An intelligent alarm method and system based on infrared thermal imaging technology

Abstract

The present invention relates to the field of smart home technology, and specifically discloses an intelligent alarm method and system based on infrared thermal imaging technology, wherein the method includes the following steps: collecting an infrared thermal image sequence of a home environment based on an infrared detector; performing threshold segmentation on each frame image in the infrared thermal image sequence based on a preset temperature threshold to extract a high-temperature area image sequence; using a target tracking algorithm between consecutive frames of the high-temperature area image sequence to form a high-temperature area tracking trajectory across frames; obtaining time-series change parameter information of each high-temperature area based on the high-temperature area tracking trajectory and the high-temperature area image sequence; distinguishing the target type of each high-temperature area based on the time-series change parameter information; generating an alarm signal if there is a high-temperature area of ​​biological target type; the method has the ability to accurately distinguish between biological targets and sunlight targets, effectively improves the reliability and accuracy of detection, and can effectively reduce the false alarm rate.

Description

Technical Field

[0001] The present application relates to the field of smart home technology, and more specifically, to an intelligent alarm method and system based on infrared thermal imaging technology. Background Art

[0002] As a non-contact detection method, infrared thermal imaging technology is playing an increasingly important role in security monitoring, particularly in home security monitoring, because it can operate in low-light or dark conditions and can reflect the temperature of objects. By capturing infrared thermal images of the home environment, the system can detect potential targets with temperature characteristics, thereby providing safety warnings.

[0003] However, in real-world applications, infrared thermal imaging alarm systems face numerous interference sources. One common and difficult-to-solve problem is sunlight interference. Sunlight enters a room through windows, glass doors, and other sources, creating hot spots on the floor or walls. These hot spots are often hotter than the ambient temperature, making them easily misidentified by the system as potential threats. Furthermore, the characteristics of hot spots are complex and dynamically changing. Their position slowly drifts and shifts with changes in the sun's angle; their shape and size can vary due to factors like window structure and obstructions caused by objects inside the room; and their temperature and brightness fluctuate with environmental factors such as external light intensity and cloud cover.

[0004] This complex interference makes it difficult for traditional infrared thermal imaging alarm systems to accurately distinguish between true biological targets (such as humans or pets) and inanimate high-temperature areas caused by sunlight. Due to the high temperatures and dynamic changes in these areas, the system can easily misidentify them as biological targets, resulting in a large number of false alarms. These false alarms not only frequently interrupt users' daily lives, reducing their trust and satisfaction with the alarm system, but are also likely to cause users to lose confidence in the system, leading to them ignoring alarm signals when actual security incidents occur, resulting in serious consequences.

[0005] There is currently no effective technical solution to the above problems. Summary of the Invention

[0006] The purpose of this application is to provide an intelligent alarm method and system based on infrared thermal imaging technology to accurately distinguish biological targets from sunlight targets, improve the reliability and accuracy of detection, and effectively reduce the false alarm rate.

[0007] In a first aspect, the present application provides an intelligent alarm method based on infrared thermal imaging technology, which is applied in a home environment. The method comprises the following steps:

[0008] S1, collecting infrared thermal image sequences of home environment based on infrared detectors;

[0009] S2. performing threshold segmentation on each frame image in the infrared thermal image sequence based on a preset temperature threshold to extract a high temperature area image sequence;

[0010] S3, using a target tracking algorithm between consecutive frames of the high-temperature region image sequence, associating the high-temperature regions between each frame of the high-temperature region image sequence to form a cross-frame high-temperature region tracking trajectory;

[0011] S4. Acquiring time-series variation parameter information of each high-temperature area according to the high-temperature area tracking trajectory and the high-temperature area image sequence;

[0012] S5. Distinguishing target types in each high-temperature area according to the time-series variation parameter information, wherein the target types are biological targets or sunlight targets;

[0013] S6. If there is a high temperature area of ​​biological target type, an alarm signal is generated.

[0014] The intelligent alarm method based on infrared thermal imaging technology of the present application obtains the over-frame high-temperature area tracking trajectory by associating high-temperature areas based on time series, tracks and analyzes the time series change parameter information of the high-temperature areas in sequence, and distinguishes biological targets from sunlight targets by analyzing the time series change parameter information of the high-temperature areas. It has the ability to accurately distinguish biological targets from sunlight targets, effectively improves the reliability and accuracy of detection, and can effectively reduce the false alarm rate.

[0015] The intelligent alarm method based on infrared thermal imaging technology, wherein the time-series change parameter information includes a center position change rate sequence, a regional area change rate sequence, a shape parameter change sequence, a regional average temperature change rate sequence, and a regional internal temperature distribution change sequence.

[0016] In this example, the rate of change sequence of the center position reflects the moving speed and direction of the high-temperature area; the rate of change sequence of the regional area reflects the change in the size of the high-temperature area; the change sequence of the shape parameter reflects the change in the shape of the high-temperature area; the rate of change sequence of the regional average temperature reflects the changing trend of the average temperature of the high-temperature area; and the change sequence of the temperature distribution inside the region reflects the change in the heat distribution inside the high-temperature area.

[0017] The intelligent alarm method based on infrared thermal imaging technology further includes the following steps between step S2 and step S3:

[0018] SA, performing morphological filtering on the high temperature area image sequence.

[0019] In this example, morphological filtering improves the quality of the high-temperature area image sequence, provides more suitable data for subsequent target tracking, reduces tracking errors caused by regional morphology problems, and thus reduces the possibility of false alarms.

[0020] In the intelligent alarm method based on infrared thermal imaging technology, the morphological filtering process includes corrosion and dilation operations, and the corrosion operation includes:

[0021] SA1, performing an erosion operation on each frame of the high-temperature region image sequence based on a predetermined set of structuring elements and a first number of iterations to eliminate small noise points and isolated areas;

[0022] The expansion operation includes:

[0023] SA2 performs an expansion operation on each frame of the corroded high-temperature region image sequence based on a predetermined set of structural elements and a second number of iterations to smooth the region edges and fill the holes inside the region.

[0024] In the intelligent alarm method based on infrared thermal imaging technology, wherein the structural element combination includes a square structural element, a circular structural element, and a linear structural element, step SA1 includes:

[0025] SA11. For each frame image in the high-temperature region image sequence, perform an erosion operation using a square structuring element, a circular structuring element, and a linear structuring element, respectively, to obtain a plurality of eroded frame images.

[0026] SA12. Calculate the edge roughness of each corroded frame image, select the frame image with the lowest edge roughness as the corrosion result of the frame image in the corresponding high-temperature region image sequence, and obtain the corroded high-temperature region image sequence.

[0027] The intelligent alarm method based on infrared thermal imaging technology, wherein step SA2 includes:

[0028] SA21, for each frame image in the high-temperature area image sequence after corrosion, use a square structuring element, a circular structuring element, and a linear structuring element to perform an expansion operation to obtain multiple expanded frame images;

[0029] SA22. Calculate the area change rate between the high-temperature area in each expanded frame image and the high-temperature area of ​​the corresponding frame image in the high-temperature area image sequence before corrosion, and select the frame image with the lowest area change rate as the expansion result of the frame image in the corresponding high-temperature area image sequence to obtain the high-temperature area image sequence after morphological filtering processing.

[0030] The intelligent alarm method based on infrared thermal imaging technology, wherein step S5 includes:

[0031] S51. For each high-temperature region, based on a pre-established parameter feature library for biological targets and sunlight targets, calculate a similarity score between the time-series variation parameter information of each high-temperature region and the corresponding parameter features in the parameter feature library for biological targets and sunlight targets, thereby obtaining a similarity score for the high-temperature region belonging to a biological target or sunlight target.

[0032] S52. Determine the target type of each high-temperature area based on the similarity scores of the biological targets and the sunlight targets and the preset classification rules.

[0033] In the intelligent alarm method based on infrared thermal imaging technology, the target tracking algorithm includes a KCF algorithm and a Kalman filter algorithm, and step S3 includes:

[0034] S31. Initialize the target tracker of each high-temperature area based on the KCF algorithm, and use the target tracker to track the frame image of the high-temperature area image, predict the new position frame by frame and calculate the confidence score. If the confidence score is higher than the preset score threshold, update the target tracker of the corresponding high-temperature area and associate the cross-frame trajectory. If the confidence score is lower than the preset score threshold, predict the potential position through the Kalman filter algorithm and reinitialize the target tracker of the corresponding high-temperature area. If the initialization is successful, continue tracking the high-temperature area, otherwise end the tracking of the high-temperature area.

[0035] S32 , generating a high-temperature region tracking trajectory across frames according to the tracking results of each target tracker.

[0036] The intelligent alarm method based on infrared thermal imaging technology, wherein step S2 includes:

[0037] S21, for each frame image in the infrared thermal image sequence, comparing the temperature value of the pixel point with the preset temperature threshold, extracting the pixel points whose temperature is higher than the preset temperature threshold, and forming a binary image;

[0038] S22, dividing the binary image into connected domains to form an initial high-temperature region;

[0039] S23 , filtering out the initial high-temperature regions smaller than a preset minimum region in the binary image, and determining each frame image in the high-temperature region image sequence.

[0040] In a second aspect, the present application also provides an intelligent alarm system based on infrared thermal imaging technology, which is applied in a home environment. The system includes:

[0041] Infrared detector, used to collect infrared thermal image sequences of the home environment;

[0042] Alarm component, used to perform alarm operations;

[0043] The control component is used to execute the intelligent alarm method based on infrared thermal imaging technology provided in the first aspect to generate an alarm signal to trigger the alarm component to perform an alarm operation.

[0044] The intelligent alarm system based on infrared thermal imaging technology of the present application analyzes the temporal variation characteristics of high-temperature areas. The system can identify sunlight targets with complex variation patterns, avoid misjudging them as intruders, and thus reduce the false alarm rate caused by sunlight interference.

[0045] From the above, it can be seen that the present application provides an intelligent alarm method and system based on infrared thermal imaging technology, wherein the intelligent alarm method based on infrared thermal imaging technology of the present application obtains the over-frame high-temperature area tracking trajectory by associating high-temperature areas based on time series, and sequentially tracks and analyzes the time series change parameter information of the high-temperature areas, and distinguishes biological targets from sunlight targets by analyzing the time series change parameter information of the high-temperature areas. It has the ability to accurately distinguish biological targets from sunlight targets, effectively improves the reliability and accuracy of detection, and can effectively reduce the false alarm rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A flowchart of an intelligent alarm based on infrared thermal imaging technology is provided for some embodiments of the present application.

[0047] Figure 2 A flowchart of an intelligent alarm based on infrared thermal imaging technology is provided for some embodiments of the present application.

[0048] Figure 3 A simplified flowchart of an intelligent alarm based on infrared thermal imaging technology is provided in an embodiment of the present application.

[0049] Figure 4 A schematic diagram of the electrical control structure of an intelligent alarm system based on infrared thermal imaging technology provided in an embodiment of the present application.

[0050] Reference numerals: 201, infrared detector; 202, alarm component; 203, control component. DETAILED DESCRIPTION

[0051] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work fall within the scope of protection of the present application.

[0052] It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and should not be understood as indicating or implying relative importance.

[0053] First, please refer to Figure 1-Figure 3 Some embodiments of the present application provide an intelligent alarm method based on infrared thermal imaging technology, which is used in a home environment to distinguish biological targets from a non-biological high-temperature area formed by sunlight passing through a window. The method includes the following steps:

[0054] S1, collecting infrared thermal image sequences of home environment based on infrared detectors;

[0055] S2. performing threshold segmentation on each frame image in the infrared thermal image sequence based on a preset temperature threshold to extract a high temperature area image sequence;

[0056] S3, using a target tracking algorithm between consecutive frames of the high-temperature region image sequence, associating the high-temperature regions between each frame of the high-temperature region image sequence to form a cross-frame high-temperature region tracking trajectory;

[0057] S4. Acquire temporal variation parameter information of each high temperature area according to the high temperature area tracking trajectory and the high temperature area image sequence;

[0058] S5. Distinguish the target type in each high-temperature area based on the time-series variation parameter information, and the target type is a biological target or a sunlight target;

[0059] S6. If there is a high temperature area of ​​biological target type, an alarm signal is generated.

[0060] Specifically, the infrared detector is preferably a non-cooling infrared detector, and step S1 obtains temperature distribution data about the home environment changing over time through the infrared detector. Step S2 limits the analysis scope to potential heat sources. Step S3 is used to track and obtain the dynamic behavior of each high-temperature area. In order to distinguish these heat sources, the method performs target tracking on the high-temperature area between consecutive image frames. Through tracking, a temporal association between each high-temperature area can be established to form a high-temperature area tracking trajectory. The time-series variation parameter information obtained in step S4 quantifies the characteristics of the high-temperature area changing over time, such as movement speed, area size change, temperature change, etc., so that step S5 can distinguish the target type corresponding to the high-temperature area according to the time-series variation parameter information, and then determine whether it is necessary to execute step S6 to trigger the alarm behavior. As a result, high-temperature areas caused by non-biological heat sources such as sunlight will not trigger an alarm, thereby reducing the false alarm rate.

[0061] More specifically, the core concept of the intelligent alarm method based on infrared thermal imaging technology in the embodiments of the present application is to distinguish biological targets from sunlight by analyzing the dynamic characteristics of high-temperature areas over time. Step S2 extracts all high-temperature areas with temperatures above a preset threshold, forming a sequence of high-temperature area images. These high-temperature areas may contain biological targets and sunlight. To distinguish between them, the method tracks these high-temperature areas between consecutive frames, establishing a tracking trajectory for each high-temperature area over time. This allows the system to track changes in the movement, size, shape, and temperature of each high-temperature area. Next, based on these tracking trajectories and the original image data, the system calculates time-varying parameter information for each high-temperature area, such as movement speed, area change rate, and temperature trend. These parameters are key features for distinguishing biological targets from sunlight, as biological targets typically exhibit specific movement patterns and temperature changes, while sunlight exhibits different time-varying characteristics. Finally, based on these time-varying parameter information, the method determines whether each high-temperature area is a biological target or a sunlight target. If a biological target is determined to be present, an alarm signal is generated. This differentiation method based on time-varying parameters effectively identifies true biological targets and significantly reduces false alarms caused by sunlight.

[0062] More specifically, if it is determined that a biological target exists, the alarm component can send an alarm message to the user's mobile phone application through the network interface based on the alarm information.

[0063] The intelligent alarm method based on infrared thermal imaging technology in the embodiment of the present application obtains the over-frame high-temperature area tracking trajectory by associating high-temperature areas based on time series, sequentially tracks and analyzes the time series change parameter information of the high-temperature areas, and distinguishes biological targets from sunlight targets by analyzing the time series change parameter information of the high-temperature areas. It has the ability to accurately distinguish biological targets from sunlight targets, effectively improves the reliability and accuracy of detection, and can effectively reduce the false alarm rate.

[0064] In some preferred embodiments, the time series variation parameter information includes a center position variation rate sequence, a region area variation rate sequence, a shape parameter variation sequence, a region average temperature variation rate sequence, and a region internal temperature distribution variation sequence.

[0065] Specifically, the rate of change sequence of the center position reflects the moving speed and direction of the high-temperature area; the rate of change sequence of the regional area reflects the change in the size of the high-temperature area; the change sequence of the shape parameters reflects the change in the shape of the high-temperature area; the rate of change sequence of the regional average temperature reflects the changing trend of the average temperature of the high-temperature area; the change sequence of the temperature distribution inside the region reflects the change in the heat distribution inside the high-temperature area.

[0066] More specifically, step S5 can characterize the dynamic characteristics of high-temperature areas by analyzing these multi-dimensional time-series parameter information. The movement, size, shape, temperature, and internal thermal distribution change patterns of biological targets are different from the drift, size, shape, temperature, and internal thermal distribution change patterns of sunlight areas. Based on these differences, it is possible to distinguish whether each high-temperature area is a biological target or a sunlight target. Through this multi-parameter threshold judgment method, it is possible to effectively distinguish between biological targets with mobility and relatively stable body temperature and sunlight areas that change with light and move slowly, thereby improving the accuracy of distinguishing biological targets in an environment with sunlight interference and reducing the false alarm rate.

[0067] In some preferred embodiments, step S4 includes:

[0068] Traverse each high temperature area tracking track and perform the following operations on the high temperature area image sequence:

[0069] S41, extracting high temperature regions corresponding to the currently selected high temperature region tracking trajectory in the high temperature region image sequence to form a high temperature region set;

[0070] S42. Extracting a center position coordinate set of the high temperature region based on the high temperature region set, calculating the Euclidean distance of the center position coordinates between adjacent frames, obtaining a center position change sequence, and then performing a first-order difference on the center position change sequence to obtain a center position change rate sequence;

[0071] S43, counting the number of pixels in the high-temperature region according to the high-temperature region set to obtain a region area sequence, and then performing first-order difference on the region area sequence to obtain a region area change rate sequence;

[0072] S44, calculating and extracting a shape parameter sequence of the high-temperature region based on the high-temperature region set, where the shape parameters of the shape parameter sequence include circularity, rectangularity, and ellipticity, and then performing a first-order difference on the shape parameter sequence to obtain a shape parameter change sequence;

[0073] S45. Extracting the temperature values ​​of all pixels in the high-temperature region according to the high-temperature region set, calculating and obtaining an average temperature sequence, and then performing a first-order difference on the average temperature sequence to obtain a regional average temperature change rate sequence;

[0074] S46. Count the number of pixels of each temperature value in the high temperature region according to the high temperature region set to obtain a temperature histogram set. Then, use Bhattacharyya distance to calculate the similarity of the temperature histograms between adjacent frames to obtain a change sequence of the temperature distribution within the region.

[0075] Specifically, the above processing method traverses each tracked high-temperature area trajectory, collects all high-temperature area images corresponding to the trajectory from the high-temperature area image sequence, and forms a high-temperature area set, which provides a basis for subsequent analysis of the changes of this specific high-temperature area over time.

[0076] More specifically, the center position change rate sequence is obtained by extracting the center coordinates of the high-temperature region in each frame, calculating the Euclidean distance between the center positions of adjacent frames, and obtaining a position change sequence. A first-order difference is then taken on the position change sequence to obtain a center position change rate sequence. This rate sequence reflects the speed and direction of movement of the high-temperature region on the image plane.

[0077] More specifically, the process for obtaining the region area change rate sequence is as follows: The number of pixels in each frame of the high-temperature region is counted to obtain a region area sequence. A first-order difference is then taken on the region area sequence to obtain a region area change rate sequence. This change rate sequence reflects the rate of change in the size of the high-temperature region over time.

[0078] More specifically, the process for obtaining the shape parameter change sequence involves calculating the shape parameters of each high-temperature region in each image frame, including circularity, rectangularity, and ellipticity, to obtain a shape parameter sequence. A first-order difference is then taken on this shape parameter sequence to obtain a shape parameter change sequence. This change sequence reflects the speed and degree of change in the high-temperature region's shape over time.

[0079] More specifically, the process for obtaining a regional average temperature change rate series is as follows: The temperature values ​​of all pixels in each frame of the high-temperature region are extracted, and the average temperature is calculated to obtain an average temperature series. A first-order difference is then taken on the average temperature series to obtain a regional average temperature change rate series. This rate of change series reflects the rate of change of the overall temperature of the high-temperature region over time.

[0080] More specifically, the process for obtaining the temperature distribution change sequence within a region involves counting the number of pixels with each temperature value within each frame of the high-temperature region to obtain a temperature histogram set. The Bhattacharyya distance is then used to calculate the similarity of the temperature histograms between adjacent frames to obtain the temperature distribution change sequence within the region. This change sequence reflects the degree of temporal change in the heat distribution within the high-temperature region.

[0081] More specifically, the above processing method extracts the time-series variation parameter information of high-temperature areas from multiple dimensions such as location, area, shape, average temperature and internal temperature distribution, and focuses on extracting the change rate or degree of change of these parameters. These dynamic characteristics provide a data basis for subsequent accurate target type differentiation, which helps to reduce the false alarm rate caused by sunlight interference.

[0082] In some preferred embodiments, the following steps are further included between step S2 and step S3:

[0083] SA, perform morphological filtering on the high temperature area image sequence.

[0084] Specifically, after extracting the high-temperature area image sequence through threshold segmentation, there may be small noise points or isolated areas in the image, the edges of the high-temperature area may not be smooth, and there may be holes inside the area. These image quality issues will affect the subsequent target tracking algorithm's recognition and association of high-temperature areas, and may cause tracking interruptions or errors. Therefore, the intelligent alarm method based on infrared thermal imaging technology in the embodiment of the present application introduces morphological filtering processing to optimize the high-temperature area image sequence. Morphological filtering processing is an image processing technology that can act on binary images, remove noise points, smooth area edges, fill holes inside the area, and connect disconnected areas. By applying morphological filtering before target tracking, the morphology of the high-temperature area can be improved, making it more suitable for target tracking. Morphological filtering processing improves the quality of the high-temperature area image sequence, provides more suitable data for processing for subsequent target tracking, reduces tracking errors caused by regional morphology problems, and thus reduces the possibility of false alarms.

[0085] In some preferred embodiments, the morphological filtering process includes erosion and dilation operations, and the erosion operation includes:

[0086] SA1, based on a predetermined set of structural elements and a first number of iterations, performing an erosion operation on each frame of the high-temperature region image sequence to eliminate small noise points and isolated areas;

[0087] The expansion operation includes:

[0088] SA2 performs a dilation operation on each frame of the eroded high-temperature area image sequence based on a predetermined set of structural elements and the second number of iterations to smooth the area edges and fill the holes inside the area.

[0089] Specifically, the erosion operation uses a predetermined set of structuring elements and a first number of iterations to process the high-temperature region image. Its function is to shrink the high-temperature region, thereby eliminating small noise points and isolated false high-temperature regions. The dilation operation is performed on the eroded image, using a predetermined set of structuring elements and a second number of iterations. Its function is to expand the high-temperature region, smoothing the region edges and filling small holes within the region. The structuring element set and the number of iterations are parameters that control the erosion and dilation effects. The first and second iterations can be set according to usage requirements, and the first iteration number is preferably equal to or less than the second iteration number.

[0090] More specifically, after threshold segmentation is performed on an infrared thermal image sequence based on a preset temperature threshold to extract a high-temperature region image sequence, the high-temperature region image sequence may contain small noise points, isolated regions, or regions with uneven edges or holes within the regions. These problems can interfere with subsequent target tracking. First, an erosion operation is performed on each frame in the high-temperature region image sequence based on a predetermined set of structuring elements and a first iteration number. The erosion operation shrinks the high-temperature region by removing boundary pixels, effectively eliminating small noise points and isolated regions in the image. Next, a dilation operation is performed on each frame in the eroded high-temperature region image sequence based on a predetermined set of structuring elements and a second iteration number. The dilation operation expands the high-temperature region by adding pixels to its boundary, smoothing the region edges and filling holes within the region. This erosion-then-dilation process removes noise from the image, making the high-temperature region more regular and continuous. After morphological filtering, the high-temperature region image sequence has clearer boundaries, a more complete interior, and less noise. This provides more reliable input for the subsequent target tracking algorithm, improving tracking accuracy and stability. After such processing, the high temperature area in the frame image of the high temperature area image sequence is more suitable for subsequent target tracking.

[0091] In some preferred embodiments, the structural element combination includes square structural elements, circular structural elements, and linear structural elements;

[0092] Step SA1 includes:

[0093] SA11. For each frame image in the high-temperature region image sequence, perform an erosion operation using a square structuring element, a circular structuring element, and a linear structuring element, respectively, to obtain a plurality of eroded frame images.

[0094] SA12, calculate the edge roughness of each frame image after corrosion, select the frame image with the lowest edge roughness as the corrosion result of the frame image in the corresponding high-temperature area image sequence, to obtain the high-temperature area image sequence after corrosion,

[0095] Specifically, using a single type of structuring element for the erosion operation may not be able to handle noise points and isolated areas of different shapes in the image. To solve this problem, the intelligent alarm method based on infrared thermal imaging technology in the embodiment of the present application uses multiple types of structuring elements to perform an erosion operation on each frame of the high-temperature area image sequence in step SA11, thereby obtaining multiple different erosion results. Each structuring element has a different erosion effect on the image, which can produce different responses to image features of different shapes. Then, step SA12 further introduces edge roughness as a criterion for selecting the optimal erosion result. Edge roughness reflects the smoothness of the region boundary. In the erosion operation, the ideal effect is to eliminate small noise points and isolated areas while preserving the shape and boundary of the main area. If the erosion operation fails to eliminate noise or over-erodes the main area, it may result in an unsmooth edge. By calculating the edge roughness of the image after erosion using different structuring elements and selecting the result with the lowest edge roughness, this solution can determine the erosion result that removes small noise points and isolated areas while maintaining a relatively smooth boundary of the main area.

[0096] More specifically, the structuring element combination includes square, circular, and linear structuring elements, each of which has a different effect on features of different shapes in the image (such as point noise, thin lines, and sharp corners). Square structuring elements are sensitive to features in both horizontal and vertical directions, circular structuring elements are sensitive to isotropic features, and linear structuring elements are sensitive to features in specific directions. Therefore, by using multiple structuring elements, different erosion results can be obtained for the same input image, each of which performs differently in removing noise of specific shapes.

[0097] More specifically, step SA12 uses edge roughness as an evaluation criterion, and edge roughness can be obtained by calculating the gradient change of the pixel points at the boundary of the region or the ratio of the boundary perimeter to the area. An image with low edge roughness usually means that noise points and fine structures have been effectively removed, while the boundaries of the main areas have been well preserved. By comparing the edge roughness of different corrosion results, the result with the lowest roughness is selected as the final corrosion output. This selection process enables the corrosion operation to adaptively select the most suitable structural elements or their combined effects according to the specific circumstances of the image content, thereby improving the effect of eliminating small noise points and isolated areas, and providing a more accurate basis for subsequent expansion operations and regional analysis.

[0098] In some preferred embodiments, step SA2 includes:

[0099] SA21, for each frame image in the high-temperature area image sequence after corrosion, use a square structuring element, a circular structuring element, and a linear structuring element to perform an expansion operation to obtain multiple expanded frame images;

[0100] SA22. Calculate the area change rate between the high-temperature area in each expanded frame image and the high-temperature area of ​​the corresponding frame image in the high-temperature area image sequence before corrosion, and select the frame image with the lowest area change rate as the expansion result of the frame image in the corresponding high-temperature area image sequence to obtain the high-temperature area image sequence after morphological filtering processing.

[0101] Specifically, the above-described processing method aims to address the issue of how to select the dilation result for the eroded high-temperature region, while achieving smooth edges and filling holes while minimizing the original shape and area of ​​the high-temperature region and avoiding deformation. Step SA21 dilates the eroded image using a variety of structuring elements of different shapes. Different structuring elements have different dilation effects on the image region, providing a variety of possible dilation results. Step SA22 uses the area change ratio as a selection criterion to determine the dilation result. This step calculates the area change ratio between the high-temperature region obtained after dilation for each structuring element and the original high-temperature region before dilation. The area change ratio reflects the extent to which the dilation operation affects the region size. Selecting a dilation result with a low area change ratio means selecting a dilation method that minimizes the change in the original region's area. Unlike erosion, each application of a structuring element produces a different dilation result image. Comparing the area change ratios of the three dilation results and selecting the dilated image with the low change ratio as the final dilation result for the frame ensures that the dilation operation minimizes the impact on the original size and shape of the high-temperature region while smoothing edges and filling holes, avoiding excessive dilation or shape distortion. In other words, the original size and shape characteristics of the high-temperature region are preserved to the greatest extent possible. Thus, a high-temperature area image sequence after morphological filtering processing is obtained. The high-temperature area in the sequence is closer to the original target, which is beneficial to subsequent analysis and processing, thereby improving the accuracy of the entire alarm method.

[0102] In some preferred embodiments, step S5 includes:

[0103] S51. For each high-temperature region, based on a pre-established parameter feature library for biological targets and sunlight targets, calculate a similarity score between the time-series variation parameter information of each high-temperature region and the corresponding parameter features in the parameter feature library for biological targets and sunlight targets, thereby obtaining a similarity score for the high-temperature region belonging to a biological target or sunlight target.

[0104] S52. Determine the target type of each high-temperature area based on the similarity scores of the biological targets and the sunlight targets and the preset classification rules.

[0105] Specifically, the parameter features of the pre-established parameter feature library of biological targets and sunlight targets include a center position change rate feature sequence, a regional area change rate feature sequence, a shape parameter change feature sequence, a regional average temperature change rate feature sequence, and a regional internal temperature distribution change feature sequence.

[0106] More specifically, the above-described processing method utilizes the temporal variation parameter information of the high-temperature region to determine the target type by comparing it with the parameter features of known target types. Specifically, a pre-established parameter feature library stores typical temporal variation patterns of biological targets and sunlight targets under different circumstances. For example, biological targets typically exhibit rapid, nonlinear changes in movement speed, fluctuating changes in area and shape, and relatively stable or slow temperature changes, while sunlight targets may exhibit slow, linear drift, slow changes in area and shape, and significant temperature changes with light intensity. Calculating a similarity score involves quantitatively comparing the temporal variation information of the current high-temperature region with these typical features. For example, methods such as Euclidean distance, cosine similarity, or correlation coefficient can be used to calculate the similarity between each temporal variation parameter sequence of the current high-temperature region and the corresponding feature sequence in the feature library. These similarity scores reflect the degree of match between the current high-temperature region and the characteristics of biological targets and sunlight targets. Subsequently, the high-temperature region's final type is determined based on pre-set classification rules, such as setting a threshold based on the difference or ratio of similarity scores, or employing classifiers such as support vector machines or neural networks. Therefore, this method can effectively utilize the dynamic characteristics of the high-temperature area for distinction, thereby improving the accuracy of judgment.

[0107] More specifically, after obtaining the temporal variation parameter information for high-temperature regions, this method compares the center position change rate sequence, area change rate sequence, shape parameter change sequence, average temperature change rate sequence, and internal temperature distribution change sequence for each tracked high-temperature region with the corresponding feature sequences in a pre-established biological target parameter signature library and a sunlight target parameter signature library. For example, a similarity score can be calculated between the current high-temperature region's center position change rate sequence and the biological target center position change rate feature sequence, as well as the sunlight target center position change rate feature sequence. Similar similarity calculations are performed for all temporal variation parameter information, resulting in an overall similarity score for the high-temperature region with the biological target feature (which can be the average or weighted average of the similarity scores corresponding to all sequences) and the sunlight target feature. These similarity scores comprehensively reflect the degree of match between the high-temperature region's dynamic characteristics across multiple dimensions and known target types. Finally, based on these similarity scores, a pre-set classification rule is applied to determine the target. For example, if the similarity score with biological target features is significantly higher than that with sunlight target features, the target is identified as a biological target. Conversely, if the similarity score with sunlight target features is significantly higher than that with biological target features, the target is identified as a sunlight target. This judgment method based on the similarity comparison of multi-dimensional temporal features can more robustly cope with the complex and changing characteristics of sunlight targets, effectively distinguishing biological targets from sunlight targets and reducing false alarm rates.

[0108] In some preferred embodiments, step S52 includes:

[0109] S521, calculating the difference between the biological target similarity score and the sunlight target similarity score in each high temperature area to obtain a similarity difference;

[0110] S522: Determine the target type of each high-temperature area based on the similarity difference and the preset classification rules. The classification rules are:

[0111] SB1. Compare the similarity difference value with a first preset difference threshold and a second preset difference threshold, where the first preset difference threshold is greater than the second preset difference threshold;

[0112] SB2. If the similarity difference is greater than the first preset difference threshold, the corresponding high-temperature area is determined to be a biological target. If the similarity difference is less than the second preset difference threshold, the corresponding high-temperature area is determined to be a sunlight target. If the similarity difference is between the second preset difference threshold and the first preset difference threshold, step SB3 is executed.

[0113] SB3. Obtain the average moving speed of the center position of the corresponding high-temperature area based on the high-temperature area image sequence, and compare the average moving speed with the preset speed threshold. If the average moving speed is greater than the preset speed threshold, the corresponding high-temperature area is determined to be a biological target, otherwise it is determined to be a sunlight target.

[0114] Specifically, the above processing method can preliminarily determine the target type by calculating the similarity difference between the biological target similarity score and the sunlight target similarity score. The similarity difference can be positive or negative, and the second preset difference threshold is negative. When the absolute value of the difference is large enough, it indicates a high classification confidence. The judgment is made directly based on the positive or negative difference. The above classification rule determines whether the classification confidence is high enough based on the first preset difference threshold and the second preset difference threshold.

[0115] More specifically, when the similarity difference is between the second and first preset difference thresholds, it indicates that the target type in the high-temperature area is difficult to distinguish based solely on the similarity difference. In this case, the average movement speed at the center of the high-temperature area is introduced as an auxiliary judgment basis. Biological targets typically move at a high speed, while the movement speed of high-temperature areas formed by sunlight is relatively low. By comparing the average movement speed with the preset speed threshold, the target's motion characteristics are used for differentiation. This hierarchical logic combining similarity judgment and movement speed judgment improves classification accuracy and reduces false alarm rates when similarity scores are close.

[0116] In some preferred embodiments, the target tracking algorithm includes a KCF algorithm and a Kalman filter algorithm, and step S3 includes:

[0117] S31. Initialize the target tracker of each high-temperature area based on the KCF algorithm, and use the target tracker to track the frame image of the high-temperature area image, predict the new position frame by frame and calculate the confidence score. If the confidence score is higher than the preset score threshold, update the target tracker of the corresponding high-temperature area and associate the cross-frame trajectory. If the confidence score is lower than the preset score threshold, predict the potential position through the Kalman filter algorithm and reinitialize the target tracker of the corresponding high-temperature area. If the initialization is successful, continue tracking the high-temperature area, otherwise end the tracking of the high-temperature area.

[0118] S32 , generating a high-temperature region tracking trajectory across frames according to the tracking results of each target tracker.

[0119] Specifically, step S3 combines the KCF algorithm and the Kalman filter algorithm in the high-temperature region tracking step. The KCF algorithm is used to initialize the target tracker and perform frame-by-frame tracking, predicting the target's position in the new frame and calculating a confidence score for the tracking result. The confidence score reflects the KCF algorithm's assessment of the reliability of the current tracking result. When the KCF algorithm's confidence score is above a preset score threshold, the KCF tracking result is reliable. The KCF predicted position is then accepted as the target's position in the new frame, and the KCF tracker model (target tracker) is updated, associating the current frame's tracking result with the historical trajectory. When the KCF algorithm's confidence score is below the preset score threshold, it indicates a potential KCF tracking issue, such as a change in target appearance, occlusion, or tracking drift. At this point, the Kalman filter algorithm is introduced. Based on the target's motion model, the Kalman filter predicts its potential position in the current frame based on its motion state in previous frames. Using the Kalman filter's predicted position, the KCF tracker is attempted to reinitialize. If the KCF tracker successfully reinitializes based on the predicted position, it is assumed that the target is likely located near the predicted position, and tracking is resumed and continued. If reinitialization fails, tracking of the high-temperature region is considered interrupted or the target has disappeared, and tracking of the high-temperature region is stopped. This collaborative approach between the KCF and the Kalman filter allows motion prediction information to assist in tracking recovery, even when the target's appearance changes or is temporarily occluded. This generates more stable and accurate cross-frame tracking trajectories for high-temperature regions. Accurate tracking trajectories provide reliable data for subsequent extraction of temporal variation parameters in high-temperature regions, thereby improving the accuracy of distinguishing biological targets from sunlight targets and reducing false alarm rates.

[0120] In some preferred embodiments, step S2 includes:

[0121] S21, for each frame image in the infrared thermal image sequence, comparing the temperature value of the pixel point with a preset temperature threshold, extracting the pixel points whose temperature is higher than the preset temperature threshold, and forming a binary image;

[0122] S22, dividing the binary image into connected domains to form an initial high-temperature region;

[0123] S23 , filtering out the initial high-temperature regions smaller than a preset minimum region in the binary image, and determining each frame image in the high-temperature region image sequence.

[0124] Specifically, step S21 divides the pixels in the image into two categories: pixels with temperatures higher than a threshold and pixels with temperatures not higher than the threshold. Pixels higher than the threshold are marked, for example, assigned a value of 1, to form a binary image containing only the outline of the high-temperature area. Step S22 divides the binary image into connected domains to form initial high-temperature areas. This operation identifies a set of interconnected high-temperature pixels in the binary image and groups them into independent regions. Each connected region is considered a potential high-temperature entity. Step S23 filters out initial high-temperature areas in the binary image that are smaller than a preset minimum area, and determines each frame image in the high-temperature area image sequence. This operation removes connected areas with too small an area by setting a lower limit on the area. These small areas are usually sensor noise or tiny, unimportant heat sources. By removing them, a cleaner and more accurate high-temperature area image sequence can be obtained, and each retained area represents an independent high-temperature entity with a certain area. As a result, the extracted high-temperature area image sequence is clearer, reducing the interference of noise and irregular areas, providing a more reliable basis for subsequent target tracking and time-varying parameter extraction, thereby improving the accuracy of distinguishing biological targets from sunlight targets and reducing the false alarm rate.

[0125] Second, please refer to Figure 4 Some embodiments of the present application further provide an intelligent alarm system based on infrared thermal imaging technology, which is applied in a home environment. The system includes:

[0126] Infrared detector 201, used to collect infrared thermal image sequences of the home environment;

[0127] Alarm component 202, used to perform alarm operations;

[0128] The control component 203 is used to execute the intelligent alarm method based on infrared thermal imaging technology as provided in the first aspect to generate an alarm signal to trigger the alarm component 202 to perform an alarm operation.

[0129] Specifically, the infrared detector 201 is configured to capture a sequence of infrared thermal images of the home environment. The alarm component 202 is configured to execute an alarm action upon receiving an alarm signal. The control component 203 is configured to process the infrared thermal image sequence, execute the intelligent alarm method, and generate an alarm signal based on the processing results. The control component 203 is connected to the infrared detector 201 and the alarm component 202, receiving image data and sending control signals to the alarm component 202. Thus, the system implements intelligent alarm functionality through the collaborative operation of these components.

[0130] Specifically, the system aims to solve the problem of false alarms caused by non-biological heat areas formed by sunlight passing through windows in a home environment. The infrared detector 201 continuously collects a sequence of infrared thermal images of the home environment and inputs these image data into the control component 203. The control component 203 internally executes an intelligent alarm method. The method first processes the input infrared thermal image sequence, such as extracting high-temperature areas in the image. Subsequently, these high-temperature areas are tracked to form a tracking trajectory across frames. Based on the tracking trajectory and the image information of the high-temperature area, the time-series change parameter information of each high-temperature area is calculated. Using this time-series change parameter information, the control component 203 distinguishes whether the high-temperature area is a biological target or a sunlight target. Only when it is identified that the high-temperature area belongs to a biological target, the control component 203 generates an alarm signal. The alarm signal is sent to the alarm component 202, triggering the alarm component 202 to perform an alarm operation.

[0131] The intelligent alarm system based on infrared thermal imaging technology in the embodiment of the present application analyzes the temporal change characteristics of high-temperature areas. The system can identify sunlight targets with complex change patterns and avoid misjudging them as intruders, thereby reducing the false alarm rate caused by sunlight interference.

[0132] In this document, relational terms such as first and second, etc. are used merely to distinguish one entity or operation from another entity or operation, but do not necessarily require or imply any actual relationship or order between these entities or operations.

[0133] The above description is merely an embodiment of the present application and is not intended to limit the scope of protection of the present application. For those skilled in the art, various modifications and variations of the present application are possible. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. An intelligent alarm method based on infrared thermal imaging technology, applied in a home environment, characterized in that: The method comprises the following steps: S1, collecting infrared thermal image sequences of home environment based on infrared detectors; S2. performing threshold segmentation on each frame image in the infrared thermal image sequence based on a preset temperature threshold to extract a high temperature area image sequence; S3, using a target tracking algorithm between consecutive frames of the high-temperature region image sequence, associating the high-temperature regions between each frame of the high-temperature region image sequence to form a cross-frame high-temperature region tracking trajectory; S4. Acquiring time-series variation parameter information of each high-temperature area according to the high-temperature area tracking trajectory and the high-temperature area image sequence; S5. Distinguishing target types in each high-temperature area according to the time-series variation parameter information, wherein the target types are biological targets or sunlight targets; S6, if there is a high temperature area of ​​biological target type, generate an alarm signal; The time series change parameter information includes a sequence of the rate of change of the center position, a sequence of the rate of change of the regional area, a sequence of the change of the shape parameters, a sequence of the rate of change of the regional average temperature, and a sequence of the change of the temperature distribution within the region; Step S4 includes: Traverse each high temperature area tracking track and perform the following operations on the high temperature area image sequence: S41, extracting high temperature regions corresponding to the currently selected high temperature region tracking trajectory in the high temperature region image sequence to form a high temperature region set; S42. Extracting a center position coordinate set of the high temperature region based on the high temperature region set, calculating the Euclidean distance of the center position coordinates between adjacent frames, obtaining a center position change sequence, and then performing a first-order difference on the center position change sequence to obtain a center position change rate sequence; S43, counting the number of pixels in the high-temperature region according to the high-temperature region set to obtain a region area sequence, and then performing first-order difference on the region area sequence to obtain a region area change rate sequence; S44, calculating and extracting a shape parameter sequence of the high-temperature region based on the high-temperature region set, where the shape parameters of the shape parameter sequence include circularity, rectangularity, and ellipticity, and then performing a first-order difference on the shape parameter sequence to obtain a shape parameter change sequence; S45. Extracting the temperature values ​​of all pixels in the high-temperature region according to the high-temperature region set, calculating and obtaining an average temperature sequence, and then performing a first-order difference on the average temperature sequence to obtain a regional average temperature change rate sequence; S46. Count the number of pixels of each temperature value in the high temperature region according to the high temperature region set to obtain a temperature histogram set. Then, use Bhattacharyya distance to calculate the similarity of the temperature histograms between adjacent frames to obtain a change sequence of the temperature distribution within the region.

2. The intelligent alarm method based on infrared thermal imaging technology according to claim 1 is characterized in that: The following steps are included between step S2 and step S3: SA, performing morphological filtering on the high temperature area image sequence.

3. The intelligent alarm method based on infrared thermal imaging technology according to claim 2 is characterized in that: The morphological filtering process includes erosion and dilation operations, and the erosion operation includes: SA1, performing an erosion operation on each frame of the high-temperature region image sequence based on a predetermined set of structuring elements and a first number of iterations to eliminate small noise points and isolated areas; The expansion operation includes: SA2 performs an expansion operation on each frame of the corroded high-temperature region image sequence based on a predetermined set of structural elements and a second number of iterations to smooth the region edges and fill the holes inside the region.

4. The intelligent alarm method based on infrared thermal imaging technology according to claim 3 is characterized in that: The set of structural elements includes square structural elements, circular structural elements and linear structural elements, and step SA1 includes: SA11. For each frame image in the high-temperature region image sequence, perform an erosion operation using a square structuring element, a circular structuring element, and a linear structuring element, respectively, to obtain a plurality of eroded frame images. SA12. Calculate the edge roughness of each corroded frame image, select the frame image with the lowest edge roughness as the corrosion result of the frame image in the corresponding high-temperature region image sequence, and obtain the corroded high-temperature region image sequence.

5. The intelligent alarm method based on infrared thermal imaging technology according to claim 4 is characterized in that: Step SA2 includes: SA21, for each frame image in the high-temperature area image sequence after corrosion, use a square structuring element, a circular structuring element, and a linear structuring element to perform an expansion operation to obtain multiple expanded frame images; SA22. Calculate the area change rate between the high-temperature area in each expanded frame image and the high-temperature area of ​​the corresponding frame image in the high-temperature area image sequence before corrosion, and select the frame image with the lowest area change rate as the expansion result of the frame image in the corresponding high-temperature area image sequence to obtain the high-temperature area image sequence after morphological filtering processing.

6. The intelligent alarm method based on infrared thermal imaging technology according to claim 1 is characterized in that: Step S5 includes: S51. For each high-temperature region, based on a pre-established parameter feature library for biological targets and sunlight targets, calculate a similarity score between the time-series variation parameter information of each high-temperature region and the corresponding parameter features in the parameter feature library for biological targets and sunlight targets, thereby obtaining a similarity score for the high-temperature region belonging to a biological target or sunlight target. S52. Determine the target type of each high-temperature area based on the similarity scores of the biological targets and the sunlight targets and the preset classification rules.

7. The intelligent alarm method based on infrared thermal imaging technology according to claim 1 is characterized in that: The target tracking algorithm includes the KCF algorithm and the Kalman filter algorithm. Step S3 includes: S31. Initialize the target tracker of each high-temperature area based on the KCF algorithm, and use the target tracker to track the frame image of the high-temperature area image, predict the new position frame by frame and calculate the confidence score. If the confidence score is higher than the preset score threshold, update the target tracker of the corresponding high-temperature area and associate the cross-frame trajectory. If the confidence score is lower than the preset score threshold, predict the potential position through the Kalman filter algorithm and reinitialize the target tracker of the corresponding high-temperature area. If the initialization is successful, continue tracking the high-temperature area, otherwise end the tracking of the high-temperature area. S32 , generating a high-temperature region tracking trajectory across frames according to the tracking results of each target tracker.

8. The intelligent alarm method based on infrared thermal imaging technology according to claim 1 is characterized in that: Step S2 includes: S21, for each frame image in the infrared thermal image sequence, comparing the temperature value of the pixel point with the preset temperature threshold, extracting the pixel points whose temperature is higher than the preset temperature threshold, and forming a binary image; S22, dividing the binary image into connected domains to form an initial high-temperature region; S23 , filtering out the initial high-temperature regions smaller than a preset minimum region in the binary image, and determining each frame image in the high-temperature region image sequence.

9. An intelligent alarm system based on infrared thermal imaging technology, used in a home environment, characterized by: The system includes: Infrared detector, used to collect infrared thermal image sequences of the home environment; Alarm component, used to perform alarm operations; A control component is used to execute the intelligent alarm method based on infrared thermal imaging technology as described in any one of claims 1 to 8 to generate an alarm signal to trigger the alarm component to perform an alarm operation.

Images