144 results about "Radar remote sensing" patented technology

The invention belongs to the technical field of radar remote sensing, and relates to an attention pyramid network-based SAR image multi-scale ship detection method. On the basis of an existing featurepyramid network, the invention provides a multi-scale feature extraction method with adaptive selection of significant features, namely the feature pyramid network based on a dense attention mechanism, and the method is applied to SAR image multi-scale ship target detection. Prominent features are highlighted from global and local ranges through the channel attention model and the spatial attention model respectively, and better detection performance is obtained; meanwhile, the attention mechanism is applied to the multi-scale fusion process of each layer, the characteristics can be enhancedlayer by layer, false alarm targets can be effectively eliminated, and the detection precision is improved.

The invention discloses an NRIET weather multisource detecting data fusion analysis system. The system comprises data collection, data quality control, data normalization treatment, background field grid interpolation, multisource data fusion, fusion analysis and three-dimensional weather information display analysis, wherein the data collection is that various weather data of ground and upper airroutine observation, satellite and radar remote sensing observation and pattern background field is acquired; the data quality control is that quality control treatment is carried out on the collected multiple observation data; the data normalization treatment is that the data normalization treatment is carried out on the data which undergos the quality control treatment, and universal standard data interfaces are provided for subsequent modules to be called; the background field grid interpolation is that module input is taken as a background field to be interpolated to three-dimensional lattice points, and three-dimensional background data is acquired; the multisource data fusion is that the observation data is used for fusion, and the three-dimensional background data is revised; the fusion analysis is that based on the data fusion, ground analysis, temperature analysis, cloud analysis, water vapor analysis and balance analysis are further utilized, and further revisal is carried out for diagnostic analysis; the three-dimensional weather information display analysis is that three-dimensional lattice point data undergoing the fusion analysis is displayed, and an interaction analysis function is provided.

The invention discloses a three dimensional polarization image laser radar remote sensor. The invention comprises a light path system, a device for the generation of polarization light and the reception of dispersion light, the generation of three dimensional data and a processing method for three dimensional data. The invention is characterized in that the light path system is the same optical system which is composed by a range finder and a device for the generation, emission and detection of the polarization light. The device for the generation, emission and detection of the polarization light comprises a laser instrument, a beam splitter, a double frequency crystal, a plurality of lenses, a code disc, a scan motor, a scanning mirror, a telescope, a polarizer, a wave plate, an echo ranging reception device, a color separation filter, a CCD and a computer. The generation device and the method for target three dimensional surface data comprises a distance measurement device and the method for distance measurement data generation, location data generation and remote sensor attitude data generation. The processing method for polarization image data and three dimensional surface data comprises the processing method for laser distance measurement data, location data and attitude data. The invention can obtain the three dimensional image of the detected target with polarization information.

The invention discloses a radar remote sensing-based detection method of soil alkalization, comprising the following steps: computing the polarization parameter based on full polarimetric SAR data, dividing bare areas and vegetation-covered areas and generating images with masks; delimiting test areas and adopting measuring data to establish a dielectric model of moist salt soil; adopting the full polarimetric SAR data to invert a complex permittivity map and carrying out RD geocoding, data information separation and conversion; based on the dielectric model and the complex permittivity, adopting the genetic algorithm for inversion to obtain moisture contents and salt contents of areas, and integrating geographic coordinate information and a vegetation cover mask map to generate a moisture content map and a salt content map; adopting the complex permittivity and the moisture contents of areas to differentiate saline soil from alkali soil. By adopting the technical proposal of the invention, the moisture content and the salt content of the soil in saline areas can be obtained, the degree of dependence on ground assistant data is reduced and simultaneously the dielectric behavior of the moist salt soil can be better depicted.

The invention, which belongs to the field of radar remote sensing application technology, in particular relates to a large-scene SAR target recognition method based on a deep neural network. A neuralnetwork structure based on a multi-scale feature map is built, multi-scale features of an SAR image are extracted, and the output of each layer of convolutional neural network is used for prediction.Meanwhile, with combination of advantages of the a deep neural network in terms of feature extraction, layer-by-layer non-linear transformation is carried out based on the neural network structure andlow-level and high-level features of an SAR image target are extracted automatically. The main four steps, including detection, authentication, feature extraction, and identification, of the traditional SAR target identification method are integrated into one neural network; and not detector, authentication device, and classifier need to be designed independently. A cutting module is added in front of the network to complete quick explanation of the large-scene SAR image.

The invention which discloses a radar remote sensing method for detecting the distribution of the water content in soil in a depth direction relates to the radar remote sensing detection technology. The method comprises the following steps: 1, respectively adopting two modes of horizontal polarization and vertical polarization to measure the scattering field of the soil under multiple visual angle and multiple frequency observation; 2, filtering out components of direct waves in radar echoes; 3, scaling echo data; and 4, utilizing the scattering model and the dielectric constant of the soil to obtain the distribution of the water content in the soil in the depth direction. The radar remote sensing method for detecting the distribution of the water content in the soil in the depth direction of the invention allows the distribution of the water content in soil with the depth under the soil surface of 0-3m to be detected, and can be applied to ground, air and space base remote sensing monitoring platforms.

The invention discloses an early identification method of potential landslide based on the remote sensing technology, and the method comprises the following steps: S1, obtaining remote sensing data, landform data and geological data of a projection region; S2, extracting a landslide development factor and a value range; S3, building a landslide factor set standard unit; S4, building a model, and dividing a landslide development high-risk region; S5, carrying out the differential interferometer radar remote sensing monitoring of the landslide development high-risk region, extracting the deformation and crack information of the high-risk region, and achieving the early recognition of the potential landslide. The method can achieve the early recognition of the potential landslide in a large range, achieves the focusing of a key region, breaks the point measurement range of a conventional method, achieves the area monitoring recognition, can greatly reduce the manpower and material resource investment, remarkably reduces the early recognition cost of the potential landslide, and improves the project operation safety and disaster prevention and disaster mitigation level.

The invention discloses a laser radar echo full-waveform acquisition device with sampling point time location, and belongs to the technical field of earth observation laser radar remote sensing. The full-waveform acquisition device comprises a main control module, a high-speed ADC sampling module, a high-speed FIFO cache module, an interpolation delay line module, a high-speed coarse counter module, and a fine measurement coding module. All the modules except the high-speed ADC sampling module are implemented inside an FPGA. The invention aims to provide a laser radar echo full-waveform acquisition device for acquiring sampling point time location by a simple method. A frequency division clock synchronous with a high-speed ADC sampling clock is used as a rough counting clock, echo sampling point data is stored in a certain time interval under control, and the time interval between a main wave pulse and a coarse counting clock edge is finely measured by use of an interpolation delay chain, thereby obtaining an echo sampling point which has precise time location relative to laser transmitting main wave pulse. The laser radar echo full-waveform acquisition device with sampling point time location has the advantages of simple peripheral circuit, concise means of realization, flexible function interface configuration, and high cost performance.

The invention relates to a multi-resource remote sensing distinguishing method and system of an underground potassium- and halide-containing water resource in a salt lake area. The method comprises the following steps: acquiring data, namely, acquiring multi-resource remote sensing image data; processing data, namely, processing multi-spectral remote sensing image data and high-resolution remote sensing image data by the multi-color transformation technology, building a remote sensing stratum lithology interpretation mark and performing remote sensing stratum lithology interpretation; extracting salt-containing abnormality information from the multi-spectral remote sensing image data by the main component analyzing technology; extracting the salt information by the interpretation extracting method; processing a radar remote sensing image data, building a geologic structure interpretation mark for potassium- and halide-containing water resource and performing geologic structure interpretation; integrating data, namely, processing three information obtained in data processing by overlapping, projection transformation and matching. According to the method, the multi-resource remote sensing image data are processed, interpreted and integrated to realize accurate distinguishing of the potassium- and halide-containing water resource.

A detecting device and method for mass concentrations of aerosol granules are disclosed. The device includes a laser emitting laser pulses of a plurality of wavebands. The laser pulses enter air through a collimation and beam expanding system and a reflector. Laser light scattering echo signals are received by a telescope, are introduced through optical fibers into a beam splitting system, and are then subjected to signal conversion through a photoelectric detector. A data collecting system performs data collecting on the signals. Finally, data processing is performed in a computer to invert an aerosol granule mass concentration. The device and the method are based on novel theories and novel methods of granule mass concentration distribution detection by multi-waveband laser radar remote sensing, can achieve high-spatial-resolution detection of granule mass concentrations in a range of a plurality of kilometers, effectively overcome a difficulty that present granule mass concentration spatial distribution monitoring techniques are not mature, and provide air scientific research, space environment monitoring and executive decisions with scientific bases and methodologies.

The invention relates to a calculation method for detecting the flow of the section of a river in a non-contact manner. The method includes the following steps that: step 1, a radar is adopted to detect river surface flow data; step 2, open channel flow cross section flow rate distribution is calculated; step 3, a relationship between a surface flow rate and a water depth is determined; step 4, a relationship between the surface flow rate and the average flow rate of the section is determined; and step 5, the total flow of the section is calculated. According to the calculation method of the invention, a radar remote sensing non-contact measurement method is adopted, so that good maneuverability, flexibility and real-time performance can be realized, and highest detection efficiency can be achieved with smallest measurement cost, a widest detection range can be realized, and measurement can be performed in a non-water-contact manner; and therefore, the calculation method is competent for hydrological surface information measurement under complex geological conditions and external adverse weather conditions.

The invention provides a new radar remote-sensing polarization imaging system. It can make use of polarization information of target backward scattered light to carry out radar remote-sensing detection for target and quick real-time polarization degree imaging. The system comprises semiconductor laser, polarization-status generating system, beam expanding system, cassegrain telescope system, polarization-status detecting system, and data collecting and imaging system. The invention realizes polarization degree detection by the polarization-status generating system and the polarization-status detecting system and adopts polarization splitting prism in the polarization-status detecting system to realize a first detection for target polarization degree and real-time imaging.

The present invention provides a synthetic aperture radar remote sensing image ocean floating raft recognition method for a deep-cooperative sparse coding network, and belongs to the field of computerremote sensing image processing and pattern recognition. The method comprises the following steps: preprocessing a synthetic aperture radar remote sensing image to obtain a remote sensing image withan accurate geographic coordinate position, low speckle noise and high image visibility; extracting texture features and contour features of the radar remote sensing image; performing super-pixel segmentation on the texture image and the contour image; and finally, inputting a super-pixel block feature image into a classification recognizer to obtain a classification recognition result. The effectand benefit of the method provided by the present invention is that: the supervised classification accuracy of the synthetic aperture radar remote sensing image can be greatly improved.

The invention discloses an SAR ship target detection method based on network pruning and knowledge distillation, and belongs to the technical field of radar remote sensing image application. Accordingto the technical scheme, firstly, a backbone network structure suitable for SAR multi-scale and large-length-width-ratio targets is designed, and then channel pruning is conducted on the network to generate a compact model. In addition, a knowledge distillation strategy is applied to make up performance degradation caused by network pruning. Different from full-feature simulation in a common detection model distillation method, the method has the advantages that the mutual relation between feature maps of different levels is used as migration knowledge in distillation, and a novel attention mechanism is designed to enhance target-related features, so that the distilled features have higher representation capability. By adopting the detector constructed by the invention, the model size of2.8 M, the reasoning speed of more than 200fps and lower calculation cost can be realized, and the detection precision is also improved to a certain extent.

The invention provides a super-high/ultra-high voltage power transmission line radar remote sensing monitoring method. A monitoring system employed by the method comprises the following three units: a data preprocessing unit, a lead and iron tower object characteristic detection unit and a line safety state assessment unit. Data acquired by a high resolution SAR (Synthetic Aperture Radar) satellite goes through the data preprocessing unit, the lead and iron tower object characteristic detection unit and the line safety state assessment unit in order, thus line safety state assessment information is generated. According to the invention, by utilizing all-weather and all-time information expression force of high resolution SAR satellite remote sensing data in wide area space, and specific reflection characteristics of a power transmission line and an iron tower in the high resolution SAR data, safety state of the power transmission line is monitored. Through fusion processing on lead and iron tower reflection speckle information and object characteristic detection data, line safety state assessment data is generated, and a scientific basis for electrical network power transmission line safety state early warning and disaster assessment is provided.

The invention belongs to the radar remote sensing application technology field and particularly relates to a synthetic aperture radar target identification method based on auxiliary decision update learning. According to the method, a small amount of initial training samples are utilized to train an initial model, newly-added unlabeled images are utilized as test samples, the identification results are taken as training samples for next training, on the basis of the existing model, iterative training is carried out till an identification system with stable and mature identification efficiencyis acquired. The method is advantaged in that the convolutional neural network is utilized as the main body to extract deep characteristics of SAR targets to carry out classification, in combination with auxiliary decision of an auxiliary classifier, so the newly-added unlabeled SAR images can be directly applied to the existing classifier, moreover, repeated sample training is avoided, and identification efficiency is improved.

The invention belongs to radar remote sensing or image processing technology, namely, analyzing radar observation information by using image processing technology, and specifically relates to a high resolution SAR image target detection method based on a part model. According to the invention, through performing characteristic compression and screening on each pixel point of original SAR image data, low-level characteristic interest points are extracted, a target part initial unit is generated from interest point local information, calculating initial unit characteristics, generating a target part combination model through an image segmentation method and realizing high resolution SAR image target detection by utilizing description of the target part model. According to the invention, distribution characteristics of a target in the high resolution SAR image are utilized fully and effectively. The target part model obtained through detection is comparatively complete, so that the integral information of a target can be kept well. The method is good in noise robustness, high in algorithm stability, accurate in detection result and can improve the self-adaptive capability of image detection.

The invention belongs to the technical field of radar remote sensing application, and particularly relates to an SAR image ship target detection and recognition integrated method based on deep learning. An important part of SAR image ship target interpretation is detection and identification, but all detection and identification are independent at present, and a detection and identification integrated method is an important research direction of SAR image interpretation in the future. Existing traditional methods and deep learning methods for target detection and recognition cannot achieve detection and recognition integration of SAR image ship targets. The invention provides an SAR image ship target detection and identification integrated method. Through an existing deep learning networkframework, a network used for SAR image ship target detection and identification integration is provided. Detection and classification tasks of ship targets are carried out at the same time by mainlyutilizing a detection and classification sub-network at the tail end of a network, so that a detection and identification integrated target is realized. Compared with a traditional ATR technology, thedetection and recognition process of the method is simpler and more efficient.

The invention discloses a method for early identification of urban ground deformation disasters, and belongs to the technical field of urban ground deformation monitoring. The method includes steps: step 1, regional general investigation based on a remote sensing technology: analyzing urban ground data obtained by radar remote sensing or optical remote sensing, preliminarily discriminating hiddenrisk regions of urban ground deformation, and determining potential deformation regions; step 2, detailed investigation of the potential deformation regions based on an unmanned aerial vehicle photography measurement technology: monitoring potential dangerous regions, identified in step 1, of ground deformation through the photography measurement technology by regarding fixed-wing and rotor-wing unmanned aerial vehicles as carriers, and performing data check and verification with the combination of a ground monitoring station; ad step 3, onsite check: organizing point-to-point check of potential geological disaster hidden risk places designated through technical means in step 1 and step 2 by personnel, and determining checked places whose ground crack widths are greater than 10 mm with sustainable development or whose sedimentation vertical displacements are greater than 30 mm with sustainable development as hidden risk regions and paying continuous attention.

The invention belongs to radar remote sensing or image processing technologies, and particularly relates to a method for detecting an SAR image cluster target based on target number posteriority. Themethod comprises the steps of extracting interesting regions from an SAR image and extracting a strong scattering center in each interest region to estimate the range of possible target number; selecting an estimated value of the target number in the range and dividing the scattering center in each interesting region into target regions by using a segmentation algorithm based on the constraint ofthe number; analyzing the similarity of the target regions through the obtained spatial size and intensity distribution of the target regions, and calculating the posterior probability of the estimated value of the target number; finally, finding the estimated value with the largest posterior probability by iteratively comparing detection results under different target numbers, and using the corresponding target region as the cluster target detection result of the method. The method for detecting the SAR image cluster target based on the target number posteriority achieves the detection and segmentation of the cluster target in the SAR image, and can more accurately distinguish different individuals in the target cluster.

The invention discloses a radar remote sensing multi-scale surface roughness measuring method based on three-dimensional microtopography reconstruction. The method comprises the following steps: (I) field measurement: (I-1) determining the direction of measurement, (I-2) limiting the shooting range, (I-3) laying coordinate control rulers, (I-4) shooting images, and (I-5) repeating (I-1) to (I-4) to carry out measurement repeatedly; (II) interior work processing: (II-1) matching image pairs with the same name, (II-2) carrying out three-dimensional calibration and precision validation, (II-3) extracting roughness sampling points, (II-4) selecting measurement points and working out a roughness curve, and (II-5) calculating roughness parameters in multiple scales. By means of the method, the roughness of a large earth surface can be measured, the roughness of a survey region in different wave length scales can be obtained, and regional surface roughness measurement can be achieved. The method has the advantages of being high in measurement accuracy, high in efficiency, flexible and convenient to implement and high in practicality, thereby being suitable for application and popularization.