77 results about "Spatial distribution pattern" patented technology

A system and method for delivering therapeutic laser energy onto selected treatment locations of the retina following a predetermined spatial distribution pattern using one single laser beam. A beam steering mechanism and control system delivers the laser energy sequentially to treatment locations forming a pre-selected treatment layout pattern. The invention allows time consuming therapeutic laser procedures such as pan-retinal photo-coagulation and segmental photocoagulation to be performed with increased accuracy and in a fraction of the time currently required for such procedures.

The invention discloses an arbor three-dimensional green quantity measuring method based on three-dimensional laser scanner, comprising the steps of: fitting point cloud data obtained by a three-dimensional laser scanning system at scanning stations in three different directions by a common reference ball to form a complete point cloud of single arbor, and finally generating single arbor crown point cloud data; segmenting the single arbor crown point cloud data from the top to the bottom of the crown into a plurality of sections in same height according to height h, wherein the sections comprises one cone and n circular truncated cones; calculating area of irregular surfaces according to equal distance by the complete single arbor crown point cloud data so as to calculate the volumes of one core and n circular truncated cones; summing the volumes of one core and n circular truncated cones, thereby obtaining the three-dimensional green quantity of single arbor. The method also can be promoted for calculating the three-dimensional green quantity of trees with irregular crowns, and has essential significance for measuring the areal green quantity, the per capita green quantity and the spatial distribution pattern of the green quantity.

A forecast method of forest fire danger of coupling climate and vegetation belongs to the computer digital image processing technical field. The forecast method of the forest fire danger of the coupling climate and vegetation comprises performing remote sensing classification on forest combustible in Zhejiang Province and performing grade division on forest fire combustible; generating a spatial forecast database of meteorological factors in a forest according to thin plate smoothing spline interpolation with meteorological data as the base; calculating to obtain a spatial database of climatic indexes of the forest fire danger in Zhejiang Province by utilizing a module of the climatic indexes of the forest fire danger and simultaneously performing the grade division on climate fire danger; performing coupling on the spatial database of the climatic indexes of the forest fire danger and a remote sensing classification database of the forest combustible, performing analysis and process by utilizing an ArcGIS ( Arc Geographic Information System) software, combining the occurrence intensity, spatial distribution pattern and regional characteristics of the forest fire disasters in Zhejiang Province, performing comprehensive analysis on fire occurrence and harm and combining with the media to establish an early warning and forecast system of the forest fire danger in Zhejiang Province to achieve forest fire danger forecast of different regions and different forest types.

The invention belongs to the field of agricultural remote sensing and relates to a remote sensing, crop model and meteorological forecast fused crop maturity stage prediction method. The method comprises the steps that S1 a crop planting area is partitioned; S2 the WOFOST model parameter of crop growth of each area is marked; S3 a filtering method is used to eliminate cloud pollution in a time series remote sensing image; S4 parameters sensitive to the maturity stage are selected, and a cost function based on normalized MODIS LAI and WOFOST simulation LAI is built to acquire an optimization parameter data set; S5 a WOFOST crop growth model is driven by the acquired optimization parameter data set, and the crop maturity stage is predicted 10 days ahead of schedule; and S6 an area crop maturity stage prediction spatial distribution pattern is generated to guide timely harvesting of the crop. The method provided by the invention fuses the advantages of remote sensing data and the crop model. The maturity stage can be predicted at the area scale, and the prediction accuracy of the crop model is improved. The application of meteorological forecast data improves the accuracy of dynamic prediction, so that the practicality is higher.

The present invention discloses a PDBSCAN (Parallel Density-Based Spatial Clustering of Applications with Noise) algorithm-based lightning approach trend prediction method. According to the defects of a traditional DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm, a VDBSCAN (Variable Density-Based Spatial Clustering of Applications with Noise) algorithm and an IDBSCAN algorithm are compounded from the two aspects of parameter assignment selection and memory structure optimization; a PDBSCAN algorithm put forwards after the compounding of the two algorithms is used to carry out equal-time interval clustering on lightning data; and linear fitting is carried out on the space centers of largest lightning clusters under equal time intervals, so that the space coordinates of a lightning occurrence region with the largest lightning density at a next time point are obtained. According to the algorithm of the invention, a condition that density changeS in the lightning clusters are monotonous is considered, a condition that the DBSCAN algorithm easily makes mistakes under a density changing condition can be eliminated, and parameter selection of various kinds of possible spatial distribution patterns of lightning is improved; and the storage structure of the algorithm is optimized, and computational efficiency is improved.

The invention discloses a method for determining the spatial distribution of the suitable transplanting period of flue-cured tobacco in a complicated hilly area. MODIS (Moderate Resolution Imaging Spectroradiometer) remote sensing image data and DEM (Dynamic Effect Model) altitude data are utilized and combined with ground observation data of the tobacco-growing area, a simulation method of the ten-day meteorology of the tobacco-growing area and the spatial and temporal distribution of soil moisture is constructed through mathematical modeling, main meteorological elements and the spatial and temporal distribution of soil moisture are extracted, the suitable meteorological factor and the soil moisture range, required by tobacco transplanting, are combined, and the spatial distribution of the suitable transplanting period of the tobacco-growing area can be determined. The method is simple, can be used conveniently, and can solve the problems that the traditional data based on sparse meteorological stations and less soil sampling points cannot reflect the spatial distribution of the meteorology and the soil moisture of the complicated hilly area easily, and the suitable transplanting period of the different areas of the complicated hilly area cannot be accurately determined.

The invention relates to a method for establishing a high-precision three-dimensional ground stress model for a heterogeneous formation. The method includes the first step of establishing a three-dimensional framework model for the heterogeneous formation according to drilling data, logging data and three-dimensional seismic data, the second step of establishing a three-dimensional rock physical parameter model for the heterogeneous formation according to three-dimensional seismic data inversion and the logging data, the third step of conducting three-dimensional finite element numerical simulation calculation on high-precision ground stress spatial distribution patterns of the heterogeneous formation, and the fourth step of establishing the high-precision three-dimensional ground stress distribution model for the heterogeneous formation. By means of the method for establishing the high-precision three-dimensional ground stress model for the heterogeneous formation, reliable reference can be provided for compact low-permeability oil and gas field development and sweet spot prediction and exploration and development of unconventional oil and gas, exploration and development efficiency of compact low-permeability and nonconventional oil and gas is effectively improved, and the risk cost of exploration and development of compact low-permeability and nonconventional oil and gas is lowered.

The invention discloses a wetland vegetation information analysis method, which comprises the following steps: step 100, preprocessing a remote sensing image of wetland vegetation collected by a satellite, and obtaining actually measured data; Step 200, performing classification processing on the remote sensing images by using various remote sensing image classification methods; Step 300, calculating the classification precision of each remote sensing image classification method, and determining the goodness of fit between the remote sensing image and the measured data; Step 400, comprehensively determining an optimal remote sensing image classification method according to the classification precision and the goodness of fit; Step 500, selecting a clustering processing method to carry outdefect processing on the classified remote sensing images; Satellite remote sensing image data in previous years is used as an analysis object; the satellite remote sensing image data are compared with the measured data, the satellite remote sensing image data are processed by using different classification methods, and the classification precision of various classification methods is obtained through calculation, so that the optimal classification method is selected, and the spatial distribution pattern of the vegetation and the vegetation distribution condition change can be obtained.

The invention relates to a simulation and prediction method of a spatial distribution changing pattern of a wetland so that technical problems of single spatial scale, poor precision, and single prediction scene in the prior art can be solved. The method comprises the steps: extracting spatial distribution information of wetlands with known years to obtain a wetland type distribution pattern; according to the wetland type distribution pattern, simulating wetland spatial distribution changes; predicting wetland spatial distribution in various scenes; and carrying out precision evaluation on prediction results. The method has the following advantages: according to the multi-time-space multi-scene wetland distribution pattern simulation and prediction method, prediction on the needed area ofthe wetland is improved by using a CA-Markov model; the simulation and prediction precision of the wetland spatial distribution pattern is enhanced; the regional adaptability is high; and the future spatial distribution pattern simulation and prediction of the wetland can be carried out stably based on a simulation model.

The brightness of a LCD backlight, comprising a plurality of LEDs or LED chains (221—1 to 221—n), is controlled by a processor (203) for calculating the number of LEDs of a plurality of LEDs to be illuminated for a determined brightness and selecting the calculated number of LEDs or LED chains to form a spatial distribution pattern; and drivers (219—1 to 219—n) for illuminating the calculated number of LEDs in a random or pseudo-random sequence.

The invention discloses a motor imagery electrocorticogram (EEG) signal classification method based on independent component analysis, which comprises the following steps: S1, collecting the EEG signals and preprocessing the EEG signals, and randomly dividing the preprocessed EEG signals into a training set and a test set; S2, sequentially selecting the training set data to carry out independent component analysis and calculation on the single test sample data, and automatically identifying and acquiring the motion related component based on the spatial distribution pattern of the source; S3:classification and recognition of motor imagery based on zero training classifier; S4: optimizing the selection of leads by using the training set data, substituting the optimized leads into the testset, and looping the steps S2 and S3 to obtain the final classification recognition rate. The invention can reduce the spatial model matching problem caused by the difference between the collected EEGdata, and has high recognition accuracy to the motor imagery EEG signals.

The invention discloses an evaluation method for a landscape fragmentation index, and belongs to the technical field of landscape ecology. The evaluation method comprises the steps of obtaining satellite remote sensing image data of the land under test; interpreting according to the satellite remote sensing image data to obtain multiple landscape indexes of the land under test; and calculating thelandscape fragmentation index according to the multiple landscape indexes of the land under test. The landscape fragmentation index obtained according to the method disclosed by the invention can accurately reflect the fragmentation degree of the landscape in an area, thereby facilitating the study for the landscape fragmentation. The method integrates multiple landscape indexes, thereby being capable of effectively avoiding a problem of index deviation caused by the regional heterogeneity. The method is conducive to the analysis for a spatial distribution pattern of the regional landscape fragmentation. The extraction method of the invention has the characteristics of universality, high efficiency simplicity and convenience.

The invention relates to the technical field of water pollution physical treatment and prevention and control of algal bloom, and provides a manual flow making device. The manual flow making device comprises an engineering ship, a sewage pump, a hydrologic winch and the like, wherein a high-power sewage pump, a diesel oil power machine and a generator set are arranged in a cabin of the engineering ship, a water inlet and a water outlet of the high-power sewage pump are respectively connected in a compression manner through a hose, a direction is adjusted through the hydrologic winch so as to change the inlet water current and the outlet water current. According to the invention, flow field interference is taken as a core technology, through actions of the high-power sewage pump and the engineering ship, hydrologic characteristics of a working water area can be changed, and flow state characteristics of a water area with water bloom can be destroyed, so that a water body is accelerated to flow and mix, the content of dissolved oxygen of the water body in the working region is increased, the temperature property and the optical property of the water body are changed, a space distribution pattern of biogenesis elements is adjusted, substance circulation and energy flow are promoted, environmental conditions necessary for dominant species of algae are destroyed, coverage cannot be kept, and competitive capacity is lost and is gradually faded.

The invention provides a floating car data processing method based on improved amplitude limiting average filtering. The floating car data processing method comprises the following steps, carrying outdata input, characteristic analysis, state analysis based on FCD (Floating Car Data) distribution, road intersection structure analysis, FCD and road mapping analysis so as to obtain a spatial distribution pattern of a floating car track; carrying out FCD data preprocessing, carrying out three-stage partitioning on floating car data on the basis of a city road vector diagram and road intersectionturnout planning specifications according to road nodes, road types and road gaps respectively, and taking partitioned sections as basis study units; carrying out track center line fitting and data conversion and integration, and with the combination of floating car track geometrical characteristics, motion characteristic change rules and spatial distribution rules of floating car data, carryingout floating car data filtering based on improved amplitude limiting average filtering. The technical scheme of the invention has the advantages of being convenient in operation and efficient in algorithm, practical data sources can be provided for later techniques that lanes are widened according to floating car data exploration, and exploration precision is correspondingly improved.

The invention relates to the technical field of forest analysis. The invention and particularly relates to a method for analyzing a carbon sink structure of a small-khingan acicular-broad-leaved mixedforest based on a carbon reserve distribution state. The method comprises the following steps: 1, preparing raw materials; setting a standard sample plot setting, adopting a sample method and an arbor positioning method to carry out are adopted. Sspecies identification, listing and positioning are carried out on arbor and small arbor plant individuals; measuring the tree height, the diameter atbreast height, the height under branches and the crown breadth; calculating carbon reserves are calculated, acquiring the spatial distribution pattern differentiation characteristic of the forest carbon reserves is obtained according to the carbon reserves, analyzing the sample plot spatial structure is analyzed through the vertical projection area, the mixing angle, the angular scale and the competition index, analyzing the carbon reserve distribution state is analyzed, discussing the variation situation of carbon reserve spatial distribution is discussed, and providing the carbon reserve spatial distribution characteristics of the coniferous and broadleaf mixed forest are provideds. The method invention has important significance for mastering the relationship between the forest space structure and the carbon reserves, and provides reference for discovering the defects existing in the development of the forest carbon reserves and searching for a higher sink increasing space.

The invention discloses a vegetation spatial distribution pattern investigation method based on an unmanned aerial vehicle technology. The method comprises the following steps: S1, obtaining an orthographic spliced image of a research area; S2, performing supervised classification and result verification on the orthographic spliced image in S1; S3, making a species classification extraction distribution diagram; and S4, performing spatial distribution pattern analysis on the species. The method has the beneficial effects that vegetation species classification is carried out on the orthographicspliced images by utilizing a supervised classification method, so that specific ecological statistical analysis is carried out on individuals of each species; manpower and material resources are saved, meanwhile, a pattern analysis quadrat can be expanded by tens of times, and a spatial pattern analysis conclusion can be obtained more comprehensively and accurately; multi-angle spatial distribution pattern analysis can be more conveniently carried out on the species classification extraction distribution diagram, such as a sample method, a point pattern analysis method and the like, so thatpattern analysis results are more comprehensive and specific.

The invention discloses a natural disaster induction technology accident disaster risk assessment method and system, and belongs to the technical field of risk assessment. In order to solve the problems of immature accident disaster risk assessment method, unclear risk spatial distribution pattern, low risk assessment accuracy and the like of a natural disaster induction technology in the prior art, the invention provides a risk assessment method and system. The method comprises the following steps of enlarging four types of first-grade indexes including a risk source, a natural disaster causing factor, a control mechanism level and receptor vulnerability to corresponding secondary indexes, finishing indexes calculations according to scores acquired by calculations of an analytic hierarchyprocess, constructing a risk assessment system, and calculating risk indexes. The assessment method is high in universality, a basis can be provided for further evaluation and risk management of theNatch risk, priority determination and key point evaluation are carried out on a Natch high-risk-level area, and the evaluation method can also be used as a basis for evaluating and analyzing the Natch risk in the area, the drainage basin or within the nation.

The invention discloses a primer pair, a castanopsis hystrix SSR7 (Simple Sequence Repeat 7) marker and a preparation method and application thereof. According to the preparation method, the total DNA (Deoxyribonucleic Acid) of castanopsis hystrix is subjected to PCR (Polymerase Chain Reaction) amplification by using the specific primer pair HZ7, thereby obtaining the SSR7 marker. The method is simple and feasible and is simple and convenient in operation; shown by tests which are carried out in members of two natural castanopsis hystrix populations and among the members, the obtained SSR has relatively high polymorphism. The castanopsis hystrix SSR7 marker is a co-dominant marker; compared with other molecular markers, the SSR marker is good in repeatability and high in reliability, can be applied to the genetic diversity evaluation of castanopsis hystrix, the genetic map construction of castanopsis hystrix, the research on the spatial distribution pattern of the genetic variation of castanopsis hystrix population and the origination and evolution of castanopsis hystrix and the standardization of target genes, and can also be applied to the molecular marker assisted breeding and QTL (Quantitative Trait Locus) research of castanopsis hystrix.

The invention provides an ecological risk assessment method for a nature reserve. The ecological risk assessment method comprises the steps of S1, identifying ecological risk types of the natural reserve and constructing evaluation indexes of different types of ecological risks; S2, respectively calculating the weight of each evaluation index corresponding to different ecological risks of the natural reserve by adopting a mean square error weighting method, and metering the weight of the natural reserve harmed by different ecological risks; S3, visually evaluating the ecological risk indexes of different types of ecological risks of the natural reserve by adopting a GIS software platform to obtain an ecological risk index spatial distribution pattern diagram of different types of ecological risks; and S4, multiplying different types of ecological risk indexes of the natural reserve by the hazard weight of the ecological risk indexes on the GIS platform to obtain a comprehensive ecological risk index of the natural reserve, and grading the risk indexes to obtain a comprehensive ecological risk grading graph of the natural reserve. The method has the advantages of being accurate, scientific, comprehensive, systematic, wide in application range and the like for the ecological risk of the natural reserve.

The invention provides a method for performing extraction in a high turbidity area of a lake or a reservoir based on a remote sensing image, comprising steps of: constructing a remote sensing estimation model of the concentration of total suspended solids based on measured data and remote sensing reflectance of blue, green and red wavebands of an environmental satellite (HJ1A / B) to obtain the spatial distribution pattern of the concentration of the total suspended solids; constructing an extraction index of the high turbidity area of the lake or the reservoir based on the remote sensing reflectance of blue, green and red wavebands of the HJ1A / B, and performing extraction in the high turbidity area of the lake or the reservoir; analyzing the validity of the remote sensing estimation model and the extraction index of the high turbidity area. The method for performing extraction in a high turbidity area of a lake or a reservoir based on a remote sensing image is suitable for various inland lakes or reservoirs, simple and convenient to operate and easy to popularize and apply.

The invention discloses a space-based early warning method for critical phase transformation of an electric power system, and relates to the field of electric power systems. According to the method, the unbalance degree of power flow distribution and the distribution pattern of transmission capacity are captured through spatial autocorrelation and variance indexes for early warning the condition that the electric power system gets into a critical state; it shows that the spatial distribution pattern of the system line load rate is difficult to measure, but the spatial autocorrelation and variance of the line load rate are easy to measure, and therefore the spatial autocorrelation and variance of the system line load rate have the good early warning characteristics and can serve as an early warning signal for the condition that the system approaches critical phase transformation.

The invention discloses a Zelkova SSR1 marker and a primer pair and preparation methods and applications thereof. PCR amplification is carried out on Zelkova total DNAs of JS1 by an inventor utilizing a specific primer, so that the SSR1 marker with a simple and repeated sequence is obtained. The method is simple and practicable and simple and convenient to operate, the obtained simple and repeated sequence shows higher polymorphism for detection inside or among populations of three Zelkova natural colonies. The Zelkova SSR1 marker is a co-dominant marker; compared with other molecular markers, the SSR marker is good in repeatability and high in reliability, can be applied to Zelkova genetic diversity evaluation, Zelkova genetic map construction, Zelkova population heritable variation spatial distribution pattern and Zelkova origin and evolution research as well as target gene marking, and can also be applied to Zelkova molecular marker assistant breeding and QTL (quantitative Trait Locus) research.

The invention discloses a primer pair, a castanopsis hystrix SSR5 (Simple Sequence Repeat 5) marker and a preparation method and application thereof. According to the preparation method, the total DNA (Deoxyribonucleic Acid) of castanopsis hystrix is subjected to PCR (Polymerase Chain Reaction) amplification by using the specific primer pair HZ5, thereby obtaining the SSR5 marker. The method is simple and feasible and is simple and convenient in operation; shown by tests which are carried out in members of two natural castanopsis hystrix populations and among the members, the obtained SSR has relatively high polymorphism. The castanopsis hystrix SSR5 marker is a co-dominant marker; compared with other molecular markers, the SSR marker is good in repeatability and high in reliability, can be applied to the genetic diversity evaluation of castanopsis hystrix, the genetic map construction of castanopsis hystrix, the research on the spatial distribution pattern of the genetic variation of castanopsis hystrix population and the origination and evolution of castanopsis hystrix and the standardization of target genes, and can also be applied to the molecular marker assisted breeding and QTL (Quantitative Trait Locus) research of castanopsis hystrix.

The invention relates to a carbon emission space drawing method and a density spatial distribution determination method and device, and belongs to the technical field of environment monitoring. The drawing method comprises the steps defining resident living density by combining coverage rate of residential construction land in the horizontal direction and the density difference in the vertical direction caused by different floors of different buildings; then, establishing a regression model representing a mapping relationship between the two according to the correlation between the resident living density and the resident settlement index, and the resident living density and the resident settlement index of a small area; using the regression model, and combining the resident settlement index of the whole target area to obtain the accurate resident living density of the whole target area; and on the basis, obtaining the high-resolution resident living carbon emission spatial distribution pattern of the target area by combining the carbon emission under the unit living density, so that resident living carbon emission spatial distribution can be rapidly determined, and the acquired spatial distribution map has high precision and relatively strong applicability.

The invention discloses a primer pair, a castanopsis hystrix SSR3 (Simple Sequence Repeat 3) marker and a preparation method and application thereof. According to the preparation method, the total DNA (Deoxyribonucleic Acid) of castanopsis hystrix is subjected to PCR (Polymerase Chain Reaction) amplification by using the specific primer pair HZ3, thereby obtaining the SSR3 marker. The method is simple and feasible and is simple and convenient in operation; shown by tests which are carried out in members of two natural castanopsis hystrix populations and among the members, the obtained SSR has relatively high polymorphism. The castanopsis hystrix SSR3 marker is a co-dominant marker; compared with other molecular markers, the SSR marker is good in repeatability and high in reliability, can be applied to the genetic diversity evaluation of castanopsis hystrix, the genetic map construction of castanopsis hystrix, the research on the spatial distribution pattern of the genetic variation of castanopsis hystrix population and the origination and evolution of castanopsis hystrix and the standardization of target genes, and can also be applied to the molecular marker assisted breeding and QTL (Quantitative Trait Locus) research of castanopsis hystrix.

The invention discloses a filter tip trapped phenol concentration distribution pattern measuring method which comprises the following steps: concentrically cutting a sucked filter tip along the axis after radially cutting with a laser, or radially cutting a sucked filter tip after concentrically cutting along the axis; after finishing cutting, separating the annular filter tip sections from the corresponding number of cylindrical filter tip sections, carrying out quantitative analysis on the phenol content in the cut and separated filter tip sections; and after calculating the phenol concentration, carrying out polynomial fitting on filter tip trapped phenol concentration distribution data, or carrying out polynomial fitting and interpolation analysis to obtain the filter tip trapped phenol concentration distribution pattern. By measuring the longitudinal, cross section radial and spatial distribution pattern of different filter tip trapped phenol concentrations, the invention can master the filtration trapped distribution characteristics of phenol in filter tips in different structures, can quantitatively analyze the phenol trapping efficiency and contribution of all parts of filter tips in different structures in flue gas, and can be used for evaluation of flue gas phenol filtering quality of functional filter tips and design improvement of the functional filter tips.

An image processing circuit includes a memory, a gamma converter which converts a first image data signal of a frame to a current image data signal thereof corresponding to first or second gamma types based on a spatial distribution pattern, a compression circuit which separates the current image data signal into first and second gamma signals corresponding to the first and second gamma types and compresses the first and second gamma signals to first and second compression gamma signals to store the first and second compression gamma signals, a decompression circuit which output a previous image data signal of a previous frame by decompressing the first and second compression gamma signals and combines first and second decompression gamma signals, and a gamma correction circuit which performs a gamma adjustment based on current and previous image data signals to output a second image data signal.

The invention discloses a filter tip trapped HCN concentration distribution pattern measuring method which comprises the following steps: concentrically cutting a sucked filter tip along the axis after radially cutting with a laser, or radially cutting a sucked filter tip after concentrically cutting along the axis; after finishing cutting, separating the annular filter tip sections from the corresponding number of cylindrical filter tip sections, carrying out quantitative analysis on the HCN content in the cut and separated filter tip sections; and after calculating the HCN concentration, carrying out polynomial fitting on filter tip trapped HCN concentration distribution data, or carrying out polynomial fitting and interpolation analysis to obtain the filter tip trapped HCN concentration distribution pattern. By measuring the longitudinal, cross section radial and spatial distribution pattern of different filter tip trapped HCN concentrations, the invention can master the filtration trapped distribution characteristics of HCN in filter tips in different structures, can quantitatively analyze the HCN trapping efficiency and contribution of all parts of filter tips in different structures in flue gas, and can be used for evaluation of flue gas HCN filtering quality of functional filter tips and design improvement of the functional filter tips.

The invention discloses a river facies deposition dispersed oil sand body identification and exploitation method and application thereof, and belongs to the technical field of oil and gas field development. The method comprises the steps of: determining a river facies deposition area based on logging information and seismic data of a drilled well; according to the river facies deposition area, determining the spatial distribution pattern of river facies deposition sand bodies; and determining the distribution of the river facies deposition oil sand bodies according to the spatial distributionpattern of the river facies deposition oil sand bodies; and according to the distribution of the river facies deposition oil sand bodies, adopting a horizontal well exploitation mode, and connecting the river facies deposition oil sand bodies in series for exploitation. According to the method, the dispersed oil sand bodies deposited by the river facies are identified based on the logging information and the seismic data, and then a horizontal well exploitation is adopted, the oil sand bodies with large volume, high reserve and close distance are connected in series along the riverway to be exploited by adopting a horizontal well exploitation mode, so that the exploitation cost can be reduced, the oil well control reserves is increased, and the oilfield development benefits are improved.