53 results about "Infrared remote sensing" patented technology

The invention discloses a deep learning-based satellite remote sensing collapsible loess foundation evaluation method, which comprises the following steps of: acquiring an optical remote sensing image, a radar remote sensing image and an infrared remote sensing image of a target area, and preprocessing; collapsibility feature weighted data are screened, and feature weights are distributed; feature extraction is carried out through a local space exhibition structure and a multi-layer Transform structure of the collapsibility foundation evaluation network; collapsibility area self-supervised collaborative learning is carried out, and self-supervised training is carried out based on spatial correlation, pseudo labels and spatial consistency regular terms; and carrying out risk grade division on the target area, and outputting a collapsibility risk result of each spatial position. According to the method, multi-mode remote sensing and deep learning are fused, high-precision intelligent partitioning of the collapsible loess foundation is achieved, and the method has the advantages of being self-adaptive, low in manpower and high in spatial resolution.

The invention provides a soybean growth monitoring method based on satellite remote sensing, and relates to the technical field of crop growth monitoring, and the method specifically comprises the steps: setting two monitoring nodes according to a soybean growth period, and collecting multispectral and thermal infrared remote sensing images and normal growth leaf area density and stem diameter sample data; a soybean area and a non-soybean area are distinguished by using a vegetation index threshold, cloud / shadow pixels are eliminated by using a multiband combination-threshold method, effective pixels are screened, and two types of remote sensing data are matched. Extracting leaf area density, canopy temperature and collaborative inversion stem diameter from the effective data; calculating the relative growth rate of the two types of indexes based on samples, and constructing a collaborative growth reference equation and a residual standard deviation; and substituting the growth rate of the planting area into an equation to obtain a theoretical value, calculating a deviation value, and judging the soybean growth state by combining the comparison between the deviation and a normal fluctuation interval and the positive and negative characteristics of the growth rate.

The invention discloses an infrared target detection method based on a bidirectional receptive field attention feature network, which is realized by constructing a model comprising a feature coding network, a feature fusion network and a detection head, and improves the detection precision and model generalization ability of an infrared remote sensing image, especially a small target. A receiving field attention convolution module is introduced into the feature coding network, different spatial positions in a receptive field are independently weighted, fine-grained feature extraction is enhanced, and the problem that multi-scale target feature extraction is insufficient is solved; further, the feature coding network optimizes feature extraction and fusion operations by alternately stacking RFAConv modules and standard convolution modules, so that key features of a target are reserved and enhanced to the greatest extent, and the problem of poor adaptability of a complex background is solved; the feature fusion network adopts a bidirectional feature pyramid network, high-level semantics and low-level detail information are bidirectionally and efficiently fused, and the problem of information loss in the fusion process is solved.

The invention discloses a lightweight unmanned aerial vehicle infrared remote sensing target detection method and device based on knowledge distillation, and belongs to the technical field of infrared weak and small target image recognition and deep learning model compression. In order to solve the problems of low detection precision, high omission ratio, high calculation complexity and difficulty in deployment at edge equipment of an existing unmanned aerial vehicle infrared remote sensing image weak and small target detection algorithm, an infrared remote sensing image data set is constructed, and a basic network model containing an intra-scale feature efficient interaction and fusion module is constructed; a teacher model and a student model are designed, a lightweight student network is trained by using a channel dimension knowledge distillation mechanism, and optimization is performed by using a WIoU loss function based on a distance attention mechanism and a soft-NMS, so that multi-scale feature fusion and model compression are realized. According to the method, the accuracy and robustness of weak and small target detection are improved, the model parameter quantity and the calculation quantity are reduced, and the method is suitable for real-time application in the fields of environmental monitoring, military reconnaissance and the like.

The invention provides a transformer substation noise traceability and diffusion simulation method fusing multi-source remote sensing data. Belongs to the technical field of power environment protection and remote sensing science. The method comprises the following steps: acquiring thermal infrared remote sensing image data of a transformer substation, after preprocessing, finding out a temperature anomaly region of main noise source equipment by using a temperature anomaly recognition algorithm, determining the position of the main noise source equipment in combination with equipment layout information, and generating position data of the noise source equipment; a laser radar is used for scanning a transformer substation and surrounding areas to obtain high-precision point cloud data, the high-precision point cloud data are processed to generate high-precision digital elevation model data, and a transformer substation three-dimensional scene model containing detailed topographic features and noise source positions is constructed in combination with noise source equipment position data. Temperature abnormity of main noise source equipment is identified by means of thermal infrared images, noise sources of specific equipment such as a main transformer and an electric reactor can be accurately positioned, equipment-level traceability is realized, the situation that traceability is difficult in a traditional method is changed, and a clear direction is provided for targeted noise treatment.

This invention provides a method for creating an infrared small target detection dataset, including S1, segmenting a satellite infrared remote sensing image to obtain at least two background images with different environments; S2, creating simulated target points; S3, adding simulated target points to the background images and fusing them with the background images; and S4, expanding the dataset of simulated target points. This invention uses space-based remote sensing satellite images as a background, providing a large data foundation for remote sensing infrared small target detection; it achieves diversity in complex backgrounds, motion characteristics of small targets, morphological diversity of small targets, and type diversity of small targets, making the simulated data closer to the required real data and greatly enriching the content of the dataset.

The application relates to the technical field of image processing, and discloses an infrared image enhancement method and system based on local phase correlation, wherein the infrared image enhancement method based on local phase correlation is used for acquiring continuous multiple infrared images, performing intelligent block division on a reference frame, calculating variance features, selecting an information-rich region of interest, independently performing phase correlation operation in each region to extract a local translation vector, obtaining global displacement estimation through weighted fusion, improving robustness by using an outlier detection algorithm, and finally realizing sub-pixel level image alignment and intelligent weighted fusion. The application is suitable for real-time enhancement processing of spaceborne infrared remote sensing images, can guarantee processing quality while meeting the resource constraint requirements of an embedded platform, and provides an efficient and reliable technical scheme for space remote sensing image processing.

The invention is suitable for the technical field of infrared telemetering gas detection and data processing optimization, and provides a multi-source infrared telemetering gas identification optimization method and system, and the method comprises the steps: sorting samples from small to large according to the average value of the actual conveying speed corresponding to a speed increase event in each sample, and analyzing the change trend of the compensation residual optimization degree along with the actual conveying speed based on the sample feature information so as to judge whether a turning interval which firstly ascends and then descends exists or not, and if yes, adjusting the set conveying speed of the belt type conveying device based on the actual conveying speed of the speed increasing event of each sample in the turning interval. According to the method, a technical thought of reversely optimizing the infrared telemetering methane recognition performance by utilizing micro speed fluctuation caused by the maintenance period is provided for the first time by systematically analyzing a slight speed increase event naturally generated by the belt type conveying device in the maintenance period.

The application discloses a kind of multi-wavelength pumped mid-infrared upconversion spectral detection methods, its characteristics are that the method uses multi-wavelength narrowband laser as the pump source of nonlinear conversion, simultaneously adopt multi-period quasi-phase matching crystal as nonlinear upconversion medium, realize mid-infrared signal broadband and efficient conversion, and combine high-performance near-infrared spectral measurement and control element, finally obtain wide band, high resolution, ultra-sensitive mid-infrared spectral detection.The application has the characteristics of large bandwidth and high efficiency compared with existing upconversion spectral technology, simultaneously can high fidelity resolution mid-infrared spectral information, for realizing wide band, high resolution, ultra-sensitive mid-infrared spectral measurement provides effective means, is expected to be applied to molecular spectroscopy, trace substance detection, infrared remote sensing and atmospheric monitoring and other important fields.

The utility model is suitable for the technical field of electric power engineering, and provides an electric power engineering construction investigation device which comprises an unmanned aerial vehicle body, a camera, a spectrograph, an infrared remote sensing analyzer and a glass protective cover are installed on the bottom face of the unmanned aerial vehicle body, and a cleaning mechanism is further arranged on the bottom face of the unmanned aerial vehicle body. The cleaning mechanism comprises a mounting shaft rotationally mounted on the glass protective cover, sweeping rods fixedly mounted on the mounting shaft and a driving assembly for driving the multiple sweeping rods to synchronously rotate, the sweeping rods are in clearance fit with the outer side face of the glass protective cover, and wiping strips are arranged on the inner walls of the sweeping rods; the glass protective cover provides protection for the camera, the spectrograph and the infrared remote sensing analyzer, and equipment is prevented from being impacted by foreign objects and polluted by dust, water vapor, rain, snow and the like. And the cleaning mechanism can clean the glass protective cover to ensure that the surface of the glass protective cover is clean, so that internal equipment can obtain clear and accurate data, and the problem of equipment performance reduction caused by lack of protection and cleaning of an existing device is solved.

A diaphragm device for simulating an infrared target and a control circuit thereof relate to the technical field of infrared remote sensing external field radiometric calibration, and comprise a driving assembly, a base, a diaphragm disc, a vertical rod and photoelectric switches, a step-shaped cylinder is arranged at the center of the base, and a plurality of photoelectric switches are uniformly distributed along the circumferential direction of the edge of the base. A vertical rod is arranged between any two optoelectronic switches, the diaphragm disc is rotationally connected to the step face of the cylinder, a plurality of diaphragm pieces are arranged in the circumferential direction of the diaphragm disc, the diameters of diaphragm holes in the centers of the diaphragm pieces are different, a shielding rod is fixed to the position, between any two diaphragm pieces, of the diaphragm disc, and the shielding rod is fixed to the position, between any two diaphragm pieces, of the diaphragm disc. The lower part of the shielding rod penetrates through the diaphragm disc and is in contact with the vertical rod; the diaphragm disc is driven by the driving assembly to rotate, and elastic sheet assemblies are correspondingly arranged on the two sides of the upper face of the diaphragm disc. According to the invention, the automatic switching of different diaphragm apertures is realized, the precise control of the diaphragm aperture position is realized through the cooperative work of the driving circuit, the decoding precision is high, and the operation is convenient.

This invention relates to the fields of data processing and remote sensing information processing technology, specifically to a remote sensing carbon source and sink monitoring and assessment method based on a multimodal hybrid expert model. The method includes: feature extraction from multi-source remote sensing data (hyperspectral, visible, and infrared); multimodal fusion of the extracted features; construction of a large-scale multimodal hybrid expert carbon source and sink model using a hybrid expert partitioning sparsity processing mechanism to detect and segment carbon source and sink changes, obtaining a carbon source and sink distribution map; assessment of annual carbon emissions / absorption rates for different types of land cover using a combination of source and sink distribution and ground measurement data; and calculation of regional carbon emissions and absorption based on land cover type area and annual carbon emissions / absorption rates to obtain carbon flux assessment results. This invention can be applied to carbon metering, regional carbon emission inventory compilation, and carbon neutrality effectiveness assessment, providing technical support for achieving the "dual carbon" goal and improving the effectiveness of remote sensing carbon source and sink monitoring and assessment.

This invention discloses a wireless network-based intelligent fire monitoring method and system, relating to the field of fire detection. The method includes: deploying infrared thermal imaging monitoring nodes based on a wireless mesh network; periodically monitoring through these nodes to obtain infrared image data; acquiring temperature data from the infrared image data and calculating surface humidity data; obtaining a two-dimensional image of fire hazard level distribution based on the surface humidity and temperature data; and conducting manual or robotic patrols based on the two-dimensional fire hazard level distribution image. This invention uses infrared remote sensing to calculate the surface humidity and ground temperature of combustible materials within the detection area, combined with multi-node collaborative monitoring via a wireless mesh network and a quasi-effective timing correction algorithm. This fundamentally transforms fire prevention from a passive response to proactive prevention, significantly improving early warning accuracy and the efficiency of resource utilization in fire control.

This application relates to the field of infrared remote sensing instrument calibration technology, and provides a high-dynamic-range, large-aperture, high-precision on-board blackbody radiation source and its temperature control method. The main body of this on-board blackbody radiation source is temperature-controlled by a refrigerator, exhibiting a large dynamic range, capable of temperature control from 200K to 310K. Through a three-layer structure consisting of the radiation source cavity, a cooling inner shell, and an outer shell, zoned temperature control via a heat transfer bracket, and a three-layer thermal insulation installation design, the temperature uniformity and stability of the blackbody radiation source are significantly improved, with temperature uniformity exceeding 30mK. Simultaneously, it possesses high emissivity (spectral emissivity reaching 0.998) and an aperture as large as 310mm, meeting the requirements for large field-of-view, long-distance use.

A high-precision radiometric calibration method for very high sensitivity infrared sea imaging data is proposed. Firstly, the blackbody calibration data on the infrared remote sensor satellite are used to calculate the radiometric calibration coefficient and detect the blind pixels. Then, the image data are corrected by the primary relative radiometric correction and blind pixel compensation correction. Secondly, a radiometric correction method based on small sample statistics is proposed to remove the residual stripe noise after the primary correction, which further improves the radiometric quality of the infrared image. Finally, the blind pixels in the image are detected and removed.

The application provides a near real-time fire point dynamic monitoring method, device, equipment and medium, relates to the technical field of infrared remote sensing monitoring, and the method comprises the following steps: acquiring thermal infrared remote sensing data and land cover product of a forest and grass monitoring area, and performing data preprocessing; the thermal infrared remote sensing data is segmented to obtain tile data, and cloud extraction is performed based on the tile data; the extracted cloud is used for mask of the thermal infrared remote sensing data, and the medium infrared band and the far infrared band are segmented to obtain an isothermal graph spot layer; the isothermal graph spot layer is masked through water body and building area corresponding to the land cover product; the masked isothermal graph spot is comprehensively identified based on the statistical characteristics of the isothermal graph spot and the comparison characteristics of the isothermal graph spot and adjacent background graph spots, and a fire point detection result in the forest and grass monitoring area is obtained. The application alleviates the technical problems of pixel-by-pixel identification, large amount of calculation, low efficiency and limited precision in the prior art.

This invention discloses a lightweight method, device, and medium for small target recognition in infrared remote sensing images. The method includes the following steps: Step S1: Constructing and preprocessing an infrared remote sensing image dataset; Step S2: Constructing a lightweight backbone network HLNet, training the HLNet based on the infrared remote sensing image dataset, and obtaining a trained HLNet network; Step S3: Constructing a lightweight multi-scale feature fusion network LSF, inputting the feature map output by the trained HLNet network into the LSF for multi-scale feature fusion, and obtaining an enhanced multi-scale feature map; Step S4: Inputting the enhanced multi-scale feature map into a detection head module for small target recognition, and obtaining the recognition result. This invention can effectively recognize small infrared targets in complex backgrounds while reducing model complexity.

The application discloses a kind of dam leakage passage digital twin systems and diagnostic methods based on flow-heat coupling inversion reconstruction, the diagnostic method includes: obtaining the time series thermal infrared remote sensing observation data of dam to be measured, environmental driving data and digital twin dam model;Under the assumption of no leakage, seepage-heat coupling simulation is carried out, and the benchmark temperature field that is continuous in space-time is generated;Based on the difference between time series thermal infrared remote sensing observation data and benchmark temperature field, extract thermal infrared anomaly index;Seepage-heat coupling inversion model containing physical constraint and structured constraint is constructed, and three-dimensional leakage parameter field inside dam is obtained by minimizing the difference between observation and simulation inversion;It is written back to digital twin dam model to carry out seepage stability calculation, and the leakage risk classification early warning result of dam is obtained.The application effectively solves the problem of ill-conditioned and environmental interference of thermal infrared inversion, realizes the structured reconstruction and risk quantitative evaluation of dam leakage passage.

The application provides an infrared remote sensing image target classification method and device, and relates to the technical field of infrared target detection and identification. The method comprises the following steps: obtaining a target slice containing a ship target from an infrared remote sensing image; based on the target slice, extracting infrared radiation features and gray statistical features of the ship target; associating and fusing the infrared radiation features and the gray statistical features of the same target slice to generate double-domain fusion features which take into account spatial details and thermal physical mechanisms; and based on the double-domain fusion features, determining the category of the ship target.

The invention discloses a semantic understanding method based on cross-modal image target detection, and relates to the technical field of target detection, and the method comprises the steps: carrying out the feature extraction of a visible light remote sensing image and an infrared remote sensing image through a multi-branch convolutional neural network; performing multi-scale fusion operation on the extracted features through a feature pyramid network; inputting the feature map obtained by the fusion operation into a region proposing network, generating a visible light remote sensing image feature candidate region and an infrared remote sensing image feature candidate region, and extracting features from the candidate regions; matching the candidate region features through a cross-modal similarity measurement network to obtain an association matching result; determining a visible light semantic feature map and an infrared semantic feature map according to an association matching result; and obtaining a target detection result and target semantic description based on the visible light semantic feature map and the infrared semantic feature map. Therefore, the target detection accuracy and the target description accuracy are improved.

The application provides a kind of all-weather ozone inversion method and system based on Fengyun three microwave data, including data acquisition and preprocessing, ozone inversion sample set establishment, ozone inversion model establishment and training, ozone inversion model verification four steps, through the fusion Fengyun three satellite microwave load multi-band data, ozone column total inversion is carried out based on machine learning method, XGBoost gradient boosting decision tree model is constructed using the training set, and the model training is completed by optimizing the hyperparameter through cross validation method, the precision and all-weather inversion ability of the model are verified using the test set, the inversion result of all-weather ozone column total is output, high-precision, all-weather ozone column total monitoring is realized.The application effectively overcomes the limitations of traditional ultraviolet and infrared remote sensing, which depends on light conditions and is easily disturbed by clouds, fills the technical blank of ozone column total inversion based on satellite microwave load, and realizes truly all-weather, all-time high-precision monitoring.

The invention discloses a control method and device for automatic irrigation of farmland crops. The method comprises the steps that infrared remote sensing image data, collected by an unmanned aerial vehicle in the air, of the farmland crops and ground text data, collected by ground equipment, of the farmland crops are obtained; performing noise reduction processing and background elimination processing on the infrared remote sensing image data to obtain canopy temperature data of the farmland crops; based on the ground text data and the canopy temperature data, the actual water deficit of the farmland crops is determined; and according to the actual water deficit of the farmland crops, controlling irrigation equipment in the field to execute a corresponding water application strategy. According to the scheme, water can be accurately supplemented to the crops according to the actual conditions of the farmland crops, so that good growth condition guarantee is provided for the farmland crops while water resources are saved.

The application provides an ozone profile inversion method based on thermal infrared remote sensing and belongs to the technical field of atmospheric environment remote sensing. The technical scheme is as follows: S1, data collection and processing; S2, an ozone prior profile model; S3, a temperature and humidity profile error correction model; S4, thermal infrared ozone profile optimization inversion; and S5, determination of a proportional factor. The application has the beneficial effect that the stratospheric column total amount and the tropospheric column total amount are respectively used for constraint in the optimization inversion, so that the accuracy of the ozone profile, especially the accuracy of the lower troposphere, is improved, and the deficiency of the thermal infrared sensor in the nadir observation mode is made up.

The present application relates to a kind of laboratory warm drainage infrared remote sensing measurement geometric calibration device and method, comprising: warm drainage experimental water tank, infrared camera, thermal resistance geometric calibration point, the setting position of each thermal resistance geometric calibration point is: along water tank left and right sides at least one column thermal resistance geometric calibration point is distributed;Thermal resistance geometric calibration point along water tank long direction The spacing needs to be guaranteed in the infrared camera frame upper and lower direction at least can accommodate two rows of thermal resistance geometric calibration point;Thermal resistance geometric calibration point size is: at least 1 pixel in the frame of infrared camera.The present application uses thermal resistance in water body, by adjusting current, makes thermal resistance show obvious mark as geometric calibration point in infrared photo, to obtain accurate calibration on infrared photo, using these accurate calibration points to carry out a series of processing to the infrared photo taken, through these processing, high-quality warm drainage thermal field distribution diagram is obtained, the precision of warm drainage experiment is greatly improved.

The invention discloses a surface water body intelligent extraction method and system based on multi-source remote sensing image data, and belongs to the technical field of multi-source remote sensing. High-spatial-resolution visible light and near-infrared remote sensing images in different time phases are collected, image registration and radiometric calibration are carried out, and a unified time sequence image sequence is constructed. And based on the normalized water body index, extracting water body response change characteristics of each pixel, identifying a water body change salient region and an occlusion mutation existence region, and generating a corresponding binary mask graph. Constructing three-channel image input data and a change guide layer by combining spectral characteristics and change information, guiding and prompting to generate network output points, frames or mask prompt information, inputting the information into an SAM mask decoder to obtain water body mask graphs of all time points, dividing permanent and seasonal water bodies according to stability indexes, generating a complete water body distribution graph, and outputting the water body distribution graph to a water body distribution graph. The method improves the intellectualization and stability of water extraction in a complex earth surface scene, and is suitable for various remote sensing application scenes such as water resource monitoring and water environment evaluation.

The invention relates to the technical field of surface temperature monitoring, and discloses an unmanned aerial vehicle thermal infrared remote sensing-based surface temperature inversion method, which comprises the following steps of S1, constructing a comprehensive flight parameter optimization model according to environmental conditions monitored in real time so as to calculate and set the optimal flight height and flight speed of an unmanned aerial vehicle; s2, in the flight process of the unmanned aerial vehicle, ground surface thermal infrared remote sensing data are collected through a thermal infrared sensor carried by the unmanned aerial vehicle, real-time flight parameters and environment parameters of the unmanned aerial vehicle are synchronously recorded, and by optimizing the flight parameters, the thermal infrared data quality stability is improved, and the inversion accuracy problem is solved; calibrating and optimizing model parameters by means of a ground target, and eliminating system errors without field calibration; and through multi-source correction, optimization inversion and noise reduction processing, high-resolution and high-precision surface temperature data is obtained, the multi-field high-precision requirement is met, and the thermal infrared remote sensing potential of the unmanned aerial vehicle is fully played.

The application discloses a semantic understanding method based on cross-modal image target detection, relates to the technical field of target detection, and extracts features of visible light remote sensing images and infrared remote sensing images through a multi-branch convolutional neural network; performs multi-scale fusion operations on the extracted features respectively through a feature pyramid network; inputs a feature map obtained through the fusion operation into a region proposal network to generate visible light remote sensing image feature candidate regions and infrared remote sensing image feature candidate regions, and extracts features from the candidate regions; matches the candidate region features through a cross-modal similarity measurement network to obtain an associated matching result; determines visible light semantic feature maps and infrared semantic feature maps according to the associated matching result; and obtains target detection results and target semantic descriptions based on the visible light semantic feature maps and the infrared semantic feature maps. In this way, the target detection accuracy and the target description accuracy are improved.

The invention relates to an optical calibration system. The optical calibration system comprises a base; the optical assembly comprises a supporting frame and a reflecting mirror, the reflecting mirror is arranged on the supporting frame in the mode that the mirror surface area is adjustable, and the supporting frame is movably arranged on the base so that the reflecting mirror can reflect the received light rays to the remote sensing satellite. By moving the support frame relative to the base, the mirror surface angle of the reflector can be adjusted, so that the reflector can reflect the received light to a remote sensing satellite. By adjusting the mirror surface area of the reflector, the energy of the light reflected by the reflector can be adjusted. And when the area of the reflecting mirror is large, more light can be reflected by utilizing the large area. And when the area of the reflecting mirror is small, less light is reflected by utilizing the small area. Moreover, the calibration system can reflect the visible light band radiation of the sun to serve as a visible light point source to calibrate the visible light remote sensing load, and can also reflect the infrared band radiation of the sun to serve as an infrared point source to calibrate the infrared remote sensing load.