Small animal repelling trigger method and repelling system

Abstract

The present application relates to the technical field of animal expelling, and more particularly to a small animal expelling triggering method and an expelling system, wherein the method comprises a collection stage, a learning stage, a determination stage and an expelling triggering stage, the determination stage comprises forming a candidate thermal anomaly region, real heat source determination and living target determination. The present application sets the collection stage, the learning stage, the determination stage and the expelling triggering stage, identifies the abnormal pixels to form the candidate thermal anomaly region in the determination stage, carries out the real heat source determination and the living target determination on the candidate thermal anomaly region to determine the living heat source target, and outputs the expelling triggering signal to the expelling module interface when it is determined that there is the real heat source and the living target. A detection triggering method chain is constructed under the conditions of low-resolution thermal array and limited resources, which simultaneously considers the thermal anomaly early warning, the living discrimination and the triggering control.

Description

Technical Field

[0001] This invention relates to the technical field of animal repulsion, and more specifically, to a method and system for triggering the repulsion of small animals. Background Technology

[0002] In locations such as power facilities, farmland, warehouses, and residences, the intrusion of small animals (such as rats, snakes, and cats) can cause short circuits, line damage, food loss, and even hygiene and safety issues. Therefore, deploying a reliable small animal intrusion detection and active repelling system is of great significance. Real-world application environments often present multiple adverse factors, including nighttime conditions, low light, temperature fluctuations, and complex backgrounds (such as vegetation and equipment shadows). Traditional detection solutions based on visible light cameras or ordinary infrared beams are prone to generating numerous false alarms and missed alarms under such conditions, making it difficult to meet actual protection requirements.

[0003] In existing technologies, small animal intrusion detection often employs passive infrared (PIR) sensors or pyroelectric sensors, triggering alarms by detecting temperature changes. Some solutions combine microwave radar or active infrared beams to improve detection reliability. However, non-living heat sources often exist in the field, such as those generated by sunlight, equipment heat dissipation, or pipe heat transfer. Relying solely on absolute temperature values, single-frame spatial morphology, or common thermal anomaly thresholds cannot effectively distinguish between living targets and non-living heat sources. Furthermore, edge nodes typically have limited computing resources, making it difficult to directly run heavy-duty deep learning recognition chains. Thermal anomaly detection alone is insufficient to determine triggering conditions and provide reliable signals to subsequent expulsion modules. Simply stacking detection algorithms, actuator actions, and networking functions can easily lead to unclear main method chains, making it difficult to form a stable and deployable technical solution under resource-constrained conditions. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies, such as difficulty in effectively distinguishing between living and non-living heat sources under limited edge computing resources, and the inability to form a stable and reliable technical solution by relying solely on thermal anomaly detection. This invention provides a small animal expulsion triggering method and expulsion system, which constructs a detection triggering method chain that simultaneously takes into account thermal anomaly early warning, liveness identification, and trigger control under low-resolution thermal array and resource-constrained conditions.

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: According to one aspect of the present invention, a method for triggering the expulsion of small animals is provided, comprising the following steps: Acquisition phase: Acquire thermal imaging temperature frames of the target area; Learning phase: Based on the thermal imaging temperature frames of the target area during the learning phase, a pixel-level background model is established, and the maximum temperature difference between each pixel in the thermal imaging temperature frames and each pixel in the pixel-level background model is statistically analyzed to generate a pixel-level threshold matrix. Judgment phase: includes: Forming candidate thermal anomaly regions: Identifying anomalous pixels based on the current thermal imaging temperature frame, pixel-level background model, and pixel-level threshold matrix; forming candidate thermal anomaly regions based on the anomalous pixels; Real heat source determination: Set a threshold for the rate of change of thermal energy conservation and collect the ambient baseline temperature; calculate the candidate rate of change of thermal energy conservation between consecutive frames of the candidate thermal anomaly region based on the local temperature of the candidate thermal anomaly region and the ambient baseline temperature; when the candidate rate of change of thermal energy conservation is not greater than the threshold, it is determined that a real heat source exists in the candidate thermal anomaly region; when the candidate rate of change of thermal energy conservation is greater than the threshold, it is determined that no real heat source exists in the candidate thermal anomaly region. Live target determination: Set a live target detection threshold and collect the ambient baseline temperature; calculate the candidate temperature following coefficient of the candidate thermal anomaly region within a preset time window based on the local temperature of the candidate thermal anomaly region and the ambient baseline temperature; when the candidate temperature following coefficient is less than the live target detection threshold, it is determined that there is a live target in the candidate thermal anomaly region; when the candidate temperature following coefficient is not less than the live target detection threshold, it is determined that there is no live target in the candidate thermal anomaly region. Expulsion triggering phase: When there is a real heat source and a living target in the candidate thermal anomaly area, the expulsion triggering phase is entered; an expulsion triggering signal is output to the expulsion module interface.

[0006] In one alternative approach, a preheating phase is included before the learning phase; the preheating phase is initiated after system startup to wait for the thermal array and scene temperature distribution in the target area to stabilize.

[0007] In one alternative approach, identifying anomalous pixels based on the current thermal imaging temperature frame, pixel-level background model, and pixel-level threshold matrix specifically includes the following steps: When the absolute temperature difference between the current thermal imaging temperature frame and the corresponding pixel in the pixel-level background model is greater than the threshold value at the corresponding position in the pixel-level threshold matrix, the pixel is an abnormal pixel. The pixel is considered a normal pixel when the absolute temperature difference between the current thermal imaging temperature frame and the corresponding pixel in the pixel-level background model is not greater than the threshold value at the corresponding position in the pixel-level threshold matrix.

[0008] In one alternative approach, forming a candidate thermal anomaly region based on the anomalous pixels specifically includes the following steps: The target region is divided into multiple subpages, and an abnormal distance threshold is determined. Within each subpage, an anomaly center is determined based on the number of anomalous pixels and the spatial clustering of the anomalous pixels. When two adjacent subpages have anomaly centers at the same time, and the anomaly centers of the two adjacent subpages meet the anomaly distance threshold condition, the subpages that meet the condition will be formed into candidate thermal anomaly regions.

[0009] In one alternative approach, calculating the candidate thermal energy conservation rate of change between consecutive frames of the candidate thermal anomaly region based on the local temperature of the candidate thermal anomaly region and the ambient baseline temperature specifically includes the following steps: The total local thermal energy of the candidate thermal anomaly region at the current moment is calculated based on the temperature of each pixel within the candidate thermal anomaly region and the ambient baseline temperature at the current moment. The conserved rate of change of candidate thermal energy between consecutive frames of the candidate thermal anomaly region is calculated based on the total local thermal energy at the current moment and the total local thermal energy at the previous moment.

[0010] In one alternative approach, calculating the candidate temperature following coefficient of the candidate thermal anomaly region within a preset time window based on the local temperature of the candidate thermal anomaly region and the ambient baseline temperature specifically includes the following steps: The candidate average temperature is calculated based on the temperature of each pixel within the candidate thermal anomaly region. The candidate temperature following coefficient for the candidate thermal anomaly region within a preset time window is calculated based on the candidate average temperature and the environmental baseline temperature.

[0011] In one alternative approach, a rollback phase is also included, wherein when problematic frames occur consecutively during the learning phase, or when the judgment phase continues to be abnormal and reaches the relearning condition, the process rolls back to the learning phase to reconstruct the pixel-level background model and the pixel-level threshold matrix.

[0012] In one optional approach, the expulsion triggering phase further includes an expulsion module condition determination. When the expulsion module meets the expulsion module condition determination, the expulsion triggering phase is initiated. The expulsion module condition determination includes at least one of the following: the expulsion module has been started; the expulsion module is not in a paused state; the expulsion module is not within an existing expulsion execution window; and the power of the expulsion module allows for the execution of the expulsion command.

[0013] According to a second aspect of the present invention, a small animal repulsion system is provided for implementing the above-described small animal repulsion triggering method, comprising: Thermal imaging acquisition module: used to acquire thermal imaging temperature frames of the target area; Thermal background learning module: used to establish a pixel-level background model based on the thermal imaging temperature frames of the target area during the learning phase, to statistically analyze the maximum temperature difference between each pixel in the thermal imaging temperature frames and each pixel in the pixel-level background model during the learning phase, and to generate a pixel-level threshold matrix. Abnormal liveness verification module: used to generate candidate thermal anomaly regions, determine real heat sources, and determine live targets; The process of forming candidate thermal anomaly regions includes identifying anomalous pixels based on the current thermal imaging temperature frame, a pixel-level background model, and a pixel-level threshold matrix; and forming candidate thermal anomaly regions based on the anomalous pixels. The determination of a real heat source includes setting a threshold for the rate of change of thermal energy conservation and collecting the ambient baseline temperature; calculating the candidate rate of change of thermal energy conservation between consecutive frames of the candidate thermal anomaly region based on the local temperature of the candidate thermal anomaly region and the ambient baseline temperature; determining that a real heat source exists in the candidate thermal anomaly region when the candidate rate of change of thermal energy conservation is not greater than the threshold; and determining that no real heat source exists in the candidate thermal anomaly region when the candidate rate of change of thermal energy conservation is greater than the threshold. The live target determination includes setting a live target discrimination threshold and collecting the ambient baseline temperature; calculating the candidate temperature following coefficient of the candidate thermal anomaly region within a preset time window based on the local temperature of the candidate thermal anomaly region and the ambient baseline temperature; determining that a live target exists in the candidate thermal anomaly region when the candidate temperature following coefficient is less than the live target discrimination threshold; and determining that no live target exists in the candidate thermal anomaly region when the candidate temperature following coefficient is not less than the live target discrimination threshold. Trigger output module: When there is a real heat source and a living target in the candidate thermal anomaly area, it enters the expulsion triggering stage; and outputs an expulsion trigger signal to the expulsion module interface.

[0014] In one alternative embodiment, the expulsion module interface is coupled to an expulsion device; the expulsion device includes a housing and a button assembly, a power supply assembly, an indicator light, an infrared thermal imaging sensor, an acoustic expulsion mechanism, a light expulsion mechanism, and a controller, all disposed on the housing. The controller is coupled to the expulsion module interface and to the button assembly, the power supply assembly, the indicator light, the infrared thermal imaging sensor, the acoustic expulsion mechanism, and the light expulsion mechanism.

[0015] Compared with the prior art, the beneficial effects of the present invention are: This invention discloses a small animal expulsion triggering method and expulsion system. By setting up a collection phase, a learning phase, a judgment phase, and an expulsion triggering phase, the judgment phase first identifies abnormal pixels to form candidate thermal anomaly regions. Then, it determines the actual heat source and the living target within these candidate thermal anomaly regions. This achieves the construction of a detection triggering method chain that simultaneously considers thermal anomaly early warning, liveness detection, and trigger control under conditions of low-resolution thermal arrays and limited resources. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the small animal expulsion triggering method of the present invention; Figure 2 This is a flowchart illustrating the event recording process in the small animal expulsion triggering method of the present invention; Figure 3 This is a schematic diagram of the wireless backhaul process within the RTC wake-up window in the small animal expulsion triggering method of the present invention; Figure 4 This is a schematic diagram of the small animal repelling system of the present invention; Figure 5 This is a first-view structural schematic diagram of the expulsion device in this invention; Figure 6 This is a second-view structural schematic diagram of the expulsion device in this invention.

[0017] In the attached diagram: 100, housing; 200, button assembly; 300, power supply assembly; 400, indicator light; 500, infrared thermal imaging sensor; 600, acoustic expulsion mechanism; 700, optical expulsion mechanism. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. The present invention will be further described below with reference to specific embodiments. The accompanying drawings are for illustrative purposes only, representing schematic diagrams only, not actual pictures, and should not be construed as limiting this patent. To better illustrate the embodiments of the present invention, some parts in the drawings may be omitted, enlarged, or reduced, and do not represent the actual product dimensions. It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0019] In the accompanying drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components. In the description of the present invention, it should be understood that if terms such as "upper," "lower," "left," "right," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the terms used to describe positional relationships in the drawings are only for illustrative purposes and should not be construed as limiting the present patent. For those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0020] Example 1 This embodiment is the first embodiment of the small animal expulsion triggering method, such as Figure 1 As shown, it includes the following steps: S1: Acquisition Phase: Acquire thermal imaging temperature frames of the target area. In some embodiments, the thermal imaging sensor outputs... The system employs a pixel temperature matrix; multiple thermal imaging sensors can cover multiple detection directions; the tasks in the acquisition phase and the subsequent judgment phase can be run separately to improve real-time performance. Specifically, two front-end controllers can be used, each connected to its respective thermal imaging sensor; each front-end controller can connect to its corresponding thermal imaging sensor via an independent hardware I2C interface; the front-end status can be transmitted to the main controller via a shared signal line; a single device can cover multiple detection directions, and multiple devices can be deployed in key areas. This structure can support the deployment and implementation of the method in multi-directional monitoring scenarios.

[0021] S2: Warm-up Phase: After system startup, a warm-up phase is initiated to allow the thermal array and scene temperature distribution in the target area to stabilize. By setting a warm-up phase, the learning or decision-making process can be avoided directly before the thermal array has stabilized and the ambient temperature has converged. Specifically, the warm-up time can be set to approximately... .

[0022] S3: Learning Phase: After the warm-up phase, the learning phase begins. A pixel-level background model is established based on the sliding window of the thermal imaging temperature frame of the target area during the learning phase. The maximum temperature difference between each pixel in the thermal imaging temperature frame and each pixel in the pixel-level background model is statistically analyzed, and a pixel-level threshold matrix is ​​generated.

[0023] In some embodiments, a pixel-level background model can be represented by the following formula:

[0024] In the formula, This indicates the result obtained after the learning phase ends. The average background value per pixel; This indicates the length of the learning window, which can be set to 20 frames. This represents the set of valid frames during the learning phase. Indicates the first The first frame Temperature value per pixel.

[0025] The pixel-level threshold matrix can be represented by the following formula:

[0026] In the formula, Indicates the first The threshold corresponds to each pixel. The pixel-level threshold matrix corresponds one-to-one with the pixel position and is used pixel by pixel in the subsequent judgment stage; the pixel-level threshold matrix reflects the maximum temperature difference of the pixel relative to the background mean during the learning stage.

[0027] The two sets of formulas above are used to transform the differences in the thermal array scene into pixel-level background representation and pixel-level threshold matrix, so as to perform anomaly identification pixel by pixel in the subsequent judgment stage.

[0028] S4: Judgment Phase: In the judgment phase, the current temperature frame is compared pixel-by-pixel with the pixel-level background model and pixel-level threshold matrix to identify abnormal pixels. The abnormal center is determined based on the number of abnormal pixels and spatial clustering results. Local thermal energy conservation verification and temperature following coefficient discrimination are further performed on the formed candidate thermal anomaly regions to confirm whether the candidate target is a real heat source and a living target. This includes: S41: Formation of candidate thermal anomaly regions: S411: When the absolute temperature difference between the current thermal imaging temperature frame and the corresponding pixel in the pixel-level background model is greater than the threshold at the corresponding position in the pixel-level threshold matrix, the pixel is an abnormal pixel; when the absolute temperature difference between the current thermal imaging temperature frame and the corresponding pixel in the pixel-level background model is not greater than the threshold at the corresponding position in the pixel-level threshold matrix, the pixel is a normal pixel.

[0029] In some embodiments, pixel-level anomaly determination can be expressed as the following formula:

[0030] In the formula, Indicates at time t No. i Anomaly detection value for each pixel, Indicates time The One pixel was identified as an abnormal pixel. Indicates time The One pixel was determined to be a normal pixel.

[0031] S412: Divide the target area into multiple subpages and determine the abnormal distance threshold; Within each subpage, the anomaly center is determined based on the number and spatial clustering of abnormal pixels. Specifically, the abnormal pixel set is clustered and statistically analyzed to determine the anomaly center location. The minimum number of abnormal pixels can be set to... .

[0032] When two adjacent subpages have anomaly centers at the same time, and the anomaly centers of the two adjacent subpages meet the anomaly distance threshold condition, the subpages that meet the condition will be formed into candidate thermal anomaly regions.

[0033] In some embodiments, the target area may be divided into multiple regions, and an anomaly center may be determined in each region.

[0034] The triggering of candidate thermal anomaly regions can be expressed by the following formula:

[0035] In the formula, Indicates the trigger value for the candidate thermal anomaly region. This indicates the current formation of a candidate thermal anomaly region. This indicates that no candidate thermal anomaly region has formed at present; and Representing time respectively The abnormal center obtained between two adjacent subpages or regions; Indicates the empty set; This represents the abnormal distance threshold, which can be set to 5 pixels.

[0036] S42: Determining the True Heat Source: S421: Set the threshold for the rate of change of thermal energy conservation. , can be set to The ambient baseline temperature is collected; this temperature is obtained by a sliding update method based on the average value of edge pixels, and is used to provide an ambient temperature reference for subsequent calculations of candidate region thermal energy and temperature following coefficients without adding additional sensors. Specifically, it is expressed as follows:

[0037] In the formula, Indicates time The ambient baseline temperature; Indicates the smoothing update coefficient; Indicates time Average temperature of edge pixels; Indicates time - 1. Ambient baseline temperature.

[0038] S422: Calculate the candidate thermal anomaly region based on the temperature of each pixel within the candidate thermal anomaly region at the current moment and the ambient baseline temperature. The total local thermal energy at the current moment can be expressed as follows:

[0039] In the formula, Indicates time t The total local thermal energy of the candidate thermal anomaly region relative to the ambient baseline temperature; Indicates time Candidate thermal anomaly regions; express t Time of the first Temperature value per pixel.

[0040] The conserved rate of change of candidate thermal energy between consecutive frames is calculated based on the total local thermal energy at the current moment and the total local thermal energy at the previous moment. Specifically, it can be expressed as the following formula:

[0041] In the formula, This represents the rate of change of candidate thermal energy between consecutive frames; This indicates that a positive number is set to prevent the denominator from being too small; it can be set to approximately 0.3 ^circC.

[0042] S423: When the candidate rate of change of conserved heat energy is not greater than the threshold rate of change of conserved heat energy, that is... When a candidate thermal anomaly region is identified, a real heat source is determined to exist; when the rate of change of the candidate thermal energy conservation exceeds the threshold of the rate of change of thermal energy conservation, that is... When the candidate thermal anomaly region is not found to contain a real heat source, it is determined that there is no real heat source within the region.

[0043] S43: Live target determination: S431: Set the liveness detection threshold , can be set to Collect ambient baseline temperature The ambient baseline temperature can be consistent with the determination of the actual heat source.

[0044] S432: Calculate the candidate average temperature based on the temperature of each pixel within the candidate thermal anomaly region; The candidate temperature following coefficient for the candidate thermal anomaly region within a preset time window is calculated based on the candidate average temperature and the ambient baseline temperature. Specifically, it can be expressed as the following formula:

[0045] In the formula, Indicates time The candidate temperature follower coefficient; Indicates time The average temperature of the candidate thermal anomaly region; This indicates the number of historical frames corresponding to the time window.

[0046] S433: When the candidate temperature following coefficient is less than the liveness detection threshold, that is... When the candidate thermal anomaly coefficient is low, it indicates that the candidate thermal anomaly region has a low degree of following changes in ambient temperature, indicating the presence of a living target within the candidate thermal anomaly region; when the candidate temperature following coefficient is not less than the liveness detection threshold, that is... At that time, it was determined that there were no living targets within the candidate thermal anomaly area.

[0047] S5: Rollback Phase: When problematic frames appear consecutively during the learning phase, or when the decision phase continues to be abnormal and the relearning condition is met, the system rolls back to the S3 learning phase to reconstruct the pixel-level background model and pixel-level threshold matrix. The relearning condition includes relearning conditions during the learning phase and relearning conditions during the decision phase.

[0048] The relearning condition during the learning phase is as follows: In the S3 learning phase, if the deviation of any pixel or a preset number of pixels in the current temperature frame from the average value of the sliding window background exceeds the pre-screening threshold of the corresponding pixel, then the temperature frame is recorded as a problem frame; when the number of consecutive occurrences of problem frames reaches the first learning threshold, the learning window is reinitialized and learning frames are re-acquired. The first learning threshold can be set to 10 frames.

[0049] The relearning condition for the judgment phase is as follows: In the S4 judgment phase, if the number of abnormal pixels, the area of ​​abnormal regions, or the change in the ambient baseline temperature continuously meets the abnormal conditions, but the candidate thermal abnormal region does not simultaneously meet the determination of a real heat source and a living target, or the abnormal region exhibits a large-area, long-term stable background drift characteristic, then the number of abnormal frames in the judgment phase is accumulated; when the number of abnormal frames reaches the second learning threshold, the generated pixel-level threshold matrix and the learning frame count are cleared, and the process reverts to the S3 learning phase. The second learning threshold can be set to 2000 frames.

[0050] The first learning threshold, the second learning threshold, the pre-screening threshold, the abnormal pixel count threshold, and the abnormal region area threshold can all be set according to the sensor frame rate, the installation scenario, and the ambient temperature drift.

[0051] S6: Expulsion Trigger Phase: When a real heat source and a living target exist within the candidate thermal anomaly area, the expulsion trigger phase begins; an expulsion trigger signal is output to the expulsion module interface.

[0052] The trigger signal can be a level signal, a pulse signal, a control word, or other triggering form that can be recognized by subsequent modules; the expulsion module interface can be connected to one or more of the following: acoustic expulsion module, optical expulsion module, jet expulsion module, or mechanical expulsion module; the expulsion module can perform corresponding expulsion actions according to the trigger signal.

[0053] In some embodiments, The working principle of this small animal expulsion triggering method in this embodiment is as follows: By setting up acquisition, learning, judgment, and expulsion triggering stages, the system first identifies abnormal pixels to form candidate thermal anomaly regions during the judgment stage. Then, it determines the actual heat source and the live target within these candidate thermal anomaly regions. When both a real heat source and a live target are identified, an expulsion trigger signal is output to the expulsion module interface. This approach leverages the good adaptability of thermal imaging to environmental temperature drift and establishes a clear, stable, and implementable detection triggering chain on edge devices, including a liveness detection step. It achieves the construction of a detection triggering method chain that simultaneously handles thermal anomaly warning, liveness detection, and trigger control under conditions of low-resolution thermal arrays and limited resources.

[0054] In some embodiments, the learning window length, the anomaly center distance threshold, the minimum number of anomaly pixels, and the warm-up time can also be approximately [values ​​to be filled in]. frame, Pixels Pixels and The threshold values ​​for the rate of change of thermal energy conservation, the minimum change threshold for the ambient baseline temperature, and the threshold value for the temperature following coefficient can be determined based on the calibration at the deployment site. The above values ​​are example parameters and can be adjusted according to the deployment scenario in actual applications.

[0055] By determining the anomaly center relationship, verifying thermal energy conservation, and discriminating the temperature following coefficient, the pixel-by-pixel anomaly detection results can be further converged into live thermal anomaly events, thereby reducing the probability of isolated hot spots, local noise, and non-live heat sources directly triggering the output.

[0056] Example 2 This embodiment is a second embodiment of a small animal repulsion triggering method. This embodiment is similar to the first embodiment, except that, as shown in the following... Figure 1As shown, the expulsion triggering phase also includes expulsion module condition determination. When a real heat source and a living target exist within the candidate thermal anomaly area, and the expulsion module simultaneously meets the expulsion module condition determination, the expulsion triggering phase begins. The expulsion module condition determination includes at least one of the following: the expulsion module has been activated; the expulsion module is not in a paused state; the expulsion module is not within an existing expulsion execution window; and the expulsion module's power level allows for the execution of the expulsion command. In some embodiments, the expulsion module condition determination may further include determining whether the current environmental or temperature state permits triggering. A radar module can also be added, with radar results serving as auxiliary input in the permission determination, but without altering the main thermal imaging early warning chain.

[0057] In some embodiments, such as Figure 2 As shown, before outputting the expulsion trigger signal, an event record corresponding to this trigger can be generated first and written to a circular FIFO, and then the trigger signal output can be executed. The event record includes at least the trigger source, timestamp, device identifier, power status, and verification information; the event FIFO control block stores it separately and includes a verification value; when the FIFO is full, the earliest event is overwritten. Through this implementation, event logging can be completed before the trigger output, facilitating subsequent status recording and information feedback.

[0058] In some embodiments, such as Figure 3 As shown, the event log can perform CAD listening, session establishment, collision avoidance transmission, and event backhaul via the wireless link within the subsequent RTC cycle wake-up window. After wake-up, CAD listening is performed first; upon detecting a valid signal, session establishment begins; the transmitting side uses CSMA / CA; when a busy channel is detected, backoff and countdown freeze recovery are performed; the backhaul content includes the trigger event log and its state context. This implementation method enables low-power event backhaul in multi-node deployment scenarios.

[0059] Example 3 This embodiment is a first embodiment of a small animal repelling system, such as Figure 4 As shown, the method for triggering the expulsion of small animals provided in Embodiment 1 or Embodiment 2 includes: Thermal imaging acquisition module: used to acquire thermal imaging temperature frames of the target area; Thermal background learning module: used to build a pixel-level background model based on thermal imaging temperature frames of the target area during the learning phase, to statistically analyze the maximum temperature difference between each pixel in the thermal imaging temperature frames and each pixel in the pixel-level background model during the learning phase, and to generate a pixel-level threshold matrix. Abnormal liveness verification module: used to generate candidate thermal anomaly regions, determine real heat sources, and determine live targets; Forming candidate thermal anomaly regions involves identifying anomalous pixels based on the current thermal imaging temperature frame, pixel-level background model, and pixel-level threshold matrix; and forming candidate thermal anomaly regions based on the anomalous pixels. The determination of a real heat source includes setting a threshold for the rate of change of thermal energy conservation and collecting the ambient baseline temperature; calculating the rate of change of candidate thermal energy conservation between consecutive frames in the candidate thermal anomaly region based on the local temperature of the candidate thermal anomaly region and the ambient baseline temperature; determining that a real heat source exists in the candidate thermal anomaly region when the rate of change of candidate thermal energy conservation is not greater than the threshold; and determining that no real heat source exists in the candidate thermal anomaly region when the rate of change of candidate thermal energy conservation is greater than the threshold. The live target determination process includes setting a live target discrimination threshold and collecting the ambient baseline temperature; calculating the candidate temperature following coefficient of the candidate thermal anomaly region within a preset time window based on the local temperature of the candidate thermal anomaly region and the ambient baseline temperature; determining that a live target exists in the candidate thermal anomaly region when the candidate temperature following coefficient is less than the live target discrimination threshold; and determining that no live target exists in the candidate thermal anomaly region when the candidate temperature following coefficient is not less than the live target discrimination threshold. Trigger output module: When there is a real heat source and a living target in the candidate thermal anomaly area, it enters the expulsion trigger stage; and outputs the expulsion trigger signal to the expulsion module interface.

[0060] The expulsion module interface is coupled to the expulsion device, and the expulsion module interface is used to provide trigger signals to the expulsion device. For example... Figure 5 and Figure 6 As shown, the expulsion device includes a housing 100 and a button assembly 200, a power supply assembly 300, an indicator light 400, an infrared thermal imaging sensor 500, an acoustic expulsion mechanism 600, a light expulsion mechanism 700, and a controller, all disposed on the housing 100. The controller is coupled to the expulsion module interface and to the button assembly 200, the power supply assembly 300, the indicator light 400, the infrared thermal imaging sensor 500, the acoustic expulsion mechanism 600, and the light expulsion mechanism 700. In some embodiments, the controller may be a microcontroller or a PLC. The button assembly 200 may include touch buttons and mechanical switches, which are located on the top of the housing 100 for easy operation by the user. The power supply assembly 300 may be a battery housed inside the housing 100; alternatively, the power supply assembly 300 may be an external power source, as in the prior art, with a DC 12V socket on the housing 100 for connecting an external power source. The indicator light 400 may be a red, blue, or green light to inform the user of the device's current status. The infrared thermal imaging sensor 500 uses a sensor from the prior art for thermal imaging. The sound-repelling mechanism 600 may include a mechanical bell and a sonic horn. The light-repelling mechanism 700 may include a white light lamp mounted on the side of the housing 100.

[0061] In some embodiments, a condition verification module is further included, which verifies whether the expulsion module meets the expulsion module condition determination. That is, it verifies the running status, permission conditions, and optional auxiliary inputs to determine whether entry into the triggering process is permitted.

[0062] In some embodiments, an event logging module is also included: for generating an event log before triggering the output.

[0063] It also includes an event persistence module: used to write event records to a circular FIFO and overwrite the oldest event when the queue is full.

[0064] It also includes a wireless backhaul module: used to perform CAD listening, session establishment, collision avoidance transmission and event backhaul within the RTC window.

[0065] The above modules can be implemented by the same controller in conjunction with the corresponding program, or by multiple cooperating hardware units, but their functional boundaries should remain consistent with the method steps.

[0066] This invention discloses a small animal expulsion triggering method and expulsion system. Targeting low-resolution thermal arrays and environmental temperature drift scenarios, it employs a three-stage processing approach for tissue detection input: preheating, learning, and judgment. Anomaly judgment is performed using a pixel-level background model and pixel-level threshold matrix, rather than relying on a unified global threshold, which improves the detection stability of low-resolution thermal arrays in environmental temperature drift scenarios. Anomaly center relationship determination narrows pixel-by-pixel anomalies into candidate thermal anomaly regions. Verification of the local thermal energy conservation of candidate regions relative to the environmental baseline temperature confirms that candidate targets are real heat sources rather than random noise, reducing false triggers caused by isolated hotspots, local noise, or transient hot spots. A temperature following coefficient is used to determine the degree to which candidate targets follow changes in environmental temperature, distinguishing between living targets and non-living heat sources, thus differentiating living targets from non-living heat sources formed by sunlight, equipment heat dissipation, or pipe heat transfer. A condition verification step is set before trigger output, unifying the living thermal anomaly results with equipment operating permit conditions into the same triggering main chain. The trigger signal is output through an external expulsion module interface, maintaining a clear interface between detection judgment and expulsion execution. In a preferred embodiment, a recording-before-triggering and cyclic FIFO approach are used. Persistence improves the traceability of the triggering process; the CAD + CSMA / CA backhaul link within the RTC window improves the stability of low-power communication during multi-node deployments.

[0067] The algorithms or displays provided herein are not inherently related to any particular computer, virtual system, or other device. Furthermore, the embodiments of this invention are not directed to any particular programming language.

[0068] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. Similarly, for the sake of brevity and to aid in understanding one or more aspects of the invention, in the description of exemplary embodiments of the invention above, various features of the embodiments are sometimes grouped together in a single embodiment, figure, or description thereof. The claims, which follow the detailed description, are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of the invention.

[0069] Those skilled in the art will understand that the modules in the device of the embodiment can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiment can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components, except that at least some of such features and / or processes or units are mutually exclusive.

[0070] It should be noted that the above embodiments are illustrative of the invention and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be construed as limiting the order of execution.

Claims

1. A method for triggering the expulsion of small animals, characterized in that, Includes the following steps: Acquisition phase: Acquire thermal imaging temperature frames of the target area; Learning phase: Based on the thermal imaging temperature frames of the target area during the learning phase, a pixel-level background model is established, and the maximum temperature difference between each pixel in the thermal imaging temperature frames and each pixel in the pixel-level background model is statistically analyzed to generate a pixel-level threshold matrix. Judgment phase: includes: Forming candidate thermal anomaly regions: Identifying anomalous pixels based on the current thermal imaging temperature frame, pixel-level background model, and pixel-level threshold matrix; forming candidate thermal anomaly regions based on the anomalous pixels; Real heat source determination: Set a threshold for the rate of change of thermal energy conservation and collect the ambient baseline temperature; calculate the candidate rate of change of thermal energy conservation between consecutive frames of the candidate thermal anomaly region based on the local temperature of the candidate thermal anomaly region and the ambient baseline temperature; when the candidate rate of change of thermal energy conservation is not greater than the threshold, it is determined that a real heat source exists in the candidate thermal anomaly region; when the candidate rate of change of thermal energy conservation is greater than the threshold, it is determined that no real heat source exists in the candidate thermal anomaly region. Live target determination: Set a live target detection threshold and collect the ambient baseline temperature; calculate the candidate temperature following coefficient of the candidate thermal anomaly region within a preset time window based on the local temperature of the candidate thermal anomaly region and the ambient baseline temperature; when the candidate temperature following coefficient is less than the live target detection threshold, it is determined that there is a live target in the candidate thermal anomaly region; when the candidate temperature following coefficient is not less than the live target detection threshold, it is determined that there is no live target in the candidate thermal anomaly region. Expulsion triggering phase: When there is a real heat source and a living target in the candidate thermal anomaly area, the expulsion triggering phase is entered; an expulsion triggering signal is output to the expulsion module interface.

2. The small animal expulsion triggering method according to claim 1, characterized in that, Before the learning phase, there is also a preheating phase; the preheating phase: after the system is started, the system first enters the preheating phase to wait for the thermal array and scene temperature distribution in the target area to tend to stabilize.

3. The small animal expulsion triggering method according to claim 1, characterized in that, The identification of abnormal pixels based on the current thermal imaging temperature frame, pixel-level background model, and pixel-level threshold matrix specifically includes the following steps: When the absolute temperature difference between the current thermal imaging temperature frame and the corresponding pixel in the pixel-level background model is greater than the threshold value at the corresponding position in the pixel-level threshold matrix, the pixel is an abnormal pixel. The pixel is considered a normal pixel when the absolute temperature difference between the current thermal imaging temperature frame and the corresponding pixel in the pixel-level background model is not greater than the threshold value at the corresponding position in the pixel-level threshold matrix.

4. The small animal expulsion triggering method according to claim 1, characterized in that, The process of forming candidate thermal anomaly regions based on the anomalous pixels specifically includes the following steps: The target region is divided into multiple subpages, and an abnormal distance threshold is determined. Within each subpage, an anomaly center is determined based on the number of anomalous pixels and the spatial clustering of the anomalous pixels. When two adjacent subpages have anomaly centers at the same time, and the anomaly centers of the two adjacent subpages meet the anomaly distance threshold condition, the subpages that meet the condition will be formed into candidate thermal anomaly regions.

5. The small animal expulsion triggering method according to claim 1, characterized in that, The calculation of the candidate thermal energy conservation rate of change between consecutive frames of the candidate thermal anomaly region based on the local temperature of the candidate thermal anomaly region and the environmental baseline temperature specifically includes the following steps: The total local thermal energy of the candidate thermal anomaly region at the current moment is calculated based on the temperature of each pixel within the candidate thermal anomaly region and the ambient baseline temperature at the current moment. The conserved rate of change of candidate thermal energy between consecutive frames of the candidate thermal anomaly region is calculated based on the total local thermal energy at the current moment and the total local thermal energy at the previous moment.

6. The small animal expulsion triggering method according to claim 1, characterized in that, The step of calculating the candidate temperature following coefficient of the candidate thermal anomaly region within a preset time window based on the local temperature of the candidate thermal anomaly region and the environmental baseline temperature specifically includes the following steps: The candidate average temperature is calculated based on the temperature of each pixel within the candidate thermal anomaly region. The candidate temperature following coefficient for the candidate thermal anomaly region within a preset time window is calculated based on the candidate average temperature and the environmental baseline temperature.

7. The small animal expulsion triggering method according to any one of claims 1 to 6, characterized in that, It also includes a rollback phase, wherein when problematic frames occur consecutively during the learning phase, or when the judgment phase continues to be abnormal and reaches the relearning condition, the system rolls back to the learning phase to reconstruct the pixel-level background model and pixel-level threshold matrix.

8. The small animal expulsion triggering method according to any one of claims 1 to 6, characterized in that, The expulsion triggering phase also includes an expulsion module condition determination. When the expulsion module meets the expulsion module condition determination, the expulsion triggering phase is entered. The expulsion module condition determination includes at least one of the following: the expulsion module has been activated. The expulsion module is not in a paused state; the expulsion module is not within an existing expulsion execution window; the expulsion module has sufficient power to execute the expulsion command.

9. A small animal repelling system, characterized in that, To implement the small animal expulsion triggering method according to any one of claims 1 to 8, comprising: Thermal imaging acquisition module: used to acquire thermal imaging temperature frames of the target area; Thermal background learning module: used to establish a pixel-level background model based on the thermal imaging temperature frames of the target area during the learning phase, to statistically analyze the maximum temperature difference between each pixel in the thermal imaging temperature frames and each pixel in the pixel-level background model during the learning phase, and to generate a pixel-level threshold matrix. Abnormal liveness verification module: used to generate candidate thermal anomaly regions, determine real heat sources, and determine live targets; The process of forming candidate thermal anomaly regions includes identifying anomalous pixels based on the current thermal imaging temperature frame, a pixel-level background model, and a pixel-level threshold matrix; and forming candidate thermal anomaly regions based on the anomalous pixels. The determination of a real heat source includes setting a threshold for the rate of change of thermal energy conservation and collecting the ambient baseline temperature; calculating the candidate rate of change of thermal energy conservation between consecutive frames of the candidate thermal anomaly region based on the local temperature of the candidate thermal anomaly region and the ambient baseline temperature; determining that a real heat source exists in the candidate thermal anomaly region when the candidate rate of change of thermal energy conservation is not greater than the threshold; and determining that no real heat source exists in the candidate thermal anomaly region when the candidate rate of change of thermal energy conservation is greater than the threshold. The live target determination includes setting a live target discrimination threshold and collecting the ambient baseline temperature; calculating the candidate temperature following coefficient of the candidate thermal anomaly region within a preset time window based on the local temperature of the candidate thermal anomaly region and the ambient baseline temperature; determining that a live target exists in the candidate thermal anomaly region when the candidate temperature following coefficient is less than the live target discrimination threshold; and determining that no live target exists in the candidate thermal anomaly region when the candidate temperature following coefficient is not less than the live target discrimination threshold. Trigger output module: When there is a real heat source and a living target in the candidate thermal anomaly area, it enters the expulsion triggering stage; and outputs an expulsion trigger signal to the expulsion module interface.

10. The small animal repelling system according to claim 9, characterized in that, The expulsion module interface is coupled with an expulsion device; the expulsion device includes a housing (100) and a button assembly (200), a power supply assembly (300), an indicator light (400), an infrared thermal imaging sensor (500), an acoustic expulsion mechanism (600), a light expulsion mechanism (700), and a controller, all disposed on the housing (100). The controller is coupled to the expulsion module interface and to the button assembly (200), the power supply assembly (300), the indicator light (400), the infrared thermal imaging sensor (500), the acoustic expulsion mechanism (600), and the light expulsion mechanism (700).

Images