A method and system for vector-borne disease quarantine at ports of entry and exit based on the Internet of Things

Abstract

This invention provides a method and system for vector-borne disease quarantine at entry and exit ports based on the Internet of Things (IoT). It utilizes IoT technology to synchronously collect environmental parameters and biological activity data within containers, and combines this with infrared thermal imaging to capture the thermal radiation characteristics of organisms obscured by cargo, constructing a dynamic monitoring model. By dynamically calibrating data through correlation with internal and external environmental parameters, it intelligently identifies vector-borne disease activity areas, divides monitoring zones according to cargo distribution density, and uses an obscuration compensation algorithm to correct the boundaries of target areas, generating an accurate biological activity distribution map. Finally, based on spatial positioning and density thresholds, it outputs visualized quarantine results, achieving effective detection of concealed vector-borne organisms. The technical solution provided by this invention, through IoT multi-source data fusion and dynamic calibration technology, significantly improves the detection accuracy and spatial positioning capability of concealed vector-borne organisms in port quarantine.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet of Things sensing, and in particular to a method and system for quarantine of vectors at entry and exit ports based on Internet of Things. BACKGROUND

[0002] In the quarantine scene of containers at entry and exit ports, the hidden activities of vectors (such as rodents and insects) may cause the risk of cross-border biological invasion or epidemic transmission. Due to the complex stacking of goods inside the container, dynamic changes in the environment (fluctuations in temperature and humidity, differences in ventilation conditions), and weak biological activity signals, an automated quarantine technology that can penetrate the obstruction of goods, capture biological activity characteristics in real time, and adapt to multiple interference environments is urgently needed to solve the problems of high false alarm rate and low positioning accuracy in traditional methods.

[0003] The typical scheme for this demand at present is a static threshold detection system based on single infrared thermal imaging: infrared cameras are arranged inside the container, and the thermal radiation area higher than the environmental background is identified through a fixed threshold, the suspected target area is screened in combination with a preset biological thermal feature template (such as the temperature range of rodents), and a thermal imaging alarm result is output. This scheme relies on the direct correlation between thermal radiation intensity and biological size, and does not fuse environmental dynamic parameters or analyze the obstruction effect of goods.

[0004] The main defect of the existing scheme is that it uses a fixed threshold detection method, relies on static analysis of biological thermal radiation intensity based on single infrared thermal imaging data, and has poor adaptability under dynamic environmental interference: when the temperature and humidity inside and outside the container change suddenly or the heat absorption / heat dissipation characteristics of the goods differ significantly, the preset thermal radiation threshold cannot be dynamically calibrated, and the false alarm rate increases; at the same time, the obstruction and attenuation effect of goods stacking density on infrared signals lacks quantitative modeling, the position of the biological hot spot in the thermal imaging map is distorted from the real spatial distribution, and it is difficult to accurately locate the vector activity area through complex goods obstruction environment; in addition, the system does not fuse biological activity sensing data (such as vibration and sound wave) and multi-dimensional environmental parameters (temperature and humidity, gas concentration) for collaborative analysis, and it is difficult to distinguish between mechanical vibration, goods deformation and other interference signals and real biological activity characteristics, and the risk of missed detection is significant. SUMMARY

[0005] The present application provides a method and system for quarantine of vectors at entry and exit ports based on Internet of Things, to solve the problems of high false alarm rate and high missed detection rate caused by inflexible detection, inaccurate positioning, and single data processing in the prior art.

[0006] In a first aspect, the present application provides a method for quarantine of vectors at entry and exit ports based on Internet of Things, comprising:

[0007] The target environmental parameter and biological activity sensing data inside the cargo container are synchronously collected to generate an environmental monitoring data set representing dynamic changes of the environment inside the cargo container;

[0008] The infrared sensing data is used to capture the thermal radiation distribution characteristics inside the cargo container to generate a thermal imaging map containing the hot spot position of biological activity and the distribution density of goods;

[0009] The environmental monitoring data set and the thermal imaging map are time-series correlated, and the environmental interference parameters outside the cargo container are combined to dynamically calibrate the correlation model of the biological activity sensing data and the thermal radiation distribution characteristics, and the target biological activity area matching the activity mode of rodents or insects is screened out;

[0010] According to the shielding effect parameters of the distribution density of goods on infrared thermal radiation in the thermal imaging map, the cargo container is divided into multiple monitoring partitions, and the boundary position of the target biological activity area is spatially calibrated based on the shielding compensation coefficient of each monitoring partition to generate a biological activity distribution map;

[0011] According to the spatial positioning information and density threshold of the target biological activity area in the biological activity distribution map, a visual evaluation result of the quarantine state of the vector organism inside the cargo container is output.

[0012] Optionally, the environmental monitoring data set and the thermal imaging map are time-series correlated, and the environmental interference parameters outside the cargo container are combined to dynamically calibrate the correlation model of the biological activity sensing data and the thermal radiation distribution characteristics, and the target biological activity area matching the activity mode of rodents or insects is screened out, including:

[0013] The biological activity intensity data in the environmental monitoring data set and the temperature data of each thermal radiation point in the thermal imaging map are assigned the same time label to form a synchronous time series pair;

[0014] According to the temperature change rate and airflow disturbance frequency in the environmental interference parameters outside the cargo container, the external interference coefficient of each thermal radiation point in the thermal imaging map is calculated;

[0015] In the correlation model of the biological activity sensing data and the thermal radiation distribution characteristics, an initial weight value is set for the temperature data of each thermal radiation point, and the corresponding relationship between the initial weight value and the biological activity intensity data is determined by a predefined mapping rule;

[0016] Based on the external interference coefficient, the initial weight value of each thermal radiation point is dynamically adjusted, wherein for each increase of a preset unit value of the external interference coefficient, the weight value of the corresponding thermal radiation point is reduced at a preset decay rate to generate a calibrated dynamic weight distribution;

[0017] superimpose the biological activity intensity data in the synchronous time sequence pair with the calibrated dynamic weight distribution to calculate a comprehensive matching degree of the biological activity intensity and the thermal radiation temperature at each time marker;

[0018] extract a variation curve of the comprehensive matching degree within consecutive time markers, and perform morphological matching on a thermal imaging map area corresponding to the variation curve and a pre-stored rodent activity characteristic curve and an insect activity characteristic curve, and retain a thermal imaging map area corresponding to a curve segment with a matching degree higher than a preset similarity as a target biological activity region.

[0019] Optionally, the thermal radiation distribution characteristics inside the cargo container are captured based on infrared sensing data to generate a thermal imaging map containing a biological activity hotspot position and a cargo distribution density, including:

[0020] A plurality of infrared sensor arrays are arranged inside the cargo container to collect temperature data of each monitoring point at a fixed sampling period to form an initial temperature distribution map;

[0021] The initial temperature distribution map is dynamically analyzed to calculate a temperature change rate of each monitoring point within consecutive sampling periods, and a region with a temperature change rate exceeding a set threshold is marked as a dynamic heat source area, and a three-dimensional space coordinate system is established in combination with container internal structure parameters to map the dynamic heat source area into a three-dimensional space model;

[0022] According to cargo loading information, a cargo stacking area is marked in the three-dimensional space model to exclude static high-temperature areas caused by cargo heat generation, and clustering analysis is performed on the remaining dynamic heat source areas to combine regions with similar spatial positions and temperature change characteristics into candidate biological activity areas;

[0023] According to the cargo stacking height and distribution, an infrared shielding coefficient of each candidate biological activity area is calculated to correct temperature measurement errors caused by cargo shielding;

[0024] Regions in the corrected candidate biological activity area that meet preset biological characteristic parameters are determined as final biological activity hotspots, and a thermal imaging map with a layered display function is generated based on the spatial position relationship between the cargo distribution density and the biological activity hotspots.

[0025] Optionally, the boundary positions of the target biological activity region are spatially calibrated based on a shielding compensation coefficient of each monitoring subregion to generate a biological activity distribution map, including:

[0026] A shielding compensation coefficient corresponding to each monitoring subregion is obtained, and the shielding compensation coefficient reflects the shielding degree of the subregion to infrared detection;

[0027] Determine the monitoring zone where each target biological activity area is located, and calculate the boundary correction amount of the area based on the shading compensation coefficient of the monitoring zone; adjust the boundary position of the monitoring zone by expanding or shrinking according to the boundary correction amount;

[0028] The boundary overlap between adjacent areas of each target biological activity area is detected after adjustment, and the boundary fusion process is performed on adjacent areas with overlap.

[0029] The spatial location information of each target biological activity area after processing is integrated with the temperature distribution information in the thermal imaging atlas to generate a biological activity distribution map containing the distribution characteristics of biological activity intensity. In the biological activity distribution map, different marking methods are used to distinguish and display the boundary of the area adjusted by occlusion compensation and the boundary of the original detection area.

[0030] Optionally, based on the spatial location information and density threshold of the target biological activity area in the biological activity distribution map, a visual assessment result of the quarantine status of disease vectors inside the cargo container is output, including:

[0031] The center coordinates of each target biological activity area are extracted from the biological activity distribution map, and the average biological activity signal intensity of each area is calculated.

[0032] The center coordinates of each target biological activity area are mapped onto the three-dimensional spatial model of the cargo container. The projection position of the center coordinates in the vertical direction is adjusted according to the shielding effect parameter of infrared thermal radiation on the cargo distribution density.

[0033] Compare the average biological activity signal intensity of each target biological activity area with a preset density threshold. If the threshold is exceeded, the area is marked as a high-risk area; otherwise, it is marked as a low-risk area.

[0034] In the three-dimensional spatial model, high-risk areas and low-risk areas are marked with different colors, and a color block distribution map covering the internal structure of the cargo container is generated based on the center coordinate projection position of each area.

[0035] The color block distribution map is overlaid with the cargo type data of the cargo containers, and a comprehensive assessment result including the risk level and spatial location of disease vectors is output in the visualization interface.

[0036] Optionally, based on the temperature change rate and airflow disturbance frequency in the environmental interference parameters outside the cargo container, the external interference coefficient of each thermal radiation point in the thermal imaging map is calculated, including:

[0037] Based on the rate of temperature change in the environmental disturbance parameters outside the cargo container, for each thermal radiation point in the thermal imaging spectrum, the difference between the rate of temperature change at the thermal radiation point and the rate of temperature change outside is calculated.

[0038] Based on the airflow disturbance frequency in the environmental interference parameters outside the cargo container, calculate the degree of its influence on the temperature of the thermal radiation point.

[0039] The difference and the degree of influence are added together according to a preset weight to obtain the external interference coefficient of each thermal radiation point; the external interference coefficient of each thermal radiation point is output.

[0040] Optionally, based on the spatial relationship between cargo distribution density and biological activity hotspots, a thermal imaging atlas with layered display capabilities is generated, including:

[0041] Based on the density values ​​of the cargo stacking area, the three-dimensional spatial model is divided into multiple vertically distributed density levels. Each level corresponds to a preset density range, generating a level division plane parallel to the side wall of the cargo container. The division plane isolates the cargo stacking areas of different density levels into independent display layers.

[0042] The coordinates of the biological activity area are automatically matched with the boundary range of each density level to determine the level to which it belongs; hotspot markers are dynamically rendered on the level segmentation plane to generate a heat source distribution map bound to each density level.

[0043] The heat source distribution map is overlaid with the cargo stacking model, so that the hot spot markers of each level coincide with the cargo area location, and an initial thermal imaging map is generated based on the overlay result.

[0044] The initial thermal imaging map is configured for hierarchical visibility according to user operation instructions. The overlay display status is updated according to the configuration results, and the thermal imaging map that supports hierarchical interaction is output.

[0045] Secondly, this application provides an Internet of Things-based vector-borne disease quarantine system for entry and exit ports, comprising:

[0046] The data acquisition module is used to generate an environmental monitoring dataset that characterizes the dynamic changes of the environment inside the cargo container by synchronously acquiring target environmental parameters and biological activity sensor data inside the cargo container.

[0047] The first generation module is used to capture the thermal radiation distribution characteristics inside the cargo container based on infrared sensing data, and generate a thermal imaging map containing the location of biological activity hotspots and the density of cargo distribution.

[0048] The filtering module is used to correlate the environmental monitoring dataset with the thermal imaging map over time, and combine the environmental interference parameters outside the cargo container to dynamically calibrate the correlation model between the biological activity sensing data and the thermal radiation distribution characteristics, and filter out target biological activity areas that match the activity patterns of rodents or insects.

[0049] The second generation module is used to divide the interior of the cargo container into multiple monitoring zones based on the shading effect parameters of the cargo distribution density on infrared thermal radiation in the thermal imaging spectrum, and to spatially calibrate the boundary position of the target biological activity area based on the shading compensation coefficient of each monitoring zone to generate a biological activity distribution map.

[0050] The output module is used to output a visual assessment result of the quarantine status of disease vectors inside the cargo container based on the spatial location information and density threshold of the target biological activity area in the biological activity distribution map.

[0051] Thirdly, embodiments of this application provide a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement an IoT-based vector-borne disease quarantine method at entry and exit ports as described in the first aspect above.

[0052] Fourthly, embodiments of this application provide a computer storage medium storing a computer program, which, when executed by a computer, implements an Internet of Things-based method for vector-borne disease quarantine at ports of entry and exit as described in the first aspect.

[0053] In this embodiment, by simultaneously collecting environmental parameters and biological activity sensor data within the container, a multi-dimensional dynamic monitoring set can be constructed, solving the problem of single-data-dimensional bias. By generating thermal imaging maps under cargo obstruction based on infrared thermal imaging, it is possible to penetrate complex cargo stacking environments and capture biological thermal radiation characteristics. By correlating internal and external parameters over time and dynamically calibrating the correlation model, environmental interference can be suppressed, and rodent / insect activity areas can be accurately screened. By segmenting monitoring zones based on cargo density and calibrating boundaries with occlusion compensation, the true spatial distribution of organisms can be restored, reducing positioning errors. By generating visualized evaluation results through spatial positioning and density thresholds, quarantine status-based early warning can be achieved, improving response efficiency.

[0054] Furthermore, by assigning synchronized time stamps to environmental monitoring data and thermal imaging data, and quantifying the interference coefficient by combining the external temperature change rate and airflow disturbance frequency, the weights of thermal radiation points are dynamically adjusted and the intensity of biological activity is superimposed to calculate the comprehensive matching degree. Based on the morphological characteristics of the matching curves within a continuous time series, a difference comparison is performed with pre-stored biological activity patterns to screen for highly similar target regions. The dynamic weight calibration mechanism suppresses the interference of sudden environmental changes on the detection signal and reduces the false alarm rate; time series feature analysis effectively distinguishes between real biological activity and instantaneous noise, reducing the risk of missed detection; and the combination of difference matching of biological activity patterns improves the accuracy of target classification and enhances the targeting of vector identification.

[0055] These or other aspects of this application will become more apparent in the following description of the embodiments. Attached Figure Description

[0056] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 A flowchart of a vector-borne disease quarantine method based on the Internet of Things provided in this application is shown;

[0058] Figure 2 A schematic diagram of the structure of an Internet of Things-based vector-borne disease quarantine system at an entry-exit port is shown below.

[0059] Figure 3 A schematic diagram of the structure of a computing device provided in this application is shown. Detailed Implementation

[0060] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0061] In some of the processes described in the specification, claims, and accompanying drawings of this application, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not themselves represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a chronological order, nor do they limit "first" and "second" to different types.

[0062] Researchers have found that existing vector-borne disease detection technologies, relying on single infrared static threshold analysis, suffer from high false alarm rates under dynamic environmental interference, distortion of biological location due to cargo obstruction, and missed detection risks caused by the lack of multi-dimensional data collaboration. Based on this, an IoT-based vector-borne disease quarantine method for entry and exit ports is proposed. This method can accurately locate biological activity areas by penetrating cargo obstruction through multi-source data fusion and dynamic weight calibration. Combined with temporal feature matching, it effectively distinguishes between real biological activity and mechanical noise, significantly reducing false alarm and missed detection rates. The technical solution of this application can be applied to scenarios such as automated customs container quarantine and cross-border logistics biosafety monitoring.

[0063] The entire R&D process demonstrates how to simultaneously collect multi-dimensional environmental parameters and biological activity sensing data within containers, dynamically capture thermal radiation distribution characteristics using infrared thermal imaging, and dynamically calibrate thermal radiation weights using time-series correlation and external environmental interference parameters to effectively suppress interference from sudden temperature and humidity changes and cargo obstruction on detection signals. Based on cargo density segmentation and the introduction of an obstruction compensation algorithm, the spatial distribution of biological activity areas is accurately reconstructed. Target areas are selected through intelligent matching of time-series feature curves and pre-stored biological activity patterns to achieve rodent / insect classification and identification. Finally, a visualized biological distribution heatmap and quarantine assessment results are output, significantly improving the accuracy, positioning precision, and quarantine decision-making efficiency of vector-borne disease detection in complex obstructed environments.

[0064] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0065] Figure 1 This application provides a flowchart of an IoT-based vector-borne disease quarantine method for embodiments of this application, such as... Figure 1 As shown, the method includes:

[0066] 101. By synchronously collecting target environmental parameters and biological activity sensor data inside the cargo container, an environmental monitoring dataset characterizing the dynamic changes of the environment inside the cargo container is generated.

[0067] An environmental monitoring dataset refers to a dynamic data set formed by fusing internal environmental parameters of a container, collected synchronously by temperature, humidity, gas, and vibration sensors, with biological activity sensing data. Biological activity sensing data specifically refers to physical signals generated by rodent or insect activity captured by piezoelectric or acoustic sensors. Time series correlation refers to aligning data from different sensors according to a unified timestamp for analyzing the causal relationship between environmental changes and biological activity.

[0068] In this embodiment, temperature and humidity sensors, gas sensors, and piezoelectric vibration sensors are first deployed inside the cargo container to collect target environmental parameters and biological activity sensing signals at a frequency of once per second, and then transmitted to an edge computing node via an IoT gateway. Next, timestamp alignment technology based on the NTP protocol is used to unify the data streams from different sensors to the same time base, eliminating data deviations caused by transmission delays. Finally, a Kalman filter algorithm is used to denoise and fuse the target environmental parameters and biological activity sensing data, generating an environmental monitoring dataset containing three-dimensional features including timestamps, target environmental parameter values, and vibration intensity. For example, when the vibration intensity exceeds 50Hz and the humidity suddenly increases by 5%, the system automatically marks it as a suspicious biological activity event.

[0069] In an inbound container inspection scenario at an international port, the system deploys temperature and humidity sensors, gas sensors, and piezoelectric vibration sensors at multiple points inside the containers. The sensors synchronously collect data with millisecond-level precision. For example, if a sudden temperature fluctuation from a stable value is detected within a certain time period, and the vibration sensor simultaneously captures irregular pulse signals at a specific frequency band, the system uses a timestamp alignment module to match the temperature fluctuation curve with the peak time axis of the vibration signal, finding a high degree of overlap between the two in the time dimension. The data fusion module further employs a Kalman filter algorithm to eliminate sensor noise, generating an environmental monitoring dataset containing temperature-humidity-vibration correlation features. Three anomalous data windows are marked within this dataset, serving as input sources for subsequent analysis.

[0070] 102. Based on infrared sensing data, capture the thermal radiation distribution characteristics inside the cargo container to generate a thermal imaging map containing the locations of biological activity hotspots and the density of cargo distribution.

[0071] The thermal imaging map is a thermal radiation distribution map generated by scanning the interior of the container using an infrared thermal imager. Biological activity hotspots refer to areas where the local temperature is more than 2 degrees Celsius above the environmental baseline. Cargo distribution density refers to the cargo packing density parameter calculated based on the degree of thermal radiation attenuation; higher density results in lower infrared transmittance. The shading effect parameter quantifies the attenuation coefficient of the cargo's thermal radiation signal and is used for subsequent spatial calibration.

[0072] In this embodiment, a FLIR T865 thermal imager is first used to perform a panoramic scan of the cargo container. The thermal radiation distribution characteristics inside the container are obtained with a thermal sensitivity of 0.5 degrees Celsius. Hotspot areas with temperatures above 35 degrees Celsius are extracted using a threshold segmentation algorithm. Next, the cargo shading effect is calculated based on the heat conduction equation, establishing an inverse proportional relationship model between thermal radiation intensity and cargo distribution density. For example, a wooden cargo box with a density of 200 kg / m³ will reduce thermal radiation intensity by 60%. Based on this, the system generates a thermal imaging map containing cargo distribution density and biological activity hotspots.

[0073] When performing infrared scanning on the same container, the thermal imager detected an anomaly in the ring-shaped thermal radiation in the left front area. Analysis using a thermal conduction model revealed that the temperature gradient in this area did not conform to the natural heat dissipation patterns of the cargo. Combined with continuous high-frequency signals recorded by vibration sensors during the corresponding time period, the system inferred the presence of a biological heat source. Simultaneously, a thermal barrier formed by multiple stacked wooden crates on the right rear side was identified. Based on a cargo density distribution model, the system calculated the thermal attenuation coefficient and automatically labeled this area as a "low-confidence monitoring zone." In the final generated thermal image, the left front area was marked as a primary hotspot, accompanied by a transparency layer indicating the extent of cargo obstruction, providing spatial reference for subsequent analysis.

[0074] 103. The environmental monitoring dataset and the thermal imaging map are correlated over time. Combined with the environmental interference parameters outside the cargo container, the correlation model between the biological activity sensing data and the thermal radiation distribution characteristics is dynamically calibrated to screen out target biological activity areas that match the activity patterns of rodents or insects.

[0075] Environmental disturbance parameters include factors such as wind speed and light intensity outside the container that affect the internal thermal balance; the correlation model refers to the matching rules between biological activity signals and thermal radiation characteristics established by the random forest algorithm, such as the pattern that rodent activity is often accompanied by intermittent vibration and local temperature rise of 1.5 to 3 degrees Celsius.

[0076] In this embodiment, the environmental monitoring dataset generated in step 101 and the thermal imaging map generated in step 102 are first matched by timestamps to generate a spatiotemporal matrix through time series correlation. Next, external weather station data is introduced, and the influence coefficient of external wind speed on the internal temperature of the cargo container is calculated using the heat transfer equation, dynamically correcting the correlation model between the biological activity sensing data and the thermal radiation distribution characteristics. Finally, a pre-trained YOLO-Bio model is used to identify target areas that match the characteristics of rodents or mosquitoes, such as rodent activity exhibiting vibration frequencies of 70 to 120 Hz accompanied by engine temperature rise, thus filtering out target biological activity areas that match the activity patterns of rodents or insects.

[0077] In the spatiotemporal correlation analysis phase, the system overlays and compares the abnormal data windows marked in step 101 with the hotspot areas in step 102. By introducing external meteorological data, it was found that the top of the container was affected by strong sunlight during the detection period. The thermal radiation model automatically deducts the surface temperature rise interference value, correcting the temperature of the hotspot on the left front side from the initial detection value to a reasonable range for biological activity. The random forest model simultaneously analyzes the time-frequency characteristics of the vibration signal, identifies short-period pulse patterns consistent with rodent gnawing behavior (such as a 2-millisecond spike every 5 seconds), and combines this with the corrected thermal radiation data to raise the confidence level of the area to a high-risk level, generating a list of target biological activity areas with time-space dual-dimensional markers.

[0078] 104. Based on the shading effect parameters of the cargo distribution density on infrared thermal radiation in the thermal imaging spectrum, the interior of the cargo container is divided into multiple monitoring zones, and the boundary position of the target biological activity area is spatially calibrated based on the shading compensation coefficient of each monitoring zone to generate a biological activity distribution map.

[0079] The monitoring zoning divides the container into 3x3 grid units based on the cargo distribution density; the occlusion compensation coefficient refers to the compensation value for the loss of thermal radiation signal caused by cargo occlusion in each grid unit, which is calculated through the backpropagation algorithm; spatial calibration refers to the sub-pixel level position correction of the target area boundary.

[0080] In this embodiment, firstly, based on the occlusion effect parameters of the cargo distribution density gradient on infrared thermal radiation in the thermal imaging map, the watershed algorithm is used to divide the container into 9 monitoring zones. For example, when the high-density zone is greater than 150 kg per cubic meter, it is divided into a 1-meter by 1-meter grid. Then, an occlusion compensation coefficient is applied to each grid, for example, a compensation coefficient of 1.32 corresponds to a density of 200 kg per cubic meter. The Gaussian kernel interpolation algorithm is used to correct the boundary of the target biological activity area from the original coordinates to the new coordinates, and finally, a calibrated biological activity distribution map is generated.

[0081] For the hotspot area on the left front side, the system divides the container into nine monitoring sub-regions based on the cargo distribution density in the thermal imaging map. In the third sub-region, the thermal radiation signal is attenuated due to the stacking of metal components. The compensation algorithm, based on the material's thermal conductivity and layered structure, reconstructs the thermal radiation distribution of the obscured area, revealing a positioning deviation at the original hotspot boundary. Subpixel-level correction of the coordinates is performed using a Gaussian kernel interpolation algorithm. In the final output biological activity distribution map, the target area boundary shifts towards the corner of the container, and comparative heatmaps before and after compensation are marked, significantly improving the positioning accuracy of high-risk areas.

[0082] 105. Based on the spatial location information and density threshold of the target biological activity area in the biological activity distribution map, output the visual assessment result of the quarantine status of disease vector organisms in the cargo container.

[0083] The density threshold refers to the risk level standard for disease vectors set according to the national standard GB / T 23797. For example, if the area where rodents are active exceeds 0.5 square meters and they are present continuously for 30 minutes, a red alert is triggered. The visualization assessment results use a heat map overlaid with AR tags to present the quarantine status.

[0084] In this embodiment, the spatial location information, density threshold, and duration of the target biological activity area in the biological activity distribution map are first input into an SVM classifier, which outputs three levels of results: high risk requiring container inspection, medium risk requiring disinfection, and low risk allowing passage. Then, the risk area is projected onto a 3D model of the container using the Unity3D engine, with red representing rodent activity areas and yellow representing insect activity areas, and treatment suggestions are labeled. Finally, a report with a visual assessment result is generated.

[0085] During the decision-making output phase, the system inputs the calibrated biological activity distribution map into the risk quantification engine. The target area, due to its sustained activity duration exceeding a threshold and spatial distribution matching rodent nest characteristics, is classified as "Level 1 Risk." The augmented reality module projects the analysis results onto the container's 3D model, generating a red pulse warning zone in the front left corner, while simultaneously overlaying handling suggestions from similar historical cases (such as "prioritize negative pressure isolation"). When quarantine personnel view the data via mobile devices, the system simultaneously pushes a complete data chain analysis report for the container, including raw sensor data waveforms, visualizations of the thermal imaging correction process, and a risk assessment logic tree, forming a traceable chain of decision support evidence.

[0086] This solution, based on IoT multi-source sensing and intelligent analysis technology, constructs a closed-loop quarantine system from dynamic environmental monitoring to visualized decision-making. By simultaneously collecting environmental parameters and biological activity signals, combined with thermal radiation characteristic analysis, a spatiotemporal correlation model is established to eliminate the influence of external interference. A cargo shading effect compensation algorithm and a gridded spatial calibration mechanism are introduced to overcome the thermal signal distortion problem caused by cargo stacking. Finally, through a risk quantification model and augmented reality technology, complex data is transformed into intuitive quarantine decisions. The entire system achieves technological breakthroughs in non-invasive detection, biological activity positioning accuracy, and dynamic environmental adaptability, significantly improving the efficiency and accuracy of port quarantine and providing an intelligent solution for vector-borne disease control in large-scale logistics scenarios.

[0087] In some embodiments, the environmental monitoring dataset is correlated with the thermal imaging atlas over time, and combined with environmental disturbance parameters outside the cargo container, the correlation model between the biological activity sensing data and the thermal radiation distribution characteristics is dynamically calibrated to screen out target biological activity areas that match the activity patterns of rodents or insects, including:

[0088] 201. Assign the same time stamp to the biological activity intensity data in the environmental monitoring dataset and the temperature data of each thermal radiation point in the thermal imaging atlas to form a synchronized time series pair;

[0089] Synchronizing time series pairs is the process of assigning the same time stamp to biological activity intensity data in an environmental monitoring dataset and temperature data at each thermal radiation point in a thermal imaging atlas, ensuring precise alignment of the two types of data in the time dimension. Time stamping is achieved through precise time synchronization protocols, such as Network Time Protocol (NTP) or precision clock synchronization techniques, eliminating microsecond-level time differences between devices. Biological activity intensity data includes peak energy recorded by vibration sensors or spectral amplitudes captured by acoustic sensors, while thermal radiation point temperature data is a pixel-level temperature matrix acquired by an infrared thermal imager.

[0090] In this embodiment, a time synchronization server is first deployed to provide a unified clock source for devices such as vibration sensors and thermal imagers. The biological activity intensity data in the environmental monitoring dataset is continuously recorded at a millisecond-level sampling frequency; for example, vibration signals are sampled 1000 times per second, and thermal imaging data generates 30 frames of infrared images per second. A timestamp alignment algorithm is used to match the time-domain waveform of the vibration signal with the temperature data of each thermal radiation point in the thermal imaging spectrum with microsecond-level precision. Interpolation is used to handle clock drift between sensors; for example, for missing time points between thermal imaging frames, intermediate values ​​are calculated based on temperature data from adjacent frames. Finally, a three-dimensional synchronized time series pair containing a time axis, vibration energy, and temperature matrix is ​​generated as the input basis for subsequent analysis.

[0091] 202. Based on the temperature change rate and airflow disturbance frequency in the environmental interference parameters outside the cargo container, calculate the external interference coefficient of each thermal radiation point in the thermal imaging map;

[0092] The external disturbance coefficient is a parameter that quantifies the impact of the external environment on the internal thermal radiation of a container. It includes two core dimensions: the rate of temperature change and the frequency of airflow disturbance. The rate of temperature change refers to the amplitude of fluctuation in the external ambient temperature per unit time, and the frequency of airflow disturbance is calculated by collecting the periodic variation characteristics of airflow from wind speed sensors or pressure sensors.

[0093] In this embodiment, real-time environmental interference parameters, such as the rate of temperature change and airflow disturbance frequency (e.g., hourly temperature fluctuations) and the spectral analysis results of wind speed sensors, are first obtained from a weather station outside the cargo container. A thermodynamic interference model is established, and the degree of external interference at each thermal radiation point is calculated using finite element simulation. The degree of external interference is represented by an external interference coefficient. For example, considering the thermal conductivity of the container wall, the heat conduction equation is derived to calculate the impact of sudden external temperature changes on internal thermal imaging. Simultaneously, the local convective heat dissipation effect caused by airflow disturbance is analyzed, generating an external interference coefficient matrix that corresponds one-to-one with the pixels of the thermal imaging spectrum. Regions with high interference coefficients in the external interference coefficient matrix identify thermal radiation points susceptible to external environmental influences for subsequent weight adjustment.

[0094] 203. In the correlation model between the biological activity sensing data and the thermal radiation distribution characteristics, an initial weight value is set for the temperature data of each thermal radiation point, and the correspondence between the initial weight value and the biological activity intensity data is determined by a predefined mapping rule.

[0095] The initial weight values ​​are initial quantitative indicators of the correlation between thermal radiation point temperature data and biological activity signals, generated according to predefined rules. For example, when vibration energy exceeds a threshold, the initial weight of the thermal radiation point within the corresponding time window is increased, and vice versa. The mapping rules can be trained based on historical data or defined based on expert experience.

[0096] In this embodiment, a predefined rule base for biological activity characteristics is used to quantify the physical correlation between vibration energy and temperature data at each thermal radiation point into initial weight values ​​within the correlation model between the biological activity sensing data and thermal radiation distribution characteristics. First, the vibration energy and temperature data at each thermal radiation point in the synchronized time series are normalized: vibration energy is compressed to the 0-1 range using a maximum-minimum method, and the temperature data is linearly scaled based on the baseline value of the internal environment of the cargo container. Then, the rule base matching engine is invoked to dynamically assign weights based on the normalized bimodal data. For example, when the vibration energy exceeds 0.7 and the temperature value is 30% higher than the baseline, it is determined to be a high-probability event of rodent activity, and an initial weight of 0.8 is assigned to the corresponding thermal radiation point; if the vibration energy is below 0.3 and the temperature fluctuation is gradual, it is marked as environmental noise and assigned a weight of 0.1. Finally, all thermal imaging pixels are traversed to generate an initial weight matrix consistent with the spatial resolution. This initial weight matrix retains the theoretical correlation strength between biological activity characteristics and thermal radiation while providing an adjustable reference parameter layer for subsequent dynamic interference compensation.

[0097] 204. Based on the external interference coefficient, dynamically adjust the initial weight value of each thermal radiation point, wherein for every preset unit increase in the external interference coefficient, the weight value of the corresponding thermal radiation point decreases according to a preset attenuation rate, thereby generating a calibrated dynamic weight distribution.

[0098] Dynamic weight distribution is the result of real-time correction of the initial weights by incorporating external interference coefficients. The higher the interference coefficient, the greater the weight attenuation of the corresponding thermal radiation point, thereby suppressing false biological activity signals caused by environmental noise.

[0099] In this embodiment, the initial weight values ​​are dynamically calibrated by fusing the external interference coefficients to suppress the interference of environmental noise on biological activity detection. First, the external interference coefficient matrix generated in step 202 is loaded, where the interference coefficient of each thermal radiation point quantifies the degree of influence of the external environment on its thermal radiation signal. Then, an exponential decay function model is defined to correlate the initial weights with the interference coefficients: the higher the interference coefficient, the greater the weight decay. For example, a preset decay rate parameter controls the slope of the weight decrease; when the interference coefficient of a thermal radiation point exceeds a threshold, its initial weight decreases rapidly according to an exponential curve. The system traverses all thermal imaging pixels, performing element-wise operations on the initial weight matrix and the interference coefficient matrix to generate a calibrated dynamic weight distribution. In this dynamic weight distribution, the weights of areas subject to strong external interference decay significantly, while stable areas retain higher weights, thereby enhancing the thermal radiation characteristics of the real biological activity signal in the spatial dimension and providing an anti-interference data foundation for subsequent comprehensive matching degree calculations.

[0100] 205. The biological activity intensity data in the synchronized time series pair is superimposed with the calibrated dynamic weight distribution to calculate the comprehensive matching degree between biological activity intensity and thermal radiation temperature at each time mark;

[0101] The overall matching degree is a joint evaluation index of biological activity intensity and thermal radiation calibration weights, calculated by weighted summation of the matching value at each time point. This index is used to quantify the spatial correlation between biological activity and thermal radiation signals within a specific time window.

[0102] In this embodiment, the normalized vibration energy value for the current time window is first extracted from the synchronized time series, and the dynamic weight matrix generated in step 204 is loaded simultaneously. For each pixel in the thermal imaging atlas, the temperature value in its normalized biological activity intensity data is multiplied by the dynamic weight at the corresponding location to obtain a weighted temperature response value. Subsequently, the weighted response values ​​of all pixels are integrated and summed in the spatial dimension, and a scalar multiplication operation is performed with the vibration energy value to generate the comprehensive matching degree between biological activity intensity and thermal radiation temperature at the corresponding time point. For example, when the weight of a certain region is 0.6, the normalized temperature value is 0.8, and the vibration energy is 0.9, its local contribution value is 0.6 × 0.8 = 0.48. After spatial integration, multiplying by the vibration energy yields a comprehensive matching degree of 0.48 × 0.9 = 0.432. The system iterates through all timestamps and repeats the above calculation to form a comprehensive matching degree curve in the time dimension. The peak period of the curve marks the key period in which biological activity and thermal radiation are highly correlated, providing a quantitative basis for subsequent pattern matching.

[0103] 206. Extract the change curve of the comprehensive matching degree within the continuous time marker, perform morphological matching between the thermal imaging atlas region corresponding to the change curve and the pre-stored rodent activity feature curve and insect activity feature curve, and retain the thermal imaging atlas region corresponding to the curve segment with a matching degree higher than the preset similarity as the target biological activity region.

[0104] Morphological matching is the process of calculating the similarity between a matching curve and a pre-stored biological activity feature template, used to filter target regions that match the behavioral patterns of rodents or insects. The feature templates are constructed based on the temporal patterns of typical biological activities, such as the intermittent peaks of rodent activity and the continuous fluctuations of insect activity.

[0105] In this embodiment, predefined matching curve templates for rodent and insect activities are first loaded from a biometric database. For example, rodent activity is characterized by intermittent spike waveforms, while insect activity exhibits continuous low-amplitude fluctuations. The system divides the comprehensive matching curve generated in step 205 into multiple time segments and uses a dynamic time warping algorithm to non-linearly align each segment, eliminating time axis scaling interference caused by differences in biological activity rhythms. The thermal imaging atlas region corresponding to the changing curve is morphologically matched with pre-stored rodent and insect activity feature curves, and the similarity score between each segment and the template curve is calculated. For example, the matching degree is quantified by correlation coefficient or dynamic bending path distance, and segments with scores exceeding a threshold are retained. Subsequently, the qualified time segments are mapped to the thermal imaging atlas, and the spatial distribution of high-weight thermal radiation points within the corresponding time period is extracted. Combined with a connected component analysis algorithm, target biological activity regions with continuously active characteristics are identified. For example, a region that matches the rodent template in a 10-15 minute time period and has a spatial weight consistently higher than 0.6 will be marked as a high-risk target biological activity region, and the detection result with confidence score and three-dimensional coordinates is finally output.

[0106] Here is a specific example:

[0107] In a container quarantine scenario at a port, the system detected a periodic impact signal of 3 times per second on the right front side of the container using vibration sensors. Simultaneously, a thermal imager captured the temperature in this area abnormally rising from a baseline of 25 degrees Celsius to 32 degrees Celsius. The time synchronization module aligned the vibration signal with the thermal imaging frames with microsecond-level precision, generating a synchronized dataset lasting ten minutes. External meteorological data showed significant wind speed fluctuations during the detection period. The interference model calculated the external interference coefficient for the right front thermal radiation point to be 0.5, significantly higher than the 0.1 to 0.3 coefficients for other areas. The weight mapping module assigned an initial weight of 0.75 to the right front side based on the excessive vibration energy characteristics, while the dynamic calibration module, applying an exponential decay model based on the interference coefficient, reduced its weight to 0.45, while maintaining the initial weight of 0.8 for low-interference areas. Matching calculations showed that the peak matching score on the right front side reached 0.88 within a 15-20 minute period. The dynamic time warping algorithm determined that the similarity between this curve and the rodent activity template exceeded the threshold. The system identified the right front side as a high-risk area and marked its three-dimensional coordinates and processing suggestions through an augmented reality interface to guide quarantine personnel in accurately implementing fumigation treatment.

[0108] This solution constructs a full-chain quarantine system from biological activity perception to decision output through multimodal data fusion and dynamic intelligent calibration technology. Time series synchronization technology eliminates time deviations in multi-source data, ensuring millisecond-level alignment of signals such as vibration and temperature. External interference modeling quantifies the impact of environmental factors on thermal radiation, effectively suppressing misjudgments caused by airflow disturbances and temperature-related noise. A dynamic weight adjustment mechanism combined with spatial calibration algorithms accurately restores the true biological activity characteristics of areas obscured by cargo. Intelligent morphological matching technology achieves automated classification and identification of disease-carrying organisms such as rodents and insects through comparison and screening of biological behavior feature databases. The entire system significantly improves detection accuracy and anti-interference capabilities in complex port scenarios, supports non-invasive rapid screening and visualized decision-making, and overcomes the technical bottlenecks of traditional manual sampling, such as low efficiency, high false negative rate, and poor environmental adaptability, providing a highly robust and reliable quarantine solution for smart ports.

[0109] In some embodiments, based on infrared sensing data, the thermal radiation distribution characteristics inside the cargo container are captured to generate a thermal imaging map containing the locations of biological activity hotspots and the density of cargo distribution, including:

[0110] 301. Multiple infrared sensor arrays are deployed inside the cargo container to collect temperature data at each monitoring point at a fixed sampling period to form an initial temperature distribution map;

[0111] An infrared sensor array refers to multiple infrared temperature sensor nodes deployed in a grid pattern inside a shipping container. Each node collects surface temperature data at its location at a fixed frequency. The fixed sampling period refers to the time interval at which all sensor nodes synchronously sample according to a unified clock. The initial temperature distribution map is a two-dimensional matrix of the temperature values ​​of all nodes within a single sampling period.

[0112] In this embodiment, high-precision infrared sensor nodes are first deployed on the top, side walls, and bottom of the cargo container to form a detection grid with specific accuracy. Each node integrates an infrared array sensor and transmits temperature data to the central processing unit via a bus. The system synchronously collects the temperature values ​​of all nodes within a set sampling period to generate an initial temperature distribution map. For example, multiple sets of sensor nodes can be deployed in a standard container to form a two-dimensional temperature matrix to cover the internal space.

[0113] 302. Perform dynamic analysis on the initial temperature distribution map, calculate the temperature change rate of each monitoring point within the continuous sampling period, mark the area where the temperature change rate exceeds the set threshold as the dynamic heat source area, and establish a three-dimensional spatial coordinate system in combination with the internal structural parameters of the container, and map the dynamic heat source area into the three-dimensional spatial model.

[0114] The dynamic heat source zone refers to the area where the rate of temperature change exceeds a set threshold within multiple consecutive sampling periods. The three-dimensional spatial coordinate system is a coordinate system established based on the physical dimensions of the container, used to map two-dimensional temperature data into a three-dimensional heat source distribution model.

[0115] In this embodiment, the initial temperature distribution map is subjected to time-series filtering to calculate the temperature change rate of each sensor node within a continuous sampling period. Nodes with temperature change rates exceeding a threshold are marked as dynamic heat source candidate points. A three-dimensional spatial coordinate system is established in conjunction with the internal structural parameters of the container, and a spatial interpolation algorithm is used to map the two-dimensional temperature data into a three-dimensional spatial model. For example, sidewall sensor data is interpolated to generate a surface temperature layer of the internal cargo stacking area, achieving three-dimensional visualization of the temperature data.

[0116] 303. Based on the cargo loading information, mark the cargo stacking area in the three-dimensional spatial model to exclude static high-temperature areas caused by the cargo's own heat generation; perform cluster analysis on the remaining dynamic heat source areas, and merge spatially adjacent areas with similar temperature change characteristics into candidate biological activity areas.

[0117] The cargo stacking area is the physical boundary of the cargo marked according to the packing information. The static high temperature area refers to the area of ​​sustained high temperature caused by the cargo's own heat generation. The candidate biological activity area is a spatially connected region with similar temperature change characteristics merged by a clustering algorithm.

[0118] In this embodiment, cargo loading information is imported, and temperature data of the corresponding area of ​​the cargo is masked in the three-dimensional spatial model to eliminate interference from static high-temperature areas caused by the cargo's own heat generation. The remaining dynamic heat source areas are spatially grouped using a density clustering algorithm, with a neighborhood radius and a minimum point threshold set. Areas with spatially connected cargo density and consistent temperature change characteristics are merged into candidate biological activity areas. For example, if an area experiences temperature rise pulses in multiple sampling periods and is spatially close, it is merged into a single candidate biological activity area.

[0119] 304. Based on the stacking height and distribution of goods, calculate the infrared occlusion coefficient of each candidate biological activity zone and correct the temperature measurement error caused by the occlusion of goods.

[0120] The infrared obstruction coefficient is a parameter that quantifies the impact of cargo stacking on the thermal radiation propagation path. It is calculated based on the cargo height and the infrared transmittance of the material to determine the degree of thermal signal attenuation.

[0121] In this embodiment, a stacking structure model is first established based on the stacking height and distribution of goods, and the types of goods, stacking height, and gap distribution in each area are labeled. For example, in a certain area, two layers of wooden boxes with a thickness of 0.5 meters and plastic partitions with a thickness of 0.1 meters are stacked alternately. Then, virtual rays are emitted from the candidate biological activity area to the infrared sensor node, and the sequence of goods layers and the length of the penetration path through which the rays pass are analyzed. For example, the rays pass through a total thickness of 1 meter of wooden layers and 0.1 meters of plastic layers. The infrared transmittance of each layer is obtained by combining the material database. The transmittance of wooden materials is 0.6, and the transmittance of plastic materials is 0.8. The layer attenuation coefficient is calculated based on the penetration thickness. The attenuation coefficient of the wooden layer is calculated to be 0.6 by power 1, and the attenuation coefficient of the plastic layer is calculated to be approximately 0.977 by power 0.1 of 0.8. Multiplying the attenuation coefficients of each layer consecutively yields a total infrared blocking coefficient of 0.6 multiplied by 0.977, approximately 0.586. Based on this infrared blocking coefficient, the original temperature value of 30 degrees Celsius detected by the sensor is corrected to approximately 51.2 degrees Celsius (30 divided by 0.586), restoring the true thermal radiation intensity. By quantifying the combined blocking effect of cargo height, density, and material, the problem of weakened biological activity signals caused by complex stacking is solved, and temperature measurement errors caused by cargo blocking are corrected, ensuring that subsequent analysis is based on accurate temperature data.

[0122] 305. The regions in the corrected candidate biological activity areas that meet the preset biological characteristic parameters are identified as the final biological activity hotspots; based on the spatial relationship between cargo distribution density and biological activity hotspots, a thermal imaging map with layered display function is generated.

[0123] The preset biometric parameters include the temperature rise amplitude and duration. Layered display refers to overlaying multiple layers of information such as cargo density, biological hotspots, and temperature change curves on the thermal imaging map.

[0124] In this embodiment, a preset biometric parameter database is established. Temperature change curves are extracted from the corrected candidate biological activity areas. These curves are then compared with the biometric parameter database using a time-series matching algorithm to select eligible areas as the final biological activity hotspots. A thermal imaging atlas is generated based on the spatial relationship between cargo distribution density and biological activity hotspots. In the final generated thermal imaging atlas, cargo density is represented by color levels, biological hotspots are highlighted with icons, and interactive viewing of historical temperature change data is supported, enabling multi-dimensional information fusion and display.

[0125] Here is a specific example:

[0126] In an inbound container quarantine scenario at an international port, the system detected abnormal temperature fluctuations inside a container carrying electronic equipment and food. An infrared sensor array deployed on the top of the container collected data once per second. The initial temperature distribution map showed that the temperature in the right rear area rose from 25 degrees Celsius to 28 degrees Celsius over three consecutive sampling periods, a temperature change rate of 0.5 degrees Celsius per minute, triggering a dynamic heat source marker. The 3D spatial mapping module located this area within the gaps between stacked goods, eliminating static high-temperature interference caused by heat from electronic equipment. Using a density clustering algorithm, the system identified two candidate biological activity areas. Area A was automatically filtered out due to its proximity to a metal container, while area B was retained due to its sustained temperature rise and spatial distribution characteristics. Ray tracing calculations showed that the heat radiation path of area B passed through a wooden container, with an obstruction coefficient of 0.4. After correction, the temperature increased from 29 degrees Celsius to 34 degrees Celsius. The temperature change curve analysis module compared the data with the biometric database and confirmed that it matched the periodic pulse characteristics of rodent activity (a single temperature rise of 2 degrees Celsius lasting for 3 minutes). Finally, the hot spot was highlighted in red in the thermal imaging map, and the cargo density gradient layer and historical temperature change trend curve were superimposed to guide quarantine personnel to accurately implement subsequent processing.

[0127] This solution systematically improves the accuracy and reliability of vector-borne disease detection within containers through multimodal sensor fusion and intelligent analysis technologies. Based on dynamic heat source capture and 3D spatial modeling using an infrared sensor array, it overcomes the limitations of traditional 2D thermal imaging, enabling three-dimensional temperature field analysis in cargo stacking scenarios. Through ray tracing technology and occlusion effect compensation algorithms, it effectively restores the true thermal radiation signals of areas obscured by cargo, overcoming the problem of missed detection of biological activity caused by physical obstruction. Combining a temperature change feature library and dynamic clustering analysis, it intelligently distinguishes between environmental noise, cargo self-heating, and real biological activity signals, significantly reducing the false positive rate. Layered visualization maps integrate cargo density, biological hotspots, and historical temperature change curves, providing multi-dimensional data support for quarantine decisions. The entire system supports non-invasive real-time monitoring, achieving highly robust and timely vector-borne disease quarantine in challenging scenarios such as complex cargo distribution and dynamic environmental interference, driving the upgrade of port quarantine processes towards automation and intelligence.

[0128] In some embodiments, the boundary position of the target biological activity area is spatially calibrated based on the occlusion compensation coefficient of each monitoring zone to generate a biological activity distribution map, including:

[0129] 401. Obtain the occlusion compensation coefficient corresponding to each monitoring zone, wherein the occlusion compensation coefficient reflects the degree of occlusion of infrared detection by goods in the zone;

[0130] The occlusion compensation coefficient is a parameter that quantifies the degree to which cargo within a monitoring zone obstructs infrared detection signals. Its value typically ranges from 0 to 2. A coefficient of 1 indicates no obstruction; a coefficient less than 1 indicates that the infrared signal is attenuated due to obstruction (e.g., 0.6 indicates a 40% signal strength loss); a coefficient greater than 1 indicates that the signal is enhanced due to the thermal focusing effect caused by the cargo structure (e.g., 1.3 indicates a 30% signal increase).

[0131] In this embodiment, monitoring zone data from the thermal imaging atlas is retrieved from the front-end data engine. The occlusion compensation coefficient for each zone is stored in the spatial database. The occlusion compensation coefficient matrix corresponding to each monitoring zone is obtained in batches using structured query or key-value reading. The occlusion compensation coefficient reflects the degree to which cargo within the zone obstructs infrared detection. For example, the coefficient for zone Z3 is 0.7, indicating that cargo within it causes a 30% attenuation of the infrared signal. The occlusion compensation coefficient will serve as the core parameter for subsequent boundary correction and will be transmitted to the spatial calibration module via the data interface.

[0132] 402. Determine the monitoring zone where each target biological activity area is located, and calculate the boundary correction amount of the area based on the shading compensation coefficient of the monitoring zone; adjust the boundary position of the monitoring zone by expanding or shrinking according to the boundary correction amount;

[0133] The boundary correction amount is the displacement distance of the region edge calculated based on the occlusion compensation coefficient. Positive values ​​expand the boundary to compensate for signal weakening areas caused by occlusion; negative values ​​contract the boundary to eliminate misjudged expansion areas caused by signal focusing. The correction direction is determined by a preset threshold: if the compensation coefficient is less than 0.8, the boundary expands; if it is greater than 1.2, the boundary contracts.

[0134] In this embodiment, the monitoring zone to which the target biological area belongs is first determined, and the occlusion compensation coefficient of the corresponding zone is retrieved. If the occlusion compensation coefficient of the monitoring zone is 0.7, which is lower than the threshold of 0.8, the boundary correction amount is calculated according to the expansion formula: the expansion amount is equal to the expansion gain factor multiplied by the complement of the occlusion compensation coefficient and then multiplied by the original radius of the area. For example, if the expansion gain factor is preset to 2.5 and the original radius of the area is 1.5 meters, then the boundary correction amount is the difference between 2.5 and 1 minus 0.7, multiplied by 1.5 meters, resulting in 1.125 meters. According to the boundary correction amount, the boundary position of the monitoring zone is adjusted by shifting the original polygon vertices along the normal direction using the geometric transformation module, generating a new expanded boundary, and the difference between the boundary before and after adjustment is compared and displayed in the visualization interface.

[0135] 403. Detect the boundary overlap between adjacent areas of each target biological activity area after adjustment, and perform boundary fusion processing on adjacent areas with overlap;

[0136] Boundary fusion processing is an optimization operation for overlapping areas after adjustment of adjacent regions. It fills tiny gaps through morphological closing operations and applies a region growing algorithm to integrate overlapping areas into a single connected region, avoiding fragmented results.

[0137] In this embodiment, spatial overlay analysis is first performed on the expanded or shrunken target biological activity region. A ray casting algorithm is used to detect boundary overlap between adjacent regions; for example, the corrected boundary of region A encroaches on 15% of the original area of ​​region B. Then, morphological closing operations are used to fill small gaps, and dilation and erosion operations are used to smooth irregular edges. For large overlapping regions, a region growing algorithm is applied to merge overlapping pixels into connected components. Based on Delaunay triangulation, polygon vertices are reconstructed to generate the minimum hull boundary covering the overlapping region. For example, after merging regions A and B, the system retains the core activity feature points of both, calculates a new convex hull boundary, and removes redundant vertices, ultimately forming a continuous and smooth composite region. This avoids fragmentation or duplicate counting problems caused by compensation adjustments, providing a spatially consistent vector data foundation for subsequent thermal imaging integration.

[0138] 404. Integrate the spatial location information of each target biological activity area after processing with the temperature distribution information in the thermal imaging atlas to generate a biological activity distribution map containing the distribution characteristics of biological activity intensity; in the biological activity distribution map, use different marking methods to distinguish and display the boundary of the area adjusted by occlusion compensation and the boundary of the original detection area.

[0139] The biological activity distribution map is a composite map that integrates spatial calibration results and thermal imaging data. It uses dashed lines to mark the original boundaries and solid lines to mark the corrected boundaries. The color gradient represents the activity intensity, with red representing high intensity and blue representing low intensity. It supports layer overlay and transparency adjustment to enhance readability.

[0140] In this embodiment, the calibrated target area vector boundary is first geospatially aligned with the thermal imaging temperature raster data to ensure accurate matching between the corrected polygon coordinates and the temperature pixel matrix. Next, temperature distribution data within each target area is extracted, and based on a preset biological activity intensity grading rule, temperature values ​​are mapped to a color gradient visualization layer: for example, areas with temperatures between 35 and 40 degrees Celsius are marked yellow for low-intensity activity, 40 to 45 degrees Celsius are marked orange for medium intensity, and above 45 degrees Celsius are marked red for high intensity. Simultaneously, the original detection boundary and the calibrated boundary are overlaid. The original boundary is drawn with a semi-transparent gray dashed line, while the calibrated boundary is highlighted with a solid red line for comparison. For example, after expansion and correction, the solid red boundary of a certain area completely encloses the core high-temperature zone with a temperature of 46 degrees Celsius, while the surrounding areas with temperatures decreasing to 38 degrees Celsius show a gradient from orange to yellow, visually demonstrating the spatial decay characteristics of biological activity intensity. The final biological activity distribution map is generated, which integrates a temperature and thermal layer, a boundary calibration layer, and an intensity labeling layer. It supports interactive click-based query of area details. For example, clicking on the expanded boundary can display the original temperature value, calibration coefficient, and historical activity curve of the area, realizing multi-dimensional data linkage analysis and visual decision-making.

[0141] Here is a specific example:

[0142] In an international port's inbound container quarantine scenario, the system performs full-process inspection of mixed cargo containers carrying textiles and electronic equipment. The system first obtains an occlusion compensation coefficient of 0.6 for the right front monitoring zone. This coefficient is generated due to a 40% attenuation of the infrared signal caused by the stacking of multiple layers of high-density textiles. For a candidate bioactivity area within the zone with an original diameter of 1.5 meters, the system calculates the boundary correction based on an expansion gain factor of 2.5, resulting in an expansion of 0.75 meters, expanding the boundary outward to form a new area with a diameter of 3 meters. Subsequently, it detects a 25% overlap between the expanded area and the adjacent area on the left, triggering a boundary fusion algorithm to merge overlapping vertices and reconstruct a smooth, continuous minimum bounding polygon, eliminating jagged edges. In the final generated bioactivity distribution map, the original gray dashed boundary expands to a red solid boundary, completely enclosing the core high-temperature zone of 45 degrees Celsius within the thermal imaging spectrum. The surrounding area, with temperatures decreasing to 38 degrees Celsius, is displayed with an orange gradient. Users can click on the boundary to view the compensation coefficient of 0.6, the expansion of 0.75 meters, and historical temperature fluctuation curves in real time, enabling multi-dimensional interactive analysis of calibration results and temperature data.

[0143] This solution effectively addresses the issues of missed detection and misjudgment of biological activities caused by complex cargo obstruction through dynamic spatial calibration and intelligent data fusion technologies. A precise obstruction compensation model recreates accurate thermal radiation signals, while boundary fusion algorithms ensure regional integrity and analytical consistency. Multi-layered visualization maps support data traceability and interactive decision-making. The system achieves sub-meter-level positioning accuracy in non-invasive detection, significantly reducing the false detection rate and providing a highly efficient and reliable intelligent solution for port quarantine, driving the industry towards automation and high precision.

[0144] In some embodiments, based on the spatial location information and density threshold of the target biological activity area in the biological activity distribution map, a visual assessment result of the quarantine status of vector organisms inside the cargo container is output, including:

[0145] 501. Extract the center coordinates of each target biological activity area from the biological activity distribution map, and calculate the average biological activity signal intensity of each area;

[0146] The average biological activity signal intensity refers to the temperature-weighted average of all thermal radiation points within the target area, with the weights determined by the correlation between thermal radiation intensity and biological activity characteristics. The center coordinates are spatial positioning data obtained by calculating the geometric center or peak thermal radiation point of the target area.

[0147] In this embodiment, the center coordinates and vector boundary data of each target biological activity area are extracted from the biological activity distribution map, and all thermal radiation point coordinates and their corresponding temperature values ​​are traversed. A weighted average calculation is performed on the biological activity signal intensity of each area, and the weighting coefficient is dynamically adjusted according to the matching degree between the temperature value and the biological activity feature database. For example, the temperature weighting coefficient for rodent activity hotspot areas is set to 0.8, and for environmental noise areas it is set to 0.2. Finally, the center coordinates and average biological activity signal intensity of each area are output. For example, the center coordinates of a certain area are 3.2 meters on the X-axis, 1.5 meters on the Y-axis, and 2.1 meters on the Z-axis, and the average biological activity signal intensity is 42.5 degrees Celsius.

[0148] 502. Map the center coordinates of each target biological activity area to the three-dimensional spatial model of the cargo container, and adjust the projection position of the center coordinates in the vertical direction according to the shielding effect parameter of infrared thermal radiation on the cargo distribution density.

[0149] The vertical projection position adjustment is based on the influence of cargo stacking height and density on the thermal radiation propagation path, correcting the Z-axis value of the target area's center coordinates to reflect the true spatial location of biological activity. The shading effect parameters are calculated from the cargo material, the number of stacking layers, and an infrared attenuation model.

[0150] In this embodiment, cargo distribution density data is loaded from a 3D spatial model database of a cargo container. The center coordinates of each target biological activity area are mapped onto the 3D spatial model of the cargo container. A virtual ray is projected downwards vertically from these center coordinates, and the total shading effect parameter of the ray passing through each layer of cargo is calculated. The Z-axis coordinate is adjusted according to the shading effect parameter. For example, when the shading effect parameter is 0.6, it indicates that the signal comes from a deeper region, and the Z-value is corrected downwards by 0.4 meters. The original center coordinate Z-axis of 2.1 meters is corrected to 2.5 meters, corresponding to the middle layer of the cargo stack.

[0151] 503. Compare the average biological activity signal intensity of each target biological activity area with the preset density threshold. If the threshold is exceeded, the area is marked as a high-risk area; otherwise, it is marked as a low-risk area.

[0152] The preset density threshold is a risk classification standard set based on historical quarantine data and biological activity patterns. For example, the high-risk threshold is set at 40 degrees Celsius, and the low-risk threshold is set at 35 degrees Celsius.

[0153] In this embodiment, the average biological activity signal intensity of each target biological activity area is compared with a preset density threshold. If the average temperature of a target biological activity area is 42 degrees Celsius and the threshold is set to 40 degrees Celsius, it is marked as high-risk; otherwise, it is marked as a low-risk area, i.e., 37 degrees Celsius is marked as low-risk. The system also verifies the duration of continuous activity in the area. For example, a high-risk area must meet the requirement that the temperature exceeds the threshold for more than 5 minutes to avoid instantaneous noise interference.

[0154] 504. In the three-dimensional spatial model, high-risk areas and low-risk areas are marked with different colors, and a color block distribution map covering the internal structure of the cargo container is generated according to the center coordinate projection position of each area.

[0155] The color block distribution map is a visual representation that maps risk levels to a three-dimensional spatial model using color coding technology. High-risk areas are marked with red cubes, and low-risk areas are marked with yellow cubes. The size of the color blocks reflects the spatial extent of the area.

[0156] In this embodiment, a color block instance is created for each target biological activity area in the three-dimensional spatial model. A preset color coding table is invoked according to the risk level, assigning red to high-risk areas and yellow to low-risk areas. The side length of the color block is determined by the size of the largest bounding rectangle of the target biological activity area; for example, an area with a diameter of 1.5 meters corresponds to a cube with a side length of 1.5 meters. The corrected center coordinates are used as anchor points for the color blocks and superimposed onto the container's three-dimensional model to generate a layered color block distribution map.

[0157] 505. Overlay the color block distribution map with the cargo type data of the cargo container, and output a comprehensive assessment result including the risk level and spatial location of disease vectors in the visualization interface.

[0158] The comprehensive evaluation results are presented in an interactive visualization interface that integrates color block distribution and cargo type data. It allows users to click on color blocks to view detailed parameters and adjust the transparency to perform multi-layer comparative analysis.

[0159] In this embodiment, the cargo loading list of the cargo container is imported into the visualization platform, and the cargo types are marked in the 3D spatial model and displayed with semi-transparent gray wireframes. A click event is added to each colored block, triggering a pop-up window to display the comprehensive assessment results of the corresponding target biological activity area parameters, including the risk level and spatial location of disease vectors. For example, clicking a red colored block allows viewing the historical temperature curve, the corrected Z-axis coordinates, and the associated cargo type. A slider is provided to adjust the transparency of the cargo wireframes and colored blocks, supporting users to focus on analyzing the spatial relationship between high-risk areas and cargo stacking.

[0160] Here is a specific example:

[0161] In an international port's inbound container quarantine scenario, the system performs a full-process assessment of mixed cargo containers carrying food and timber. The system first extracts two target areas from the biological activity distribution map: Area A has center coordinates of 2.1 meters X-axis, 1.3 meters Y-axis, and 1.8 meters Z-axis, with an average signal strength of 43 degrees Celsius; Area B has center coordinates of 4.5 meters X-axis, 0.9 meters Y-axis, and 2.2 meters Z-axis, with an average signal strength of 37 degrees Celsius. For Area A, the system detects an obstruction coefficient of 0.5 due to the stacked timber on top. The vertical projection adjustment module lowers the Z-axis coordinate by 0.6 meters to 2.4 meters, accurately locating it in the middle layer of the cargo. Area B, being unobstructed, retains its original coordinates. The risk assessment module compares the risk level to thresholds; Area A, exceeding 40 degrees Celsius, is marked as high-risk, while Area B is marked as low-risk. The 3D visualization engine generates a red cube with a 1.2-meter margin covering Area A and a yellow cube with a 0.8-meter margin covering Area B. After overlaying the cargo distribution map, the system displays the food containers adjacent to Area A. Users can click on the red block to view detailed parameters: if the temperature curve shows that the activity exceeds the threshold for 5 consecutive minutes, the system recommends prioritizing fumigation treatment. At the same time, the system will link cargo data to indicate that the surrounding area is susceptible to contamination, thus realizing intelligent push of risk assessment and disposal plan.

[0162] This solution optimizes the entire process of vector-borne disease quarantine assessment within cargo containers through multi-dimensional data fusion and intelligent analysis technologies. Precise biological activity signal extraction and spatial coordinate correction techniques effectively reconstruct the true location of biological activity in areas obscured by cargo. A dynamic threshold determination mechanism combined with a historical feature database significantly improves the accuracy and anti-interference capability of risk classification. A 3D visualization engine and interactive layer management intuitively present the spatial relationship between risk areas and cargo stacking, supporting multi-dimensional data traceability and real-time decision-making. The system achieves non-invasive, rapid detection in complex cargo scenarios, greatly improving quarantine efficiency and reliability, and providing efficient and intuitive technical support for intelligent quarantine at ports.

[0163] In some embodiments, the external interference coefficient of each thermal radiation point in the thermal imaging spectrum is calculated based on the temperature change rate and airflow disturbance frequency in the environmental interference parameters outside the cargo container, including:

[0164] 601. Based on the temperature change rate in the environmental interference parameters outside the cargo container, calculate the difference between the temperature change rate of the thermal radiation point and the external temperature change rate for each thermal radiation point in the thermal imaging spectrum.

[0165] The temperature change rate difference refers to the difference between the rate of temperature change of a single thermal radiation point in a thermal imaging map and the rate of temperature change of the external environment of the container. The external temperature change rate is obtained by temperature sensors installed on the outer wall of the container at a fixed frequency, reflecting the influence of ambient temperature fluctuations on the conduction of thermal radiation inside the container.

[0166] In this embodiment, firstly, external temperature data from continuous time-series environmental disturbance parameters outside the cargo container is acquired from external sensors, and the rate of temperature change per unit time is calculated. For example, if the external temperature rises from 28 degrees Celsius to 30 degrees Celsius within 5 minutes, the rate of external temperature change is 0.4 degrees Celsius per minute. Simultaneously, the temperature time series of each thermal radiation point in the thermal imaging spectrum is extracted, and the rate of temperature change is calculated using a sliding window. For example, if a thermal radiation point rises from 32 degrees Celsius to 33.5 degrees Celsius within the same 5 minutes, the internal rate is 0.3 degrees Celsius per minute. Finally, the difference between the rate of temperature change of the thermal radiation point and the rate of temperature change of the external temperature is calculated. Subtracting 0.4 degrees Celsius per minute from 0.3 degrees Celsius per minute yields a difference of -0.1 degrees Celsius per minute, indicating that the thermal radiation point experiences a reverse conduction effect due to external temperature changes.

[0167] 602. Calculate the degree of influence of airflow disturbance frequency on the temperature of the thermal radiation point based on the environmental interference parameters outside the cargo container.

[0168] The degree of influence of airflow disturbance frequency quantifies the interference intensity of periodic changes in external airflow on the temperature stability of thermal radiation points. The airflow disturbance frequency is collected by an anemometer, the dominant frequency component is extracted by Fourier transform, and then converted into a temperature fluctuation coefficient according to an experimentally calibrated model.

[0169] In this embodiment, a Fast Fourier Transform (FFT) is first performed on the airflow disturbance frequency in the wind speed sensor data to identify the main frequency components. For example, significant frequency peaks at 0.5 Hz and 1.2 Hz are detected. Then, a relationship table between airflow disturbance frequency and temperature disturbance is established based on wind tunnel experimental data; for example, a temperature fluctuation coefficient of 0.2 corresponds to a frequency of 0.5 Hz, and a coefficient of 0.5 corresponds to 1.2 Hz. Finally, for each thermal radiation point, the total influence on the temperature of the thermal radiation point is calculated based on a weighted average of its spatial location and distance from the container vent. For example, a thermal radiation point near the vent is affected by the 1.2 Hz frequency; with a distance weighting factor of 0.8, the influence is 0.5 multiplied by 0.8, resulting in an influence of 0.4.

[0170] 603. Add the difference and the degree of influence according to the preset weight to obtain the external interference coefficient of each thermal radiation point; output the external interference coefficient of each thermal radiation point.

[0171] The preset weights are the proportions of the contribution of the temperature change rate difference and the degree of influence of airflow disturbance to the external interference coefficient, and are determined based on historical data training. For example, the weight of temperature difference is set to 0.6, and the weight of airflow influence is set to 0.4, reflecting that temperature change conduction is the main source of interference.

[0172] In this embodiment, the absolute value of the difference in step 601 and the degree of influence in step 602 are normalized and linearly superimposed according to their weights. For example, if the normalized value of the temperature difference at a thermal radiation point is 0.3 and the normalized value of the airflow influence is 0.6, then the external interference coefficient is 0.3 multiplied by 0.6 plus 0.6 multiplied by 0.4, resulting in 0.42. The system iterates through all thermal radiation points to obtain the external interference coefficient for each point, generating an external interference coefficient matrix for each thermal radiation point for subsequent thermal imaging calibration.

[0173] Here is a specific example:

[0174] In a container inspection scenario at a coastal port, the system calculates interference coefficients for external environmental temperature fluctuations and sea breeze disturbances. An external temperature sensor detects a temperature increase from 25°C to 28°C within 10 minutes, a rate of change of 0.3°C per minute. Simultaneously, the thermal imaging shows a temperature rate of 0.5°C per minute for a heat radiation point near a container door, resulting in a difference of 0.2°C per minute. An anemometer detects a dominant airflow frequency of 1 Hz, and based on a wind tunnel experimental model, its temperature fluctuation coefficient is calibrated to 0.4. Combining this with a positional weighting factor of 0.8 for the heat radiation point's proximity to a vent, the airflow influence level is calculated to be 0.32. The system then superimposes the normalized temperature difference value (0.5) and the normalized airflow influence level value (0.8) with preset weights of 0.6 and 0.4, generating an external interference coefficient of 0.62 for the heat radiation point, marking it as a high-interference area. After traversing all thermal radiation points, an external interference coefficient matrix is ​​output to guide the subsequent thermal imaging calibration module to dynamically reduce the weight of high-interference areas, ensuring that biological activity detection is not affected by environmental noise.

[0175] This solution significantly improves detection robustness in complex scenarios by dynamically quantifying the impact of the external environment on thermal imaging data. A two-factor fusion model of temperature conduction rate and airflow disturbance frequency accurately identifies and quantifies the interference intensity of external interference sources on the thermal radiation signal inside the container, effectively distinguishing between environmental noise and real biological activity signals. An adaptive calibration mechanism based on weight allocation generates an external interference coefficient matrix in real time, providing a reliable basis for dynamic noise reduction in thermal imaging and ensuring that detection results are unaffected by environmental fluctuations. In high-interference scenarios such as ports and logistics, the system significantly reduces the false alarm rate and improves the target area positioning accuracy, providing highly adaptable and reliable technical support for biological activity detection.

[0176] In some embodiments, based on the spatial relationship between cargo distribution density and biological activity hotspots, a thermal imaging atlas with layered display capabilities is generated, including:

[0177] 701. Based on the density values ​​of the cargo stacking area, the three-dimensional spatial model is divided into multiple vertically distributed density levels, each level corresponding to a preset density range, generating a level division plane parallel to the side wall of the cargo container, the division plane isolating cargo stacking areas of different density levels into independent display layers;

[0178] Density tiers are three-dimensional vertical layers defined by the numerical density of cargo stacking. For example, low-density tiers correspond to densities of 0 to 200 kg / m³, medium-density tiers to 200 to 500 kg / m³, and high-density tiers to above 500 kg / m³. The tier division plane is a virtual plane parallel to the container's sidewall, used to isolate cargo areas of different density tiers, forming independent, visually distinct display layers.

[0179] In this embodiment, firstly, 3D point cloud data of the cargo stacking area is loaded, and the stacking density value of each area is calculated. The K-means clustering algorithm is used to divide the density values ​​into preset density intervals, such as low, medium, and high levels. Then, a segmentation plane is generated vertically in the 3D spatial model, and the height of the plane is determined based on the density level boundary values. For example, the height range for the low-density layer is 0 to 1.5 meters, for the medium-density layer 1.5 to 3 meters, and for the high-density layer above 3 meters. The segmentation plane is rendered in semi-transparent gray, isolating cargo stacking areas of different density levels as independent display layers, achieving visual isolation between different density levels.

[0180] 702. Automatically match the coordinates of the biological activity area with the boundary range of each density level to determine the level to which it belongs; dynamically render hotspot markers on the level segmentation plane to generate a heat source distribution map bound to each density level;

[0181] The heat source distribution map is a marker layer generated by matching the coordinates of biological activity hotspots with the boundary range of density levels. The hotspot markers are represented by dynamic icons, such as flashing red dots, and each icon is bound to the display plane of its respective density level.

[0182] In this embodiment, the center coordinates of the biological activity area are first compared with the height boundary range of each density level to determine the corresponding level. For example, a hotspot with a Z-axis coordinate of 2.4 meters belongs to the medium-density layer. Then, on the corresponding level segmentation plane, a dynamic icon is generated using the hotspot's coordinate projection position as the anchor point. A flashing red dot instance is created using the Three.js engine, and its position is mapped to two-dimensional coordinates on the segmentation plane. Finally, the visibility of the marker icon is associated with the density level display status to generate a heat source distribution map bound to each density level, ensuring that the corresponding hotspot is displayed or hidden synchronously when switching levels.

[0183] 703. Overlay the heat source distribution map with the cargo stacking model, so that the hot spot markers of each level coincide with the cargo area location, and generate an initial thermal imaging map based on the overlay result;

[0184] The initial thermal imaging map is a composite view of the heat source distribution map and the cargo stacking model. Through transparency adjustment and layer blending technology, the hot spot markers are made to accurately coincide with the cargo area location.

[0185] In this embodiment, the WebGL rendering engine is invoked to perform data overlay: First, the two-dimensional marker coordinates of the heat source distribution map are converted into three-dimensional model space coordinates to ensure that the red dots are aligned with the stacked goods area. Then, the rendering transparency of the stacked goods model is set to 50%, and the heat source marker layer is set to opaque, achieving synchronous visibility of the goods outline and hotspots. Finally, a Gaussian blur effect is added to smooth the edges of the goods, and an initial thermal imaging map is generated based on the overlay result to enhance the visual focus of the hotspot markers.

[0186] 704. Configure the hierarchical visibility of the initial thermal imaging map according to the user's operation instructions, update the overlay display status according to the configuration result, and output the thermal imaging map that supports hierarchical interaction.

[0187] Layer visibility configuration dynamically controls the display status of each density layer and its hotspots based on user actions, supporting on-demand focus on specific layer analysis. For example, layer transparency can be adjusted by checking checkboxes or dragging sliders.

[0188] In this embodiment, interactive controls are integrated into the front-end interface: a hierarchical selection panel and a transparency slider are created, for example, the low-density layer checkbox is checked by default, and the slider ranges from 0 to 100%. User operation events are listened for, and the hierarchical visibility of the initial thermal imaging map is configured. For example, when the high-density layer is unchecked, the corresponding segmentation plane and associated hotspots are hidden. The overlay display state is updated according to the configuration state, for example, adjusting the transparency of the low-density layer to 70% to make it semi-transparent to highlight the hotspots in the medium-density layer, and outputting the thermal imaging map that supports hierarchical interaction.

[0189] Here is a specific example:

[0190] In a container inspection scenario involving food and timber, the system performs full-process layered visualization: density calculations show a low-density layer (0-1.5 meters) consisting of food packaging boxes, while a high-density layer (above 1.5 meters) consists of stacked timber. The biological activity hotspot coordinates at 1.2 meters on the Z-axis are matched to the low-density layer, rendering flashing red dots on the segmentation plane. The initial overlay map shows the red dots located between the food boxes, with a clear outline of the cargo thanks to the transparency blending effect. The user unchecks the high-density layer, hides the timber stack area, focuses on analyzing the hotspot distribution in the low-density layer, adjusts the transparency to 60%, confirming that the hotspots do not overlap with the food box locations, thus eliminating interference from the cargo's self-heating.

[0191] This solution significantly enhances the spatial correlation analysis capabilities of thermal imaging maps in complex cargo scenarios through layered visualization and dynamic interaction technologies. Based on a density-level vertical division and hotspot binding mechanism, it achieves precise spatial mapping between biological activity areas and cargo stacking structures. Intelligent matching of layered segmentation planes and heat source markers effectively distinguishes the distribution of biological activity in high-density cargo-occupied areas from low-density susceptible areas. Interactive visibility configuration and transparency adjustment functions allow users to focus on key layers as needed, reducing visual interference from multi-layer overlays. Real-time rendering optimization and hybrid display technology enhance data readability, intuitively presenting the spatial relationship between hotspots and cargo locations. It provides highly adaptable and high-precision visualization decision-making tools for port and logistics scenarios, helping to quickly identify biological activity risk areas and formulate targeted quarantine strategies, promoting the efficient implementation of intelligent quarantine management.

[0192] Figure 2This application provides a schematic diagram of the structure of an IoT-based vector-borne disease quarantine system at ports of entry, as shown in the embodiments. Figure 2 As shown, the system includes:

[0193] The acquisition module 21 is used to generate an environmental monitoring dataset that characterizes the dynamic changes of the environment inside the cargo container by synchronously acquiring target environmental parameters and biological activity sensor data inside the cargo container.

[0194] The first generation module 22 is used to capture the thermal radiation distribution characteristics inside the cargo container based on infrared sensing data, and generate a thermal imaging map containing the location of biological activity hotspots and the cargo distribution density.

[0195] The filtering module 23 is used to correlate the environmental monitoring dataset with the thermal imaging map over time, and combine the environmental interference parameters outside the cargo container to dynamically calibrate the correlation model between the biological activity sensing data and the thermal radiation distribution characteristics, and filter out target biological activity areas that match the activity patterns of rodents or insects.

[0196] The second generation module 24 is used to divide the interior of the cargo container into multiple monitoring zones according to the occlusion effect parameters of the cargo distribution density on infrared thermal radiation in the thermal imaging spectrum, and to spatially calibrate the boundary position of the target biological activity area based on the occlusion compensation coefficient of each monitoring zone to generate a biological activity distribution map.

[0197] Output module 25 is used to output a visual assessment result of the quarantine status of disease vectors in the cargo container based on the spatial location information and density threshold of the target biological activity area in the biological activity distribution map.

[0198] Figure 2 The aforementioned IoT-based vector-borne disease quarantine system at entry and exit ports can perform... Figure 1 The implementation principle and technical effects of the IoT-based vector-borne disease quarantine method at entry-exit ports described in the illustrated embodiment will not be repeated here. The specific operation methods of each module and unit in the IoT-based vector-borne disease quarantine system at entry-exit ports described in the above embodiments have been detailed in the relevant method embodiments and will not be elaborated upon here.

[0199] In one possible design, Figure 2 The illustrated embodiment of an IoT-based vector-borne disease quarantine system at an entry-exit port can be implemented as a computing device, such as... Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0200] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are invoked and executed by the processing component 32.

[0201] The processing component 32 is used for the above Figure 1 The above embodiment describes an IoT-based method for vector-borne disease quarantine at ports of entry.

[0202] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above-described method. Alternatively, the processing component may be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described method.

[0203] Storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0204] Of course, computing devices may also include other components, such as input / output interfaces, display components, communication components, etc.

[0205] Input / output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices, input devices, etc.

[0206] The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

[0207] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

[0208] This application also provides a computer storage medium storing a computer program, which, when executed by a computer, can perform the above-described functions. Figure 1 The illustrated embodiment presents an IoT-based method for vector-borne disease quarantine at ports of entry.

[0209] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0210] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0211] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0212] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for quarantine of vectors at entry and exit ports based on the Internet of Things, characterized by, The method comprises the following steps: Synchronously collecting target environmental parameters and biological activity sensing data inside the cargo container to generate an environmental monitoring data set representing dynamic changes of the environment inside the cargo container; Capturing thermal radiation distribution characteristics inside the cargo container based on infrared sensing data to generate a thermal imaging map containing biological activity hot spot locations and cargo distribution density; Time-series correlating the environmental monitoring data set and the thermal imaging map, combining environmental interference parameters outside the cargo container, dynamically calibrating the correlation model of the biological activity sensing data and the thermal radiation distribution characteristics, and screening out target biological activity areas matching the activity patterns of rodents or insects; According to the shielding effect parameters of cargo distribution density on infrared thermal radiation in the thermal imaging map, the interior of the cargo container is divided into multiple monitoring sub-zones, and the boundary positions of the target biological activity areas are spatially calibrated based on the shielding compensation coefficients of each monitoring sub-zone to generate a biological activity distribution map; According to the spatial positioning information and density threshold of the target biological activity area in the biological activity distribution map, the visual evaluation result of the quarantine status of the vector organisms inside the cargo container is output.

2. The method of claim 1, wherein, Time-series correlating the environmental monitoring data set and the thermal imaging map, combining environmental interference parameters outside the cargo container, dynamically calibrating the correlation model of the biological activity sensing data and the thermal radiation distribution characteristics, and screening out target biological activity areas matching the activity patterns of rodents or insects, comprising: Assigning the same time markers to the biological activity intensity data in the environmental monitoring data set and the temperature data of each thermal radiation point in the thermal imaging map to form a synchronous time series pair; According to the temperature change rate and airflow disturbance frequency in the environmental interference parameters outside the cargo container, calculating the external interference coefficients of each thermal radiation point in the thermal imaging map; In the correlation model of the biological activity sensing data and the thermal radiation distribution characteristics, setting an initial weight value for the temperature data of each thermal radiation point, and the corresponding relationship between the initial weight value and the biological activity intensity data is determined by a pre-defined mapping rule; Based on the external interference coefficients, dynamically adjusting the initial weight value of each thermal radiation point, wherein for each thermal radiation point, the weight value decreases at a preset decay rate for each increase of a preset unit value of the external interference coefficient, to generate a calibrated dynamic weight distribution; Superimposing the biological activity intensity data in the synchronous time series pair and the calibrated dynamic weight distribution to calculate the comprehensive matching degree of biological activity intensity and thermal radiation temperature at each time marker; Extracting the change curve of the comprehensive matching degree within consecutive time markers, and performing morphological matching between the thermal imaging map region corresponding to the change curve and the pre-stored rodent activity characteristic curve and insect activity characteristic curve, and retaining the thermal imaging map region corresponding to the curve segment with a matching degree higher than a preset similarity as the target biological activity area.

3. The method of claim 1, wherein, Capturing thermal radiation distribution characteristics inside the cargo container based on infrared sensing data to generate a thermal imaging map containing biological activity hot spot locations and cargo distribution density, comprising: A plurality of infrared sensor arrays are arranged inside the cargo container to collect temperature data of each monitoring point at a fixed sampling period, forming an initial temperature distribution map; The initial temperature distribution map is dynamically analyzed to calculate the temperature change rate of each monitoring point in the continuous sampling period, and the area with a temperature change rate exceeding a set threshold is marked as a dynamic heat source area, and a three-dimensional spatial coordinate system is established by combining the internal structure parameters of the container to map the dynamic heat source area into a three-dimensional spatial model; According to the cargo loading information, the cargo stacking area is marked in the three-dimensional spatial model to exclude the static high-temperature area caused by the heat generated by the cargo itself; the remaining dynamic heat source area is subjected to cluster analysis, and the areas with similar spatial positions and temperature change characteristics are merged into candidate biological activity areas; According to the cargo stacking height and distribution, the infrared shielding coefficient of each candidate biological activity area is calculated to correct the temperature measurement error caused by cargo shielding; The area in the corrected candidate biological activity area that meets the preset biological characteristic parameters is determined as the final biological activity hotspot; based on the spatial position relationship between the cargo distribution density and the biological activity hotspot, a thermal imaging atlas with layered display function is generated.

4. The method of claim 1, wherein, The boundary position of the target biological activity area is spatially calibrated based on the shielding compensation coefficient of each monitoring partition to generate a biological activity distribution map, including: Obtain the shielding compensation coefficient corresponding to each monitoring partition, which reflects the shielding degree of the cargo in the partition to the infrared detection; Determine the monitoring partition where each target biological activity area is located, and calculate the boundary correction amount of the area according to the shielding compensation coefficient of the monitoring partition; expand or shrink the boundary position of the monitoring partition according to the boundary correction amount; Detect the boundary overlap between the adjusted target biological activity areas, and perform boundary fusion processing on the adjacent areas with overlap; Integrate the spatial position information of the processed target biological activity areas with the temperature distribution information in the thermal imaging atlas to generate a biological activity distribution map containing biological activity intensity distribution characteristics; in the biological activity distribution map, the region boundary adjusted by the shielding compensation and the original detection region boundary are displayed in different ways.

5. The method of claim 1, wherein, According to the spatial positioning information and density threshold value of the target biological activity area in the biological activity distribution map, the visual evaluation result of the quarantine state of the vector biological in the cargo container is output, including: Extract the center coordinates of each target biological activity area from the biological activity distribution map, and calculate the average value of the biological activity signal intensity of each area; Map the center coordinates of each target biological activity area to the three-dimensional spatial model of the cargo container, and adjust the projection position of the center coordinates in the vertical direction according to the shielding effect parameters of infrared thermal radiation on cargo distribution density; Compare the average value of the biological activity signal intensity of each target biological activity area with the preset density threshold value, and if it exceeds the threshold value, mark it as a high-risk area, otherwise mark it as a low-risk area; In the three-dimensional space model, high-risk areas and low-risk areas are marked with different colors, and a color block distribution map covering the internal structure of the cargo container is generated according to the projection position of the center coordinates of each area; The color block distribution map is superimposed with the loading cargo type data of the cargo container, and the comprehensive evaluation result containing the risk level and spatial positioning of the vector organism is output in the visualization interface.

6. The method of claim 2, wherein, According to the temperature change rate and airflow disturbance frequency in the environmental interference parameters outside the cargo container, the external interference coefficient of each thermal radiation point in the thermal imaging map is calculated, including: According to the temperature change rate in the environmental interference parameters outside the cargo container, the difference between the temperature change rate of each thermal radiation point of the thermal imaging map and the external temperature change rate is calculated; According to the airflow disturbance frequency in the environmental interference parameters outside the cargo container, the influence degree on the thermal radiation point temperature is calculated; The difference and influence degree are added according to the preset weight to obtain the external interference coefficient of each thermal radiation point; the external interference coefficient of each thermal radiation point is output.

7. The method of claim 3, wherein, Based on the spatial position relationship between the cargo distribution density and the biological activity hot spot, a thermal imaging map with hierarchical display function is generated, including: According to the density value of the cargo stacking area, the three-dimensional space model is divided into multiple vertically distributed density levels, each level corresponds to a preset density interval, a hierarchical segmentation plane parallel to the side wall of the cargo container is generated, and the cargo stacking area of different density levels is isolated into independent display layers by the segmentation plane; The coordinates of the biological activity area are automatically matched with the boundary range of each density level to determine the level; the hot spot mark is dynamically rendered on the hierarchical segmentation plane to generate a heat source distribution map bound to each density level; The heat source distribution map is superimposed and displayed with the cargo stacking model, so that the hot spot mark of each level is overlapped with the cargo area position, and an initial thermal imaging map is generated based on the superposition result; According to the user operation instruction, the hierarchical visibility configuration is performed on the initial thermal imaging map, the superposition display state is updated according to the configuration result, and the thermal imaging map supporting hierarchical interaction is output.

8. An IoT-based entry and exit port quarantine system for vectors, characterized in that, Including: The acquisition module is used for synchronously acquiring the target environmental parameters and biological activity sensing data in the cargo container to generate environmental monitoring data set representing the dynamic change of the environment in the cargo container; The first generation module is used for capturing the thermal radiation distribution characteristics of the cargo container based on the infrared sensing data to generate a thermal imaging map containing the position of biological activity hot spot and the distribution density of cargo; The screening module is used for time sequence correlation between the environmental monitoring data set and the thermal imaging map, and the correlation model of the biological activity sensing data and the thermal radiation distribution characteristics is dynamically calibrated combined with the environmental interference parameters outside the cargo container, and the target biological activity area matching the rodent or insect activity mode is screened out. A second generation module is configured to divide the cargo container into a plurality of monitoring sub-zones according to the shielding effect parameter of the cargo distribution density on the infrared thermal radiation in the thermal imaging map, and to spatially calibrate the boundary position of the target biological activity region based on the shielding compensation coefficient of each monitoring sub-zone, and to generate a biological activity distribution map; An output module is configured to output a visual evaluation result of the quarantine status of the vector biological organisms in the cargo container according to the spatial positioning information of the target biological activity region in the biological activity distribution map and a density threshold.

9. A computing device, comprising: The method comprises a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement the method for the quarantine of vector biological organisms at an entry and exit port based on the Internet of Things according to any one of claims 1-7.

10. A computer storage medium, characterized in that A computer program is stored, and when the computer program is executed by a computer, the method for the quarantine of vector biological organisms at an entry and exit port based on the Internet of Things according to any one of claims 1-7 is implemented.

Images