42 results about "Spectral matching" patented technology

The invention relates to the technical field of fault detection, in particular to a heating and ventilation system fault positioning system and method based on big data, and the system comprises a multi-source sensing module, a disturbance feature module, a path modeling module, a frequency spectrum matching module and a fault positioning module. In the method, a real-time disturbance sequence is constructed through time window segmentation and parameter offset calculation, separation of an active response chain and an abnormal propagation path is realized through a directional joint state vector and a topological relation table, and time asynchronism of multi-device signal transmission is eliminated by adopting a dynamic time warping algorithm. In combination with a real-time parameter bidirectional verification mechanism of a frequency domain main frequency band energy mark, a valve opening degree and a pump rotating speed, the problem of path confusion in a multi-node parameter coupling scene of a traditional method is solved, the tracing efficiency of concurrent faults in a complex pipe network system is improved, the adaptability limitation of a single-dimensional threshold mechanism to equipment performance degradation is overcome, and the method is suitable for a complex pipe network system. And the error positioning probability caused by signal delay superposition is reduced.

The invention relates to the technical field of solid waste treatment, in particular to an intelligent waste sorting method based on multi-sensor fusion. Carrying out multi-level fusion decision making: carrying out sewage reflection suppression on the visible light image; carrying out matching identification and rejecting a liquid pollution area based on a near infrared spectrum; fusing the visible light texture, the spectral feature and the electromagnetic signal to generate a space grid feature vector; identifying the material category through a cascade classifier, and dynamically adjusting the weight of each sensor in the decision according to the environmental parameters; self-adaptive sorting execution: triggering a corresponding sorting execution mechanism according to an identification result; based on the transmission belt speed, the execution mechanism response delay and the material physical attribute, the action advance compensation amount is calculated; the action time of an execution mechanism is adaptively prolonged for the stacked materials according to the stacking hierarchy; and online feedback optimization. Through dynamic compensation and time sequence adjustment, the time sequence consistency of multi-sensor data is ensured, and the influence of band speed fluctuation is overcome.

This application provides a simulation method and system for intelligent fusion and dynamic prediction of electronic warfare intelligence. This simulation method uses multi-domain sensors to collect electromagnetic signals in real time, adopts a dynamic adversarial network (DAN) for spectrum matching and anomaly location, and combines spatiotemporal verification to eliminate interference data. It constructs a semantic association graph of radar, communication, and geographic information, fuses time-domain and spatial-domain features through spatiotemporal attention, and dynamically allocates confidence to achieve multi-source data fusion. It predicts threat evolution trends based on a spatiotemporal transformer, outputs threat level, type, and intent, and generates a dynamic situation map in combination with a hybrid density network. It adopts an online incremental learning framework to adapt to new signal patterns, prioritizes updating the prediction module, and dynamically adjusts push priorities through edge-cloud collaborative optimization of a reinforcement learning-driven multi-path distribution mechanism to achieve low-latency transmission of key threat intelligence.

The invention discloses a bridge disease unmanned aerial vehicle multi-sensor cooperative detection method, and relates to the technical field of bridge monitoring, and the method comprises the following steps: generating a corrosion layering thickness distribution condition through a spectrum matching algorithm; acquiring three-dimensional point cloud data of the surface of the bridge through a laser radar carried by an unmanned aerial vehicle, and generating deformation distribution data; calculating a local stiffness degradation coefficient matrix according to the thickness distribution condition of the erosion layer, and inputting the local stiffness degradation coefficient matrix into the finite element model to generate predicted deformation distribution data; comparing the deformation distribution data generated in the step 2 with predicted deformation distribution data to generate residual data, generating a dynamic detection priority map by the residual data, and marking a bridge area needing to be remeasured; and executing the re-flight planning detection strategy. Structural rigidity parameters are dynamically corrected according to the corrosion layering thickness, model precision is optimized by combining actual measurement deformation data feedback, closed-loop optimization is formed through data interaction, and the problem of one-sidedness of single sensor detection is solved.

The invention relates to the technical field of illuminometer calibration, and discloses a calibration device and a calibration method of an illuminometer for measuring medium and low illuminance. The calibration device comprises an integrating sphere assembly, a light source control module, a reflecting plate adjusting module, a spectrum monitoring module and a calibration execution module. A plurality of fixing holes are formed in the integrating sphere assembly, the light source control module achieves stable light intensity output through a dynamic current distribution algorithm and PID closed-loop control, the reflecting plate adjusting module compensates edge light attenuation through a self-adaptive positioning algorithm, and the spectrum monitoring module evaluates spectrum stability through a spectrum matching algorithm. And the calibration execution module adopts a nonlinear error correction algorithm to improve the calibration precision. The calibration method comprises the steps of initializing the device, adjusting the height of the reflecting plate, collecting data to generate a calibration curve, correcting errors, testing at multiple angles and the like. According to the invention, the defects of the traditional calibration technology can be effectively overcome, and the calibration precision and reliability of the medium and low illumination illuminometer are improved.

The invention provides a small sample channel state information data expansion method based on a generative adversarial network, and the method comprises the steps: generating a vivid time-frequency domain channel sample through a Wasserstein generative adversarial network based on physical constraints under the condition of rare or missing drive test data; and a time-frequency joint discriminator is combined to effectively distinguish real and forged signals. Statistical authenticity of a generated sample is improved through joint time-frequency domain feature discrimination, a physical constraint module is embedded in a generator, and channel physical characteristics are maintained through forced power normalization and a spectrum matching mechanism. A GAN framework with gradient penalty is adopted in the whole training process, and the training stability is ensured. And data sample generation when the nodes are quickly deployed to a new environment is facilitated.

The invention provides a millimeter wave radar miscellaneous point removing method and device based on frequency spectrum matching, a medium and a program product, and the method comprises the steps: obtaining a multi-channel echo signal of a millimeter wave radar array, and constructing an actual measurement angle spectrum reflecting the energy distribution in each direction through beam forming and angle traversal processing; carrying out peak value search on the actually measured angle spectrum, and determining an effective target angle corresponding to each power peak; according to geometric structure parameters and effective target angles of the millimeter-wave radar array, performing weighted stacking on each target response based on a theoretical echo signal model to obtain a theoretical angle spectrum; performing spectrum matching analysis on the theoretical angle spectrum and the actually measured angle spectrum, and calculating cross correlation coefficient distribution of the theoretical angle spectrum and the actually measured angle spectrum under different angle deviations; and when the cross-correlation obtains a maximum value at the zero offset and is greater than a threshold value, judging that the corresponding target is an effective target, otherwise, regarding the target as a miscellaneous point and removing the miscellaneous point. According to the method and the device, accurate identification of a real target is realized by utilizing the change characteristics of the cross correlation coefficient under different angle deviations.

This invention discloses a robust method, device, and storage medium for laser spot recognition and tracking in complex lighting environments, belonging to the field of computer vision and target detection. Addressing the problems of high false alarm rates and poor robustness in existing methods under interference such as strong ambient light, dynamic projector flicker, and specular reflection, this invention proposes a multi-layered processing framework of "detection-verification-tracking." First, adaptive threshold segmentation is performed on the current frame image to extract candidate regions with brightness higher than the background. Second, multi-dimensional feature verification is introduced. Based on the inherent physical properties of the laser spot (such as spectral matching degree, shape and size, and two-dimensional Gaussian intensity distribution) and temporal motion patterns, candidate regions are jointly verified and filtered to effectively remove various homogeneous and heterogeneous interferences. Finally, through a multi-target trajectory management module, the filtered candidate spots are correlated with historical trajectories (e.g., using the Hungarian algorithm), and combined with Kalman filter prediction and state machine-based trajectory lifecycle management (temporary, confirmed, lost, terminated), stable tracking and output of the laser spot are achieved. This invention does not rely on a large amount of training data, has high computational efficiency, and can significantly improve the accuracy and robustness of laser spot recognition under complex dynamic lighting conditions, providing a reliable front-end perception guarantee for vision-based high-precision laser spatial positioning and interaction systems.

A system and method using remote sensing instrument with hyper spectrum quantitatively measure metal dust elements in lubricating oil, which includes (not limited): Al, Cd, Cr, Cu, Fe, Pb, Mg, Mn, Mo, Ni, Ag, Sn, Ti, V, Zn, B (Boron, for Coolant), Ca (Calcium for water contaminant), and particle size, cone penetration, dropping point, steel mesh oil separation, moisture, PQ concentration, in few seconds. The instrument integrates near-field communication (NFC), Internet of Thing (IoT), Cloud computing, spectral matching and other data processing, and application software forming a system to easily operated and build a model enable self-learning to improve precision through collection accumulation. With the system, the instrument as FIG. 1 can provide comprehensive on-site analysis enable preventive maintenance of mission critical engine and rotating equipment. The characteristics of the system are easy to operate, get result quickly, and self-learning to improve precision.

The application provides a millimeter wave radar clutter point removal method and device based on spectrum matching, a medium and a program product. The method comprises the following steps: acquiring a multi-channel echo signal, constructing a measured angle spectrum reflecting the energy distribution of each direction through beam forming and angle traversal processing; determining each effective target angle of the measured angle spectrum; based on the geometric structure parameters of the millimeter wave radar array and the effective target angle, weighting and superimposing each target response based on a theoretical echo signal model to obtain a theoretical angle spectrum; performing spectrum matching analysis on the theoretical angle spectrum and the measured angle spectrum, performing point-by-point translation processing on the theoretical angle spectrum along the angle axis, and calculating the cross-correlation coefficient distribution of the two at different angle offsets; when the cross-correlation reaches a maximum value at zero angle offset and is greater than a threshold value, it is determined that the corresponding measured angle spectrum is effective, otherwise it is regarded as a clutter point and removed. The application realizes accurate identification of real targets by using the change characteristics of the cross-correlation coefficient at different angle offsets.

The invention relates to the field of geophysical exploration, in particular to a nuclear magnetic logging data inversion system and method based on semi-supervised adversarial learning. Comprising a T2 spectrum reconstructor which is used for training by adopting a label logging data set to obtain a nonlinear mapping relation between a spin echo signal and a preliminary T2 spectrum, and processing a non-label logging data set according to the nonlinear mapping relation to generate a predicted T2 spectrum; the false label selector generates the confidence coefficient of the predicted T2 spectrum according to a false matching data set obtained by matching the actual spin echo signal with the predicted T2 spectrum and a labeled logging data set, and selects the predicted T2 spectrum as a false label according to the confidence coefficient; mixing a pseudo label and a predicted T2 spectrum, and taking the mixture, a label logging data set and a non-label logging data set as a training set of a T2 spectrum reconstructor; and the T2 spectrum reconstructor is trained by the training set to obtain a nonlinear mapping relation between the actual spin echo signal and the predicted T2 spectrum. According to the invention, high-precision and high-efficiency intelligent nuclear magnetic logging data T2 spectrum inversion under a limited label condition is realized.

The invention provides an earth surface deformation measurement method and system based on different-satellite different-imaging-mode SAR data. The earth surface deformation measurement method comprises the steps of obtaining different-satellite different-imaging-mode SAR image data; registering the SAR image data to obtain a main and auxiliary image slope distance difference, a resampled main image and a registered auxiliary image; obtaining a main image and an auxiliary image after down-sampling; and performing interference processing to obtain and output earth surface deformation information. According to the method, an end-to-end processing framework of different-satellite different-imaging-mode SAR data is constructed, the risk of spatial baseline decoherence of a different-satellite different-imaging-mode SAR system is effectively reduced, and space-time correlation of different-satellite data is ensured; through full-link innovation of track design, joint registration, spectrum matching and terrain residual suppression, the bottleneck of cross-platform data interference incoherence is broken through, the problem of registration misalignment caused by different imaging mode data spectrum offset and resolution difference is solved, and a systematic solution is provided for high-frequency and high-precision surface deformation monitoring. Through a terrain phase separation technology, a terrain residual error is effectively inhibited.

The invention describes an unattended surveillance node and a method for identifying signal-emitting intruders across multiple physical domains. The system uses a hierarchical triggering architecture where mechanical disturbance sensors (acoustic and seismic) operate as low-power logic triggers. Upon detecting a mechanical signature matching an intrusion profile, the system activates a wide-spectrum radio frequency (RF) analysis. The key innovation lies in a correlational fusion engine that validates the intrusion only if there is a rhythmic or spectral match between the mechanical pressure fluctuations and the detected electromagnetic emissions, allowing the classification of the source (biological, mechanical, or electronic) with high immunity to environmental false alarms.

A concept is presented for adapting a predefined first mesh topology representing an organ to a different second mesh topology of the organ. One such concept includes identifying a correspondence between the first mesh topology and the second mesh topology based on a spectral matching of the first mesh topology and the second mesh topology. The predefined first mesh topology is aligned with the second mesh topology based on the identified correspondence between the first mesh topology and the second mesh topology.

This application provides a method for constructing a mainshock and aftershock sequence and a method for assessing damage to an intake tower. The method includes: obtaining an unfavorable mainshock and aftershock sequence based on ground motion data, and a database of near-field measured mainshocks, near-field measured aftershocks, far-field measured mainshocks, and far-field measured aftershocks; constructing multiple input conditions using the data from the databases and performing unified amplitude modulation processing to obtain the target mainshock and target aftershocks; determining the mainshock spectrum value of the target mainshock and the aftershock spectrum value of the target aftershock; determining the sum of squared residuals (SSE) between the mainshock spectrum value and the aftershock spectrum value, and a scaling factor. k ; Combine the sum of squared residuals (SSE) with the scaling factor k The mainshock spectrum and aftershock spectrum values ​​with a difference less than a preset difference threshold are used as matching results, thereby constructing a mainshock and aftershock combination with good spectral consistency; this application uses an improved spectrum matching method to construct a mainshock and aftershock combination with good comparability.

The application discloses a laser radar, which comprises a narrow-line-width multi-longitudinal-mode pulse laser, a photoelectric detector, a perforated mirror beam receiving system, a high-resolution narrow-line-width light splitting device and a plurality of single-photon detectors. The output beam of the narrow-line-width multi-longitudinal-mode pulse laser is emitted through the perforated mirror, and the photoelectric detector serves as a local signal. The echo signal photons reflected by a target are received through the receiving system, and are subjected to angle selection and spectral matching filtering through the light splitting device, and finally subjected to logical AND gate operation through the plurality of single-photon detectors, so that the final signal is simultaneously input into a time-correlation single-photon counter. The working performance of the photon counting laser radar in a strong noise environment is effectively improved, the false alarm probability of the laser radar is greatly reduced under the condition of ensuring the detection probability, the grating combined with the plurality of detectors can greatly reduce the background noise, the signal-to-noise ratio of the echo signal is greatly improved, and the detection capability of the laser radar is improved.

The invention discloses a hyperspectral and multispectral data fusion method of a Transform + KAN network. The hyperspectral and multispectral data fusion method comprises the following steps: S1, constructing a training data set: acquiring high-resolution and hyperspectral reference data HR-HSI-Benchmark, low-resolution and hyperspectral input data LR-HSI-Input to be fused and high-resolution and multispectral input data HR-MS-Input to be fused; after radiation correction, geometric correction, spectrum matching and down-sampling processing are carried out on the high-resolution hyperspectral reference data, low-resolution hyperspectral data LR-HSI-Train for training and high-resolution multispectral data HR-MSI-Train for training are generated; the LR-HSI-Train and the HR-MSI-Train are combined to form a low-resolution hyperspectral-high-resolution multispectral-high-resolution hyperspectral reference data training sample set; s2, constructing a multispectral and hyperspectral fusion network; s3, performing data amplification on the training sample set, and inputting the training sample set into a multispectral hyperspectral fusion network to obtain a trained fusion model; s4, registering the hyperspectral and multispectral data to be fused; and S5, image fusion and output.

The invention provides a morphology parameter measurement method, which comprises the following steps of: constructing a first weight vector set comprising a plurality of weight vectors for a first spectrum matching function, performing spectrum matching on a reference spectrum by using the first spectrum matching function for each weight vector in the first weight vector set, and obtaining a value of a corresponding loss function according to a matching result, and obtaining a first loss function set. Afterwards, a minimum loss function is obtained, because a matching result corresponding to the minimum loss function is more accurate, a weight vector corresponding to the minimum loss function is used as a target weight vector, a second spectrum matching function is constructed by using the target weight vector, the measurement spectrum and the theoretical spectrum, and a morphology parameter measurement value is obtained. The reference sample and the sample to be measured comprise the same periodic structure, so that the target weight vector determined by the reference spectrum of the reference sample can be used for obtaining a more matched theoretical spectrum for the measurement spectrum, and further more accurate morphology parameter measurement values of the sample to be measured can be obtained.

The present invention discloses a two-dimensional parameter estimation method for broadband incoherent distributed sources based on spectrum matching, comprising the following steps: decomposing the direction of arrival (DOA) estimation problem of a two-dimensional broadband incoherent distributed source into two one-dimensional estimation problems by decomposing an L-shaped array into two orthogonal uniform linear arrays; recovering two angular frequency spatial spectrum matrices from the array receiving covariance matrix based on the received signal in combination with a low-rank recovery method; and extracting two sets of key parameters from the two angular frequency spatial spectrum matrices for matching; proposing a spectrum matching-based method for determining the two-dimensional DOA of a broadband incoherent distributed source based on the frequency domain parameter invariance of the distributed source in space; and directly estimating the joint angle and spectrum to obtain the joint parameters. The present invention also eliminates the need for prior knowledge of the distributed source and can directly estimate the two-dimensional angle based on the angular frequency spatial spectrum.

The present application relates to a kind of quantitative methods of spectral imager, specifically, a kind of high spectral remote sensor cross calibration method based on spectral matching factor, existing high spectral remote sensor only relies on single spectral channel in cross calibration process, leading to the technical problem of poor cross calibration precision.This method includes the following steps:1】Obtain spectral channel characteristics;2】Calculate simulated spectral radiance;3】Calculate spectral matching factor;4】Obtain cross calibration test area;5】Calculate the real spectral radiance of each spectral channel of reference remote sensor B;6】Obtain weight coefficient;7】Obtain final spectral matching factor;8】Calculate the real spectral radiance of each spectral channel of to-be-calibrated remote sensor A;9】Calculate the radiation calibration coefficient of each spectral channel of to-be-calibrated remote sensor A, complete the cross calibration of high spectral remote sensor of spectral matching factor.This method reduces error, improves data accuracy and reliability.

This invention discloses a method and apparatus for evaluating on-board very long wave (VLW) spectral calibration based on a radiative transfer model, belonging to the field of remote sensing optics technology. This invention targets a wide-band VLW infrared hyperspectral sensor covering the wavelength range of 12-16 μm, analyzing the atmospheric absorption characteristics and sensor imaging spectral characteristics within this spectral range. The spectral calibration accuracy is evaluated through physical model simulation, including: simulation using the MODTRAN radiative transfer model; construction of a spectral drift model; solving for the center wavelength shift based on a spectral matching combined with an iterative optimization algorithm; and uncertainty analysis of the impact of factors such as the estimation error of the algorithm itself, sensor signal-to-noise ratio, inter-spectral response inconsistency, and atmospheric CO2 concentration on the estimation results. This invention is applied to evaluate the spectral calibration accuracy of on-orbit satellite data.

The application relates to a hyperspectral remote sensor cross-calibration method based on a dynamic Gaussian distribution, and solves the technical problem that the spectral matching factor obtained by the existing spectral matching factor calculation method is low in accuracy and precision, and it is difficult to improve the cross-calibration precision of the interference type hyperspectral remote sensor. The method comprises the following steps: 1, determining a cross-calibration test area; 2, calculating a simulated spectral radiance; 3, calculating the spectral matching factor of a plurality of spectral channels of a to-be-calibrated remote sensor B which have a greater influence on a to-be-calibrated remote sensor A based on a dynamic Gaussian distribution; 4, obtaining the final spectral matching factor of each spectral channel of the to-be-calibrated remote sensor A; 5, calculating the real spectral radiance of the plurality of spectral channels of the to-be-calibrated remote sensor B which have a greater influence on the to-be-calibrated remote sensor A; 6, calculating the real spectral radiance of the to-be-calibrated remote sensor A; 7, calculating the radiation calibration coefficient of each spectral channel of the to-be-calibrated remote sensor A, and completing the cross-calibration of the hyperspectral remote sensor.

The invention discloses a broadband incoherent distributed source two-dimensional parameter estimation method based on spectrum matching, and the method comprises the following steps: decomposing an L-shaped array into two orthogonal uniform linear arrays, decomposing a direction-of-arrival angle estimation problem of a two-dimensional broadband incoherent distributed source into two one-dimensional estimation problems, and based on a received signal, carrying out the estimation of a direction-of-arrival angle of the two-dimensional broadband incoherent distributed source; in combination with a low-rank recovery method, recovering two angular frequency spatial spectrum matrixes from the array receiving covariance matrix, and extracting two groups of key parameters from the two angular frequency spatial spectrum matrixes for matching; according to the frequency domain parameter invariance of the distribution source in the space, proposing a method based on spectrum matching to determine the two-dimensional direction of arrival angle of the broadband incoherent distribution source; the joint angle and the frequency spectrum are directly estimated to obtain the joint parameter. According to the method, the two-dimensional angle can be directly estimated according to the angular frequency spatial spectrum without priori knowledge of a distribution source.

This application provides a blockchain-based intelligent processing method, system, electronic device, and storage medium for cross-border e-commerce orders, relating to the field of blockchain technology. The method includes: receiving order information, material spectral data, and genuine product feature data to construct a multi-dimensional knowledge graph; triggering hyperspectral scanning based on product identification to generate material composition data, and performing spectral matching with the knowledge graph to obtain a matching degree; simultaneously extracting key region feature sets from the hyperspectral image using the U-Net image segmentation algorithm, and comparing them with an anti-counterfeiting feature library to obtain spatial similarity; when both indicators meet customs thresholds, activating the target smart contract based on an emergency identifier and generating an execution record; finally, encapsulating the verification data into a verification package, generating a hash value, and writing it to the blockchain, establishing a bidirectional index with the order through the smart contract to achieve state synchronization. This application improves the automation level and data reliability of cross-border e-commerce product customs clearance verification.

The invention discloses an AI intelligence-based English explanation system for tourism. The system comprises an environmental noise spectrum identification module, which is used for collecting exhibition hall sound signals to extract a spectrum, calculating an Euclidean distance with a preset scene feature library, adjusting an attenuation coefficient according to a comparison result, and generating an environmental noise attenuation coefficient value. According to the method, the noise suppression coefficient is adaptively adjusted through dynamic spectrum matching, the signal-to-noise ratio of a complex sound field is improved, multi-channel signal cross-correlation operation is combined with array model conversion, and millimeter-level sound source positioning and beam forming enhanced voice space separation are achieved. Knowledge graph nodes construct a dynamic association model based on multi-dimensional semantic similarity, tourist density weights are corrected in real time, explanation content is accurately adapted, a color parameterized template and artistic comments are spliced to reserve term accuracy, cross-culture text infection is enhanced, and a binaural time difference and sound level difference parameter synthesis technology endows voice with three-dimensional space characteristics, so that the voice quality is improved. And the auditory immersion and the orientation identification degree are enhanced.

A metasurface inverse design method based on a generative adversarial network (GAN) controlled by an anchor network is proposed. The method involves collecting metasurface design parameters and spectral response data, preprocessing them, and encoding them into a format that can be handled by a deep learning model. K-means clustering is used to analyze the training data to generate structured control vectors. An AcGAN framework is constructed, integrating the spectral similarity metric SOC into the loss functions of the generator and discriminator to enhance spectral matching accuracy. Through adversarial training, the generator is trained to produce designs that conform to the target spectral characteristics, while the discriminator and the AnchorNet are optimized for rapid prediction and evaluation of the spectral response. The trained generator, combined with the control vectors and random noise, generates candidate designs. The candidate designs are then screened to select the optimal design. This invention effectively improves the spectral control accuracy and design efficiency of metasurface design, solving the challenges faced by traditional methods in handling complex spectral requirements.

This application relates to a seismic motion simulation method that matches long-period response spectra. By extending the response spectrum analysis period to 20 seconds and constructing a low-frequency characteristic seismic motion database, high-precision characterization of seismic motion characteristics across the entire frequency band from 0.03 to 20 seconds is achieved. This method overcomes the limitation of traditional models that decay too quickly in the amplitude descent segment, providing a seismic motion input solution for the seismic design of long-period structures such as super high-rise buildings that combines broadband spectral matching accuracy with fidelity to the physical characteristics of the geological environment.

The invention discloses a mineral anomaly identification method and system based on spectral remote sensing, and relates to the technical field of mineral anomaly identification, and the method comprises the following steps: obtaining hyperspectral remote sensing image data of a target area; performing atmospheric correction and geometric correction on the hyperspectral remote sensing image data to obtain surface reflectance data, and extracting a spectral reflectance curve; performing weak signal enhancement processing on the spectral reflectivity curve to obtain an enhanced spectral curve, and calculating a spectral matching degree between the enhanced spectral curve and each mineral end member in a standard mineral end member spectrum library; the spectrum matching degree is analyzed, and the mineral abnormal condition is judged according to the analysis result. According to the method, a systematic weak signal enhancement technology system is constructed, so that the technical problem of low recognition accuracy of a traditional method in a deep ore body and a low-grade mineralized area is solved.

The invention provides a mineral spectrum global matching method and device based on multi-band feature vectors. The method comprises the following steps: carrying out continuum removal processing on an input spectrum to obtain a target spectrum; determining an effective spectrum band and a central wavelength thereof of the target spectrum based on a preset threshold strategy; determining an absorption characteristic parameter corresponding to the effective spectrum band; constructing and determining a first target comprehensive absorption feature vector according to the absorption feature parameter of the effective spectrum band; determining a second target comprehensive absorption feature vector according to the absorption feature parameter of the effective band of the preset reference spectrum; determining a basic Euclidean distance and an augmented Euclidean distance of the first target comprehensive absorption feature vector and the second target comprehensive absorption feature vector; and determining the mineral corresponding to the reference spectrum with the minimum basic Euclidean distance or augmented Euclidean distance as a target mineral. Through the above method, the multi-band feature vectors are fused and mineral identification is performed based on the basic Euclidean distance or the augmented Euclidean distance, so that the accuracy of mineral identification can be improved.

The application provides a kind of radiation calibration method and device for sub-meter optical remote sensing satellite data, it is related to remote sensing technical field, including: based on global atmospheric parameter observation site, extract and establish pseudo stable ground library;For each kind of pseudo stable ground, establish BRDF model and spectral curve reconstruction model, calculate the spectral matching factor of reference sensor and to be calibrated sensor;Based on the to-be-calibrated image of pseudo stable ground, the theoretical ground reflectivity of to-be-calibrated image is calculated using spectral matching factor and BRDF model;Based on the theoretical ground reflectivity, the observation angle and the measured atmospheric parameters of site, the theoretical apparent radiance of to-be-calibrated image is calculated;The theoretical apparent radiance of to-be-calibrated image on time series is fitted with DN value, the variation trend curve of cross calibration coefficient is determined, to carry out radiation calibration for sub-meter optical remote sensing satellite data.The application can realize high frequency, high precision cross calibration of sub-meter optical remote sensing satellite data.