133 results about "Remote sensing reflectance" patented technology

The invention discloses a hyperspectral water quality parameter inversion system based on a one-dimensional convolutional neural network. The system is used for measuring hyperspectral data of a waterbody at a selected point through a ground point spectrum actual measurement module; a water quality acquisition and analysis module is used for acquiring a water body sample at a selected point and analyzing to obtain water quality parameter concentration; an inversion model construction module is used for training parameters of a one-dimensional convolutional neural network by taking all the spectral band information as input and taking water quality parameter concentration as output so as to fit a complex nonlinear relationship between the spectral band information and the water quality parameter concentration; a hyperspectral data acquisition module is used for acquiring a hyperspectral remote sensing image of a monitored water area and obtaining the remote sensing reflectivity of eachpoint spectral band; and a water quality parameter inversion module is used for taking the spectral band information of each point as input to obtain the water quality parameter concentration of eachpoint in the monitored water area through inversion, and meanwhile, the system does not need to add a band screening sub-module and fully utilizes all band information.

The invention provides an MODIS satellite high-precision monitoring method for chlorophyll-a in a eutrophic lake water body. The method comprises the following steps: screening a chlorophyll-a evaluation index (BNDBI) which is sensitive to the concentration change of chlorophyll-a and not affected by highly suspended matters, thin cloud and solar flares; acquiring the quantitative relationship between BNDBI and the concentration of chlorophyll-a on the basis of biological optical model simulation; acquiring a chlorophyll-a inversion algorithm based on ground measured spectrum (Rrs) and BNDBI by referring to water body spectral information and corresponding water body chlorophyll-a concentration of 2013-2014 lake field monitoring; acquiring the quantitative relationship between the ground monitoring and remote sensing reflectance ratio (Rrs) and Rrc after simulated Rayleigh scattering correction by simulating different aerosol types and thicknesses and different solar elevation angles, satellite observation angles and azimuth angles; and extending the chlorophyll-a inversion algorithm based on ground measured spectrum data to satellite image data after Rayleigh scattering correction. Based on the method, inter-annual and inter-monthly change rules and spatial distribution of chlorophyll-a in a eutrophic lake can be acquired accurately.

The invention discloses an analysis method of remote sensing inversion of water color parameters of inland class II water. According to the method, inherent optical quantity of the class II water is inverted with an improved QAA (quasi-analytical algorithm), and concentration inversion of the water color parameters such as chlorophyll a and suspended matters is realized based on the inverted inherent optical quantity of the class II water. The method comprises specific steps as follows: (1), data of remote sensing reflectance above the surface of the water is input, parameter inversion of the inherent optical quantity of the water is realized based on the improved QAA, and absorption coefficients and scattering coefficients of the water and absorption coefficients of phytoplankton are acquired; (2), concentration data of the chlorophyll a and concentration data of the suspended matters of the water at a sampling point are input, a chlorophyll a concentration quantitative inversion model is established according to the concentration data of chlorophyll a and the absorption coefficients of the phytoplankton, and a suspended matter concentration quantitative inversion model is established according to the concentration data of the suspended matters and the absorption coefficients of the water after removal of pure water; (3), hyperspectral data finishing atmospheric correction are input, and concentration inversion of the water color parameters of the inland class II water in a monitoring area is realized according to the step (1) and the step (2).

The invention discloses a lake water body blue-green algae abundance estimation method based on remote sensing. The method comprises the following steps: carrying out high-spectrum remote sensing measurement on a lake water body by utilizing the concentration ratio of phycocyanin and chlorophyll to represent the blue-green algae abundance, selecting the ratio of remote sensing reflective rates of the two specific wave lengths and a function thereof as characteristics of blue-green algae, establishing an estimation model of the lake water body blue-green algae abundance, and estimating the lake water body blue-green algae abundance based on remote sensing. The method provided by the invention adopts a surface feature spectrograph or an onboard high-spectrum imager to carry out high-spectrum remote sensing on the lake water body, and the remote sensing reflective rate required for estimation can be obtained through performing radiation correction and atmospheric correction on aviation satellite high-spectrum data. The method provided by the invention has the advantages that the accuracy is high, the model is simple, the remote sensing reflective rates of the two wave lengths are selected, not only can the estimation on the water body blue-green algae abundance measured by the surface feature spectrograph be realized, but also the estimation on the blue-green algae abundance by virtue of aviation high-spectrum remote-sensing images is realized.

The invention provides an MODIS remote sensing monitoring method for the vertical distribution pattern of eutrophic lake water algae. The method includes: acquiring the vertical distribution pattern of the algae by means of field monitoring; on the basis of field measured water surface spectral information and environmental information, creating a ground measured spectral data (Rrs) based remote sensing monitoring method for vertical algae distribution; acquiring the quantitative relationship between ground monitoring and remote sensing reflectance Rrs and simulated Rrc after Rayleigh scattering correction by simulating different aerosol types and thicknesses and different solar altitudes, satellite observation angles and azimuthal angles; further promoting the ground measured spectral data based monitoring method for vertical algae distribution to MODIS satellite image data subjected to Rayleigh scattering correction. On the basis of the method, interannual and inter-monthly change rules and spatial distribution of the vertical distribution pattern of the eutrophic lake water algae can be acquired accurately, and scientific support is provided for scientific decision-making on water resource management and water environment protection water conservancy and environmental protection departments.

The invention discloses a spectral index model-based black and odorous water body identification method and system. The identification method comprises the steps of processing a selected high-spatial-resolution remote sensing image, superposing a river system vector document or a mask file corresponding to the processed fused remote sensing reflectance image to the fused remote sensing reflectance image, extracting river in the fused remote sensing reflectance image to obtain a river remote sensing image; obtaining blue, green, red and near-infrared bands of each pixel in the river remote sensing image and calculating a black and odorous spectral index according to the bands and a preset spectral index model; and determining the category of the water body corresponding to each pixel according to the relationship between the spectral index of each pixel unit and preset first threshold and second threshold. By adopting the black and odorous spectral index model, the identified black and odorous spectral index of the urban river is more accurate, the categories of the water bodies of the rivers are divided into a heavy black and odorous water body, a light black and odorous water body and a general water body, and differentiation of the water bodies is more detailed.

The invention discloses a water color exception object extraction method and system based on multi-spectral remote sensing. The method comprises the steps of first acquiring the reflectance of a remote sensing image, then carrying out separation of land and water and cloud removing to obtain remote sensing reflectance of a water area portion, and conducting a principal component analysis of the multiband remote sensing reflectance to obtain a first principal component; afterwards, conducting a histogram analysis of a first principal component image, setting a threshold, segmenting the first principal component into a target component and a background component by using the threshold, segmenting the first principal component by using a threshold when the between-class variance of the target and background components is maximum and by setting the multiple, and obtaining a water color exception object portion; and conducting pixel statistics for the water color exception object portion, and obtaining the area and distribution of water color exception objects. The invention can retain the amount of multiband information to the greatest extent, thereby improving the accuracy; and meanwhile, acquires the segmenting threshold completely based on automatic calculation of a computer, thus greatly saving the information extracting time, and achieving simple and accurate extraction of distribution information of the water color exception objects.

The invention provides a retrieval method and a retrieval device of chlorophyll a concentration of inland case II water, wherein the retrieval method of the chlorophyll a concentration of the inland case II water comprises the steps of firstly, classifying first remote sensing reflectance spectra to obtain a plurality of second remote sensing reflectance spectra taking different components as main components; calculating the weight matrixes of all the second remote sensing reflectance spectra, selecting all characteristic wave bands and calculating the chlorophyll spectral indexes; finally, carrying out weight fusion on the chlorophyll spectral indexes under all categories of all the spectra by weight to obtain weighted chlorophyll index, and fitting an inversion model to obtain the chlorophyll a concentration. Due to the method of firstly classifying and then carrying out weighted inversion, the limitation on the retrieval method caused by the area and the time can be reduced, and the smoothness and the stability of inversion results can be improved, and therefore, when the chlorophyll a concentration of a certain second remote sensing reflectance spectrum is inversed subsequently, the chlorophyll a concentration can be obtained by inversion through the inversion model corresponding to the second remote sensing reflectance spectrum, so that the universality of the method is improved.

The invention discloses a method and device for establishing a water body classification model. The method for establishing the water body classification model comprises: classifying an actually measured remote sensing reflectivity spectrum into N classes by a clustering algorithm; and then determining N water body classes corresponding to N reflectivity spectrum classes in combination with a water quality parameter and an inherent optical quantity, determining a sensitive waveband in a remote sensing image waveband range for each reflectivity spectrum class, associating the water body classes corresponding to the reflectivity spectrum with the sensitive wavebands; and finally generating a water body classification decision tree model. The method and device for establishing the water body classification model can be applied to image water body classification. A user can obtain the class of the water body directly by means of the water body classification decision tree model according to the remote sensing image data of the water body, and is not required to carry out actual measurement and experiments in the field and a laboratory. The method and device are simple, convenient, and good in timeliness.

The invention discloses a water body water color abnormity identification method and device based on time sequence remote sensing images, and belongs to the field of water environment pollution monitoring. The method comprises the following steps: preprocessing a multi-time-sequence remote sensing image to obtain water body remote sensing reflectivity images with different time sequences; performing spectral statistical analysis on the remote sensing reflectivity image to obtain water body pixel statistical characteristic values of different wave band long-time sequences, and establishing different wave band reference characteristic values and pixel characteristic values; making a spectral characteristic analysis on the to-be-detected remote sensing image, and creating waveband characteristic values of different wavebands; preliminarily judging whether the to-be-detected image has abnormal water color or not by comparing the wave band characteristic value of the to-be-detected image with the corresponding wave band reference characteristic value; analyzing the wave band with the abnormal water color of the to-be-detected image pixel by pixel, comparing the wave band with the pixelcharacteristic value of the corresponding wave band, obtaining the spatial distribution pixel of the abnormal water color, and obtaining the spatial distribution of the abnormal water color of the water body. According to the invention, the water color abnormity identification precision and speed of the water body are improved.

The invention discloses a remote sensing extraction method, a remote sensing extraction device and a remote sensing extraction system for polluted water body at drainage outlets and belongs to the field of pollution monitoring for water body. The method includes the steps of: a) on the basis of a high-spatial-resolution remote-sensing reflectance image, constructing spectral libraries of a polluted water body and a normal water body, calculating characteristic values of different wave bands in the spectral data of the polluted water body and the normal water body, and screening a characteristic wave band by comparing the characteristic values between the polluted water body and the normal water body, and respectively setting threshold values according to the characteristic value of the characteristic wave band of spectrum of the polluted water body; b) performing comparison to every pixels in the remote-sensing reflectance image waveband by waveband, if a comparison result is true, marking the pixel as a polluted water body pixel; and c) performing extraction of position information and counting of pixel number to the polluted water body pixels so as to acquire position and area of the polluted water body. The method has stronger pertinence to extraction of the polluted water body and greatly increases precision of recognizing the position and area of the polluted water body, thereby high-effectively, quickly and accurately extracting the distribution range and area of the polluted water body.

The embodiment of the invention discloses an urban black and odorous water recognition method and device. The method comprises the following steps: calculating the interpolation reflectivity of a water level to-be-recognized point in a green light wave band according to remote sensing reflectivity of a bluelight wave band and the remote sensing reflectivity of a redlight wave band of the water level to-be-recognized point; calculating the black and odorous water grading index according to the interpolation reflectivity of the green light wave band and the interpolation reflectivity of the water level to-be-recognized point as well as the remote sensing reflectivity of the redlight wave band; and judging whether the water level to-be-recognized point is urban black and odorous water or notaccording to the black and odorous water grading index and grading the urban black and odorous water when the water level to-be-recognized point is urban black and odorous water or. The method not only recognizes the urban black and odorous water effectively, but also recognizes the urban black and odorous water in a graded manner.

The invention discloses a remote-sensing calculation method of long-term variation of suspended sediments of a strong-tide high-turbidity estuary. The method comprises the following steps: step S1, receiving Landsat satellite remote-sensing data in different years and the same hydrological condition; step S2, processing the remote-sensing data; step S3, carrying out spectral measurement above spring/neap-tide water surfaces, carrying out synchronous surface layer suspended-sediment sampling, obtaining remote-sensing reflectivity, analyzing wavebands where spectral curve peak values are located, selecting waveband data which are of different sensors on Landsat and correspond to reflectivity peak values, establishing a statistical regression model between surface layer suspended-sediment concentration and remote-sensing reflectivity corresponding to sensitive wavebands, and then obtaining an optimal calculation mode of quantitative remote sensing through accuracy verification calculation, wherein wavebands corresponding to the peak values are the sediment sensitive-wavebands; and step S4, applying the optimal calculation mode, which is of quantitative remote-sensing and is obtained in the step S3, to the remote-sensing data after processing in the step S2, obtaining remote-sensing data of the suspended sediments of all regions in the different years and the same hydrological condition, and carrying out analysis of long-term variation of the suspended sediments.

The invention relates to an optimal distinguishing method for a red tide water body and a main algae thereof, which comprises the following steps: (1) acquisition of spectroscopic data, which comprises the processes of laboratory culture of a red tide algae, measurement of the remote sensing reflectance and the post treatment of the spectroscopic data; and (2) analysis of the spectroscopic data, i.e. Carrying out comparison of spectral absorption characteristic parameters by a spectral absorption characteristic parameter extraction method through utilizing ENVI (The Environment for Visualizing Images) software, extracting wavelengths of a reflection peak and an absorption peak by a spectral differential method and carrying out comparison analysis on the water bodies before and after the red tide main algae is inoculated. According to the invention, by the spectroscopic data of different red tide main algae, which is obtained in a laboratory and the spectroscopic data flow processing, an abnormal value is removed to obtain a remote sensing spectral curve of Karenia mikimotoi, alexandrium tamarense and AmPhidinium; and the foundation is laid for the extraction of the absorption characteristic parameters and the spectral differential method.

The invention discloses a novel mixed method for estimating the particle size of seawater suspended particulate matter by means of satellite data. The method comprises the steps that seawater measured data of a seawater area to be estimated is acquired; satellite data is collected, and remote sensing reflectance data is extracted and calculated; the remote sensing reflectance of a seawater subsurface layer is calculated, and a backscattering coefficient of the suspended particulate matter is calculated by combining a total backscattering coefficient of the suspended particulate matter with a relation expression between a total absorption coefficient and the remote sensing reflectance of the seawater subsurface layer; a relation expression between the particle size of the seawater suspended particulate matter and the backscattering coefficient is derived, and a relation model between the backscattering coefficient of the seawater suspended particulate matter and the particle size of the seawater suspended particulate matter is established; an association relation process between the particle size of the seawater suspended particulate matter and the remote sensing reflectance of the seawater subsurface layer is established, and satellite-retrieved particle size distribution information of the seawater suspended particulate matter is obtained by means of monthly-composited remote sensing reflectance data. The remote sensing estimation capacity on offshore high-turbidity water body PSD distribution information can be improved.

The invention belongs to the technical field of water color remote sensing, and relates to a method for inversely analyzing the transparency of a water body by virtue of an inherent optical parameter. The method for inversely analyzing the transparency of the water body by virtue of the inherent optical parameter comprises the following steps: (1) reading data of measured transparency and data of remote sensing reflectance; (2) calculating an absorption and scattering coefficient through the remote sensing reflectance; (3) constructing an algorithm for inversely analyzing the transparency of the water body by virtue of the inherent optical parameter; (4) calculating an error between an inverse analysis value and a measured value to estimate the accuracy of the algorithm by taking the measured transparency as a true value. Compared with the prior art, the method for inversely analyzing the transparency of the water body by virtue of the inherent optical parameter has the advantages that the inherent optical parameter of the water body can be estimated according to the remote sensing reflectance to further inversely analyze the transparency of the water body, so that the problem of time-space limits of a case-two water body color constituent inverse analysis algorithm is solved, and universal applicability is achieved.

The invention discloses a method for inverting estuarine wetland water environment elements through a hyperspectral sensor on an unmanned aerial vehicle, and belongs to the field of water environmentsimulation. According to the method for inverting the estuarine wetland water environment elements through the hyperspectral sensor on the unmanned aerial vehicle, the estuarine wetland water environment elements are quickly calculated out in combination with the use of the unmanned aerial vehicle through the steps of collecting a water body sample of a unit research area, calculating the remote sensing reflectance of the water surface of the unit research area, analyzing a water environment of the unit research area, making unmanned aerial vehicle work preparation, collecting hyperspectral data of the unit research area, building a water environment model of the unit research area, collecting hyperspectral data of a to-be-tested research area, determining a water environment of the to-be-tested research area, and the like; the above steps are repeated to establish multiple groups of water quality content distribution images of the estuarine wetland research areas; and the change rulesof the water quality of the estuarine wetland research areas are compared and obtained in the same direction, so that the tidal creek water quality change conditions which are quickly and dynamicallychanged can be accurately captured, and the time and labor are saved.

The invention discloses a sea surface salinity inversion algorithm based on a wind cloud meteorological satellite. The method comprises the following steps: S1, training independent variables (X): increasing the number of hidden layers according to the need of a regression task, building a sea surface salinity satellite inversion algorithm model for a deep neural network, and setting up four independent variables including the four-waveband satellite observation remote sensing reflectance (Rrs), aCDOM, the sea surface temperature (SST) and the suspended matter concentration (TSM); and S2, measuring the sea surface salinity (Y): establishing a relationship between aCDOM, SST and TSM and the sea surface salinity through a deep learning model, and realizing the observation of the global seasurface salinity by utilizing the FY-3D/MERSI global observation capability. According to the invention, a sea surface salinity inversion algorithm of a domestic wind cloud meteorological satellite (FY-3D/MERSI) is established based on a deep learning method; observation of high spatial-temporal resolution of sea surface salinity is achieved, and the problems that due to the limitation of low spatial-temporal resolution, data quality and the like of SMOS and Aquarius/SAC-D satellites, the sea surface salinity data coverage rate is low, and the sea surface salinity cannot be observed with the high spatial-temporal resolution are solved.