server. The method can also be applied in a distributed scenario, where multiple devices cooperate to complete the task. In such a distributed scenario, one of these devices may execute only one or more steps of the method in this embodiment, and the multiple devices will interact with each other to complete the method described.
[0101] It should be noted that the above description describes some embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in a different order than that shown in the above embodiments and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0102] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the infrared thermal image sequence analysis method described in any of the above embodiments.
[0103] Figure 2 This embodiment illustrates a more specific hardware structure of an electronic device, which may include a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, memory 1020, input / output interface 1030, and communication interface 1040 are interconnected internally via the bus 1050.
[0104] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.
[0105] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.
[0106] The input / output interface 1030 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components within the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touchscreens, microphones, various sensors, etc., while output devices may include displays, speakers, vibrators, indicator lights, etc.
[0107] The communication interface 1040 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).
[0108] Bus 1050 includes a pathway for transmitting information between various components of the device, such as processor 1010, memory 1020, input / output interface 1030, and communication interface 1040.
[0109] It should be noted that although the above-described device only shows the processor 1010, memory 1020, input / output interface 1030, communication interface 1040, and bus 1050, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this specification, and not necessarily all the components shown in the figures.
[0110] The electronic devices described above are used to implement the corresponding infrared thermogram sequence analysis method in any of the foregoing embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.
[0111] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the infrared thermogram sequence analysis method as described in any of the above embodiments.
[0112] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.
[0113] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to execute the infrared thermogram sequence analysis method as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.
[0114] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of this application (including the claims) is limited to these examples; within the framework of this application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the embodiments of this application as described above, which are not provided in the details for the sake of brevity.
[0115] Additionally, to simplify the description and discussion, and to avoid obscuring the embodiments of this application, the well-known power / ground connections to integrated circuit (IC) chips and other components may or may not be shown in the provided drawings. Furthermore, the apparatus may be shown in block diagram form to avoid obscuring the embodiments of this application, and this also takes into account the fact that the details of the implementation of these block diagram apparatuses are highly dependent on the platform on which the embodiments of this application will be implemented (i.e., these details should be fully understood by those skilled in the art). While specific details (e.g., circuits) have been set forth to describe exemplary embodiments of this application, it will be apparent to those skilled in the art that the embodiments of this application can be implemented without these specific details or with variations thereof. Therefore, these descriptions should be considered illustrative rather than restrictive.
[0116] Although this application has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed.
[0117] The embodiments of this application are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the embodiments of this application should be included within the protection scope of this application.
Claims
1. A method for analyzing the sequence of infrared thermograms, characterized in that, include: The mean filter is applied to each sample in the acquired infrared thermal image sample set to obtain the filtered temperature value; The infrared thermal image sample set includes infrared thermal image sample sets of normal people and infrared thermal image sample sets of abnormal people; Based on the filtered temperature value, a sequential feature vector for each sample in the sample set is constructed, including: based on the filtered temperature value, a horizontal sequential feature vector for each sample in the sample set is constructed, and based on the filtered temperature value, a vertical sequential feature vector for each sample in the sample set is constructed. The sequential feature vectors are dimensionality reduced to obtain a set of sample points; Density clustering is performed on the sample point set, and a minimum neighborhood circle function is constructed based on the density-clustered sample points to obtain the center and radius of the minimum neighborhood circle function. This includes: setting neighborhood parameters; obtaining core objects based on the neighborhood parameters and the sample point set; traversing all the core objects to obtain a set of category clusters; selecting the largest category cluster from the category cluster set to obtain the normal human sample category cluster; constructing a minimum neighborhood circle function containing all sample points in the normal human sample category cluster; and obtaining the center and radius of the minimum neighborhood circle function based on the minimum neighborhood circle function. Based on the center and radius, the newly acquired infrared thermal image is analyzed, including: processing the newly acquired infrared thermal image to obtain new sample points; in response to the Euclidean distance from the new sample point to the center of the minimum neighborhood circle function being greater than the radius, the newly acquired infrared thermal image is an infrared thermal image of an abnormal person; in response to the Euclidean distance from the new sample point to the center of the minimum neighborhood circle function being less than or equal to the radius, the newly acquired infrared thermal image is an infrared thermal image of a normal person.
2. The method according to claim 1, characterized in that, The filtered temperature value is calculated using the following formula: ； in, Indicates the filtered first... Line 1 The temperature values in the column, Indicates the first before filtering Line 1 The column contains temperature values, and the window size is [value missing]. .
3. The method according to claim 1, characterized in that, The step of constructing a lateral sequential feature vector for each sample in the sample set based on the filtered temperature value includes: Calculate the mean temperature of each column for each sample in the sample set; The lateral sequential feature vector of each sample in the sample set is constructed based on the mean of the temperature in each column.
4. The method according to claim 1, characterized in that, The step of constructing a longitudinal sequential feature vector for each sample in the sample set based on the filtered temperature value includes: Calculate the mean temperature of each row for each sample in the sample set; The longitudinal sequential feature vector of each sample in the sample set is constructed based on the mean temperature of each row.
5. The method according to claim 1, characterized in that, The sequential feature vector includes a horizontal sequential feature vector and a vertical sequential feature vector; The step of reducing the dimensionality of the sequential feature vector to obtain the sample point set includes: Calculate the magnitude of the horizontal sequential feature vector; Calculate the magnitude of the longitudinal sequential feature vector; The sample point set is obtained based on the magnitude of the horizontal sequential feature vector and the magnitude of the vertical sequential feature vector.
6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1 to 5.
7. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to perform the method according to any one of claims 1 to 5.