25 results about "Spectral mapping" patented technology

The invention discloses a full-automatic, rapid and high-precision three-dimensional reconstruction method and system, and relates to the field of three-dimensional dynamic reconstruction. According to the method, the spectrum-point cloud fusion model is constructed, the Monte Carlo ray tracing is introduced to calculate the dynamic reflectivity field, and the Fresnel formula and the dynamic ray tracing technology are combined, so that the reflectivity calculation is more in line with the physical law, and the accuracy of spectrum mapping is improved. The spectral data is optimized and adjusted based on an error correction matrix through a spectral error compensation network, so that error accumulation caused by factors such as a measurement angle and surface roughness is reduced; besides, in the three-dimensional reconstruction process, the continuity and spectrum consistency of the three-dimensional model are improved through a Poisson surface reconstruction and spectrum reflectivity driven grid optimization method according to spectrum enhanced point cloud data.

The invention discloses a park digital twinning modeling method based on a generative AI technology, and relates to the technical field of digital twinning modeling, and the method comprises the steps: carrying out the alignment of point cloud, image and state data, completing the preprocessing through topological adaptive filtering and multi-resolution voxelization, frequency domain harmonic fusion and hypersurface texture excitation and time delay coupling fuzzy clustering, and obtaining a digital twinning modeling result; constructing a cross-modal spine network driven by a holographic entropy film, generating a hierarchical token through graph attention, and inputting a surge tuned diffusion converter model for iterative denoising and focus decoding to obtain a three-dimensional fragment; and fusing the fragments in a voxel space by using a cross attention kernel, and adaptively updating parameters through an entropy pulse closed loop until errors converge, so as to generate a high-precision digital twinborn model. A cross-modal semantic network is constructed by constructing spectral mapping and a holographic entropy film, a three-dimensional fragment is efficiently generated in a self-adaptive surge tuned diffusion converter framework through focus fusion, and the cross-modal semantic coupling efficiency, the generative reasoning convergence speed and the multi-fragment voxel fusion continuity are improved.

The invention provides a display control method and system for a liquid crystal display screen, and the method comprises the steps: carrying out the nonlinear spectrum mapping processing of an input RGB drive value, and obtaining a theoretical emission spectrum through combining with the response characteristics of liquid crystal molecules; calculating a reflection spectrum component in the multilayer structure according to the ambient light spectrum, and superposing the reflection spectrum component with the theoretical emission spectrum to obtain an actual observation spectrum; calculating a difference value between the actual observation spectrum and the target spectrum, and converting the difference value into an RGB correction value through a constraint optimization algorithm to obtain an RGB driving value after ambient light compensation; and calculating actual white point color coordinates, and adjusting the actual white point color coordinates to a standard white point position through a progressive white point migration algorithm. Display parameters can be dynamically adjusted according to real-time ambient light conditions, the influence of ambient light reflection on colors is effectively compensated, meanwhile, color temperature consistency and color accuracy are ensured through white point correction, the display quality under different ambient lights is remarkably improved, and the visual experience of a user is improved.

This invention discloses a fully automated, high-speed, and highly accurate three-dimensional reconstruction method and system, relating to the field of three-dimensional dynamic reconstruction. The invention introduces the construction of a spectral-point cloud fusion model and the calculation of a dynamic reflectance field by Monte Carlo ray tracing. By combining this with Fresnel's equation and dynamic ray tracing techniques, the reflectance calculation conforms to physical laws, thereby improving the accuracy of spectral mapping. Furthermore, the invention optimizes spectral data based on an error correction matrix using a spectral error compensation network, reducing the accumulation of errors caused by factors such as measurement angle and surface roughness. Moreover, in the three-dimensional reconstruction process, the invention improves the continuity and spectral consistency of the three-dimensional model by using Poisson surface reconstruction and a spectral reflectance-driven mesh optimization method based on spectrally enhanced point cloud data.

The invention provides a spectrum screening method and device for deep sea rare earth element LIBS quantitative detection and a storage medium, rare earth sample LIBS spectrums under different conditions are collected, and the spectrums are mapped to a CIE L * a * b * color space in combination with spectral intensity and a CIE standard chromaticity observer spectrum tristimulus value function to realize characterization of plasma colors. Quantitative analysis is carried out on LIBS spectrums under different conditions, the condition with the optimal quantitative effect is taken as a standard, a central area of a spectrum mapping point at the moment is delimited, the spectrum mapping points under other conditions are constrained by utilizing the standard area, spectrum screening based on plasma luminescence characteristics is realized, and the linear fitting degree of a spectrum calibration curve under other conditions is improved. According to the method, a complicated data processing means is not needed, the mapping from the spectrum to the color space can be realized only by combining a mathematical formula for operation, the operation time is short, the required data volume is small, the applicability is high, and a result has an actual physical meaning and has interpretability.

The invention relates to a method for improving agricultural pest recognition or classification model performance based on a multispectral image, and relates to the technical field of agricultural pest model performance improvement. The invention discloses a method for improving the performance of an agricultural pest recognition or classification model based on a multispectral image. A spectrum mapping module is added in an agricultural disease and insect pest identification or classification model, a spectrum key quantitative analysis method based on a deep learning model is adopted, and the utilization rate of the model for each spectrum band characteristic is improved by using a channel attention mechanism, so that the model shows good performance in disease grade classification; in addition, an effective technical support is provided for intelligent disease monitoring and management in the agricultural field, practical application of multispectral image processing is promoted, and an important foundation is laid for development of a disease detection technology in the future.

The present invention discloses a method for spectral reconstruction of UAV RGB images based on band difference. First, collect UAV aerial RGB images and hyperspectral auxiliary images to construct a dataset with RGB and hyperspectral images. Secondly, use band difference and superpixel segmentation to generate training data. Then, establish a spectral mapping model based on a deep network and train to obtain the spectral mapping relationship from RGB images to band difference sequences. Further, use the spectral mapping relationship obtained by training the model to perform spectral reconstruction on any RGB image collected by the UAV to obtain the difference sequence of the target image. Finally, perform inverse difference processing on the difference sequence to obtain the reconstructed target image. By introducing the difference idea to define the relationship between adjacent bands of the hyperspectral auxiliary image, the present invention can represent both the intra-band information and the inter-band correlation information of the hyperspectral auxiliary image, and can effectively improve the quality of spectral reconstruction of RGB images.

This application provides a machine learning-based method and system for optimizing the photothermal conversion efficiency of solar desalination in seawater. The method includes: firstly, collecting solar spectral distribution data, seawater turbidity spectral data, real-time heat flow data, and semiconductor nanomaterial data for the target sea area; and constructing a material spectral mapping library based on the light absorption characteristics of the nanomaterials. By coupling analysis of the solar spectrum and seawater turbidity spectrum to form environmental constraints, a Bayesian optimization model is used in conjunction with the material spectral mapping library and real-time heat flow data to generate the optimal nanomaterial mixing ratio under these constraints. Subsequently, a deep residual network is used to predict the interfacial evaporation rate and obtain the optimal material morphology configuration scheme, which is finally applied to the preparation of semiconductor nanomaterials. This application improves the photothermal conversion efficiency of solar desalination systems in high-turbidity seawater environments.

The invention discloses a multi-parameter integrated photoelectric detection sensing system based on a wide spectrum response photoelectric material. The system relates to the photoelectric sensing technology and comprises a wide-spectrum photoelectric material pixel array, a reading circuit and a processor. The pixel array is composed of a plurality of detection units, and each detection unit comprises a plurality of sub-pixels with wavelength gradients. The readout circuit constructs spectral response distribution by collecting sub-pixel signals to determine the central wavelength and calculates the total photocurrent of the detection unit. And the processor inversely calculates the in-situ temperature of the detection unit based on the central wavelength, and performs closed-loop calibration on the total light current by using the in-situ temperature. Through the spatial spectrum mapping mechanism of the sub-pixel array and the internal self-consistent calibration algorithm, the problem of inaccurate measurement caused by ex-situ temperature compensation in the prior art is solved, and the measurement accuracy and environmental adaptability of the system are improved.

The invention provides a light-emitting diode chip transfer method and device and a display screen, and the method comprises the steps: detecting a middle carrier plate carrying a plurality of light-emitting diode chips, determining a spectrum mapping graph corresponding to the middle carrier plate, and enabling the spectrum mapping graph to be used for reflecting the light-emitting parameters corresponding to the plurality of light-emitting diode chips, the light-emitting parameters comprise light-emitting wavelength and / or light-emitting brightness; performing rasterization processing on the spectral mapping graph to generate a processed spectral mapping graph; for each grid in the processed spectrum mapping graph, determining a luminescence index corresponding to the grid according to luminescence parameters of a plurality of light emitting diode chips in the grid; and transferring the plurality of light emitting diode chips in the grids on the middle carrier plate to a target carrier plate corresponding to the index range according to the index range to which the light emitting index corresponding to each grid belongs. According to the invention, the target carrier plate with consistent wavelength and / or consistent brightness can be obtained, so that the target carrier plate can be utilized to generate a display screen with a better display effect.

The invention discloses a meridian interaction science popularization method and system based on multi-modal interaction and RGB spectral mapping, and relates to the technical field of traditional Chinese medicine science popularization education and human-computer interaction. According to the meridian interaction science popularization method and system based on multi-mode interaction and RGB spectral mapping, through three interaction modes of position sensing, gesture recognition and a touch screen, operation habits of different crowds are adapted, interaction is more immersive, gesture recognition is accurate, meanwhile, trigger deviation is avoided, traditional static display limitation is broken through, and the science popularization effect is better. Exclusive RGB colors are distributed to 14 meridians and collaterals according to the five-element theory, dynamic lighting effects are matched, abstract meridians and collaterals are converted into visual symbols, the cognition threshold is reduced, the voice broadcast function is synchronized, meridian knowledge is transmitted in multiple dimensions, small-week-day and large-week-day achievements of the 14 meridian and collateral modes are matched, the lighting effects and voice feedback are combined, and therefore the effect of the meridian and collateral channels is achieved. Public health cognition is enhanced, and interest and recognition of the public on traditional Chinese medicine culture are stimulated.

The present invention provides a spectral screening method, equipment and storage medium for LIBS quantitative detection of deep-sea rare earth elements. The LIBS spectra of rare earth samples under different conditions are collected, and the spectrum is mapped to the CIE L*a*b* color space by combining the spectral intensity and the CIE standard colorimetric observer spectrum tristimulus value function to achieve the characterization of the plasma color. The LIBS spectra under different conditions are quantitatively analyzed. The central area of ​​the spectral mapping point at this time is delineated based on the condition with the best quantitative effect. The standard area is used to constrain the spectral mapping points under other conditions to achieve spectral screening based on the luminescence characteristics of the plasma and improve the linear fit of the spectral calibration curve under other conditions. This method does not require complex data processing means. It only needs to be combined with mathematical formulas for calculation to achieve the mapping of the spectrum to the color space. It has a short running time, requires a small amount of data, is highly applicable, and the results have actual physical meaning and are interpretable.

This invention provides a display control method and system for a liquid crystal display (LCD). The method includes: performing nonlinear spectral mapping processing on the input RGB driving values ​​and combining this with the response characteristics of liquid crystal molecules to obtain the theoretical emission spectrum; calculating the reflection spectral components in the multilayer structure based on the ambient light spectrum and superimposing them with the theoretical emission spectrum to obtain the actual observed spectrum; calculating the difference between the actual observed spectrum and the target spectrum, converting it into an RGB correction value through a constraint optimization algorithm to obtain the ambient light compensated RGB driving value; and calculating the actual white point color coordinates and adjusting them to the standard white point position through a progressive white point migration algorithm. This invention can dynamically adjust display parameters according to real-time ambient light conditions, effectively compensate for the influence of ambient light reflection on color, and ensure color temperature consistency and color accuracy through white point correction, significantly improving display quality under different ambient light conditions and enhancing the user's visual experience.

The invention provides a visible light and thermal infrared remote sensing image registration method, and relates to the technical field of remote sensing data processing and analysis. The registration method of the visible light and thermal infrared remote sensing images comprises the following steps: performing spectral mapping on a visible light band image of a target area by using a spectral mapping model to obtain a mapped thermal infrared band image; performing coarse registration on the thermal infrared band image of the target area and the mapping thermal infrared band image to obtain a coarse registration thermal infrared band image; performing feature matching on the mapping thermal infrared band image and the coarse registration thermal infrared band image to determine a feature pair set of the target area; and registering the visible light band image and the thermal infrared band image according to the feature pair set. The spectral mapping model is constructed by using the spectral information of different ground objects in the visible light and thermal infrared bands as priori knowledge, and effective conversion from the visible light reference band spectral feature to the thermal infrared band can be completed.

The invention relates to a technique for generating a medical image from a spectral map produced from multispectral raw medical imaging data acquired using an energy-resolved medical imaging technique. A computer-implemented method (100), performed by a computing device (200), comprises receiving (S102) multispectral raw medical imaging data acquired using an energy-resolved medical imaging technique. An analysis scope of the received (S102) multispectral raw medical imaging data is determined (S106). At least one spectral map to be generated from the received (S102) multispectral raw medical imaging data is selected (S108). The selection (S108) is based on the determined (S106) analysis scope. The at least one selected (S108) spectral map is generated (S110).The received (S102) multispectral medical raw imaging data are processed (S112) to generate (S114) at least one medical image based on the at least one generated (S110) spectral map.

This invention relates to the field of automated laser chip heating processing control technology, and in particular to a laser heating rework system. The system includes a spectral mapping module that obtains wavelength drift data by analyzing transient photoluminescence spectra collected by a fluorescence thermometer, and uses this data to identify the coordinates of poorly heated locations to generate a transient thermal state matrix. A tensor calculation module combines the microscopic orientation data of the material's crystal lattice structure to calculate the thermal conductivity coefficient and generate a thermal conductivity tensor matrix. A polarization synthesis module retrieves photon absorption cross-section difference parameters to inversely deduce a physical compensation array and generate a spatial polarization phase map. A polarization modulation device generates a non-uniform vector polarized light field based on this map, driving a laser output device to inject the light field into the coordinates of the poorly heated location to perform targeted heat injection. This invention utilizes the difference in physical absorption rate of photons with different polarization states in the underlying crystal lattice arrangement to correct local thermal distribution anomalies while achieving zero thermal damage overflow in the surrounding area, thus improving the overall processing yield.

The invention provides a spectral information processing method, a spectral information processing device and electronic equipment. The spectral information processing method comprises the following steps: respectively acquiring detection signals acquired in different spectral ranges and corresponding pre-training spectral mapping models; performing feature matching processing on each detection signal through a spectrum mapping model to generate a plurality of pieces of first spectrum information corresponding to each detection signal; performing calibration processing on each piece of first spectral information to correct the characteristic deviation, and obtaining corrected spectral information corresponding to each piece of first spectral information; and splicing the multiple pieces of corrected spectrum information according to a preset target spectrum coverage rule to form continuous spectrum information. The problem of how to combine and integrate the spectrums in each spectral range into an ultra-wideband spectrum when a computational spectrometer is adopted is solved.

The invention discloses a depth non-rigid three-dimensional shape corresponding method, system and device and a medium. The method comprises the steps that a first feature vector matrix and a second feature vector matrix are determined; in the micro-iteration process, based on the current spectral domain mapping matrix, the first eigenvector matrix and the second eigenvector matrix, a symbolized soft mapping operator based on a kernel function and used for describing the corresponding relation between the vertexes is constructed, the symbolized soft mapping operator is re-projected back to the spectral space, and intermediate state spectral mapping estimation is obtained; performing multi-channel filtering processing on the intermediate state spectral mapping estimation to obtain a refined coupling mapping matrix, and taking the refined coupling mapping matrix as a current spectral domain mapping matrix of next iteration until a preset number of iterations is reached to obtain a target coupling mapping matrix; and converting the target coupling mapping matrix into a point-by-point mapping matrix to realize point-by-point correspondence between the source shape and the target shape. According to the invention, the accuracy corresponding to the deep non-rigid three-dimensional shape can be improved.

The invention relates to the technical field of ecological environment monitoring, in particular to an ecological environment monitoring device based on remote sensing images, which comprises a data acquisition and preprocessing module used for acquiring current and historical remote sensing images and meteorological data of a target area and constructing a time sequence data set; the land coverage identification module is used for identifying the land coverage type of the target area; the reference area selection module is used for delimiting a reference area in a preset range; the meteorological-spectral mapping model building module is used for building a mapping function from meteorological data to spectral reflection characteristics; the spectrum anomaly detection and region correction module is used for comparing the difference between the actual spectrum and the expected spectrum output by the model, and performing boundary cutting correction on an abnormal region to obtain a new target region after meteorological influence correction; and the ecological parameter inversion and output module is used for performing inversion and mapping of the ecological parameters. The adaptive capacity to different weather conditions can be improved, and the accuracy of a monitoring result is enhanced.

The present invention discloses a fully automatic, fast and high-precision three-dimensional reconstruction method and system, and relates to the field of three-dimensional dynamic reconstruction. The present invention constructs a spectral-point cloud fusion model, introduces Monte Carlo ray tracing to calculate the dynamic reflectivity field, and combines the Fresnel formula and dynamic ray tracing technology to make the reflectivity calculation more in line with physical laws, thereby improving the accuracy of spectral mapping. The present invention also optimizes and adjusts the spectral data based on the error correction matrix through a spectral error compensation network, thereby reducing the error accumulation caused by factors such as measurement angle and surface roughness; in addition, in the three-dimensional reconstruction process, the present invention enhances the point cloud data according to the spectrum, and improves the continuity and spectral consistency of the three-dimensional model through Poisson surface reconstruction and spectral reflectivity-driven grid optimization method.

The present invention discloses a spectral domain mapping method for satellite remote sensing multispectral images based on a generative adversarial model, which is characterized by the following steps: S1, collecting multiple groups of matched multispectral remote sensing images, ground object classification maps, and numerical feature product maps and performing preprocessing; S2, determining the input matrix x of the generative adversarial model according to the preprocessed multispectral remote sensing images, ground object classification maps, and numerical feature product maps, and determining the size of the output matrix y of the generative adversarial model; S3, respectively processing each group of matched image data and then sequentially inputting them into the generative adversarial model for training; S4, inputting new matched image data and using the trained generative adversarial model to complete the spectral domain migration of satellite remote sensing multispectral images. The present invention realizes the direct spectral band migration from pixel to pixel of the image, and enhances the certainty of the simulation spectral mapping of the remote sensing spectrum.

The invention discloses an intelligent agricultural information monitoring system and method based on big data, and relates to the technical field of agricultural information processing. The intelligent agricultural information monitoring method based on big data comprises the following steps: acquiring soil state time sequence data and crop spectral image time sequence data of a target greenhouse; performing hierarchical evaluation processing on the soil state time sequence data to obtain a soil moisture deficiency evaluation value; based on a pre-trained canopy moisture time sequence response model, performing time sequence spectrum mapping processing on the crop spectral image time sequence data to obtain a crop moisture loss evaluation value; according to the method, irrigation processing is performed on the target greenhouse through the irrigation urgent evaluation value, so that the state of water stress can be captured more comprehensively, water stress aggravation or water resource waste can be effectively avoided, and the irrigation demand evaluation method and the irrigation demand evaluation system have the advantages that the irrigation urgent evaluation value and the soil water shortage evaluation value are input into the pre-constructed irrigation demand evaluation model; crop water supply and demand balance is guaranteed, and the refined level of planting management is improved.

This application discloses a panoramic spectral reconstruction imaging method, comprising: lowering a device with coaxially integrated specific optical components into a borehole; illuminating the borehole wall with a light source; collecting the reflected light panoramically using a conical reflector; encoding the light spectrally through an array of annular gradient filters; and receiving the light with a detector; based on a pre-calibrated spatial-spectral mapping model, processing the detector image to simultaneously reconstruct a panoramic unfolded image and a narrowband image of the borehole wall, thereby generating a spectral image cube containing spatial and spectral information; this invention enables the one-time, in-situ acquisition of a panoramic image of the borehole wall and the continuous spectral curves of each point thereon within the borehole, achieving spectral collaborative perception of rock morphology and mineral composition.

The invention provides a generation method for accelerating a holographic video in 360-degree holographic display based on a rotation motion compensation algorithm of a segmented wave field. The method comprises the following steps: firstly, according to a given three-dimensional object, rotating to an mth reference visual angle along a horizontal direction, and rendering a three-dimensional scene under the visual angle; secondly, the three-dimensional object is divided into a plurality of basic objects in the depth direction (z-axis), each basic object comprises point cloud data with the depth range being L, and the corresponding central plane is located at the position where z = + / -nL; for each basic object, independently calculating wavefront distribution of the basic object on a holographic plane by adopting a traditional fast holographic algorithm to obtain a corresponding basic hologram; overlapping all the basic holograms to generate an mth period reference frame, and taking the mth period reference frame as an mkth frame in the holographic video at the same time; on the basis of the mth period reference frame, within a certain horizontal view angle interval, executing a rotation motion compensation operation based on frequency spectrum mapping on each basic hologram, generating a compensation frame, and performing compensation before and after the mkth frame; and repeating the above operations until all holographic frames are generated in a 360-degree full-view angle range, and finally outputting a holographic video sequence which is high in quality and frame rate and retains depth and motion parallax information. The algorithm has the advantages of high calculation speed, high-quality 360-degree holographic video generation and the like.

The present application belongs to the technical field of three-dimensional reconstruction, and particularly relates to a multi-view hyperspectral three-dimensional reconstruction and spectral mapping method. The present application establishes a unified geometric reference through explicit three-dimensional reconstruction, establishes a strict projection relationship between a space point and a hyperspectral pixel based on camera internal and external parameters, selects a target view angle in combination with a normal vector geometric constraint, and finally realizes accurate, stable and physically credible mapping of hyperspectral data on a three-dimensional point cloud through nearest surface point discrimination and voxel grid constraint screening of target pixel points. The present application can strengthen the constraint relationship between a space point and a real surface position, inhibit mixed pixel interference and same-pixel multi-depth ambiguity, and further ensure the physical authenticity of the spectral mapping result.