CN117576579A - Method, device and storage medium for improving accuracy of determining feature type information - Google Patents
Method, device and storage medium for improving accuracy of determining feature type information Download PDFInfo
- Publication number
- CN117576579A CN117576579A CN202311535765.0A CN202311535765A CN117576579A CN 117576579 A CN117576579 A CN 117576579A CN 202311535765 A CN202311535765 A CN 202311535765A CN 117576579 A CN117576579 A CN 117576579A
- Authority
- CN
- China
- Prior art keywords
- pixels
- determining
- ground object
- end members
- feature type
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 46
- 239000013598 vector Substances 0.000 claims abstract description 67
- 238000005070 sampling Methods 0.000 claims abstract description 56
- 238000001228 spectrum Methods 0.000 claims description 39
- 230000003595 spectral effect Effects 0.000 claims description 23
- 238000002310 reflectometry Methods 0.000 claims description 18
- 238000012545 processing Methods 0.000 claims description 13
- 238000013527 convolutional neural network Methods 0.000 claims description 6
- 238000007477 logistic regression Methods 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 2
- 238000001514 detection method Methods 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 15
- 238000004891 communication Methods 0.000 description 14
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 6
- 101150052583 CALM1 gene Proteins 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 239000000284 extract Substances 0.000 description 4
- 230000002238 attenuated effect Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000005484 gravity Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000013441 quality evaluation Methods 0.000 description 2
- 230000002787 reinforcement Effects 0.000 description 2
- 230000001932 seasonal effect Effects 0.000 description 2
- 101150114882 CALM2 gene Proteins 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
- G01C11/04—Interpretation of pictures
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V8/00—Prospecting or detecting by optical means
- G01V8/10—Detecting, e.g. by using light barriers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/17—Terrestrial scenes taken from planes or by drones
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Remote Sensing (AREA)
- Computing Systems (AREA)
- Medical Informatics (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Astronomy & Astrophysics (AREA)
- Radar, Positioning & Navigation (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geophysics (AREA)
- Image Processing (AREA)
Abstract
The application discloses a method for improving accuracy of determining ground object type information, which comprises the following steps: acquiring a first image corresponding to a target area based on a satellite; extracting a plurality of pixels serving as sampling points from a remote sensing image, and determining abundance values of end members in the plurality of pixels; acquiring a second image corresponding to the plurality of pixels by using the unmanned vehicle, determining a model based on the ground feature type information, and determining a plurality of ground feature type vectors corresponding to the ground feature type; determining score values corresponding to the plurality of pixels based on the abundance values of the end members in the plurality of pixels and a plurality of feature type vectors corresponding to the feature types; and under the condition that the score value corresponding to the plurality of pixels is larger than a preset score threshold value, the plurality of pixels can be used as sampling points. Thus, in the case that the score value is larger than the preset score threshold value, the selected plurality of pixels are qualified as sampling points. And then the ground object type information determined based on the plurality of pixels is more accurate.
Description
Technical Field
The present disclosure relates to the field of unmanned and satellite communications technologies, and in particular, to a method, an apparatus, and a storage medium for improving accuracy of determining ground object type information.
Background
With the continuous development of satellite technology, satellites are increasingly applied to unmanned technology. The satellite has the advantages of wide coverage, no need of installing sensors or base stations on roads, and the like, so that the satellite plays a great role in unmanned technology. For example, an unmanned vehicle can travel along a predetermined travel route, and the application of the satellite navigation system is not restricted.
The existing statistics of the ground feature type information is generally performed by a statistics person in the field, and then manually counting or counting the existing ground feature type information according to the record of the historical ground feature type information. However, whatever method is used for counting the ground object type information, omission is easy to occur, and the accuracy of counting the ground object type information is not high.
For example, the existing technical scheme is that based on a remote sensing image which is acquired by a satellite and corresponds to a target area, a plurality of pixels are extracted from the remote sensing image, and the abundance value of an end member in the plurality of pixels is determined, so that the ground object type information of the target area is determined according to the determined abundance value.
However, the above operations of extracting a plurality of pixels from the remote sensing image and determining the ground object type information of the target area based on the abundance values of the end members in the plurality of pixels may result in that the satellite does not perform quality evaluation on the pixel points in the remote sensing image, so that it cannot be determined whether the selected pixel points are qualified sampling points, and thus the accuracy of the finally counted ground object type information may be low.
Publication No. CN116543316A, entitled a method for identifying turf in paddy fields using multi-phase high resolution satellite images. The method comprises the following steps: acquiring multi-temporal satellite paddy field images of different growth periods of paddy fields of the turf in the paddy field to be identified; separating non-vegetation type ground object types from the multi-time satellite paddy field images through a machine learning algorithm, and determining multi-time vegetation coverage area image data; classifying the image data of the multi-time-phase vegetation coverage area by utilizing a plurality of vegetation index features obtained by a pre-test, and determining the image data of the multi-time-phase seasonal vegetation type; according to the time sequence change characteristics of paddy and turf and vegetation index information, respectively dividing turf areas in a hybrid paddy field in multi-temporal seasonal vegetation type image data by using an AI deep learning algorithm, and determining paddy image data and divided turf image data; a multi-temporal turf image is determined based on the split turf image data and the initial turf image data.
Publication number is CN116863330A, named as a method for intelligently detecting ground objects based on reinforcement learning and Transformer feature fusion. Comprising the following steps: s1, constructing an unstructured BIM three-dimensional ground object model by means of a digital three-dimensional design platform based on a BIM technology. Classifying the BIM three-dimensional ground object graphs, and sorting ground object type labels to obtain various ground object target data sets; s2, constructing a Vision Transformer model with fused features, separating and identifying various objects in a target scene, acquiring the features of the ground objects, and detecting the initial positions of the ground objects; s3, inputting the initial position and the feature characteristics of the feature into a reinforcement learning module, and outputting a feature position diagram with finer granularity by an evaluation network; and S4, visualizing algorithm results and providing an external use interface.
Aiming at the technical problems that in the prior art, the satellite does not evaluate the quality of the pixel points in the remote sensing image, so that whether the selected pixel points are qualified sampling points or not cannot be determined, and the accuracy of the finally counted ground object type information is low, no effective solution is proposed at present.
Disclosure of Invention
The embodiment of the disclosure provides a method, a device and a storage medium for improving accuracy of determining ground object type information, which at least solve the technical problem that in the prior art, as satellites do not evaluate quality of pixel points in an image video received from an unmanned vehicle, whether the selected pixel points are qualified sampling points cannot be determined, so that the accuracy of finally counted ground object type information is low.
According to an aspect of the embodiments of the present disclosure, there is provided a method for improving accuracy of determining feature type information, including: acquiring a first image corresponding to a target area based on a satellite; extracting a plurality of pixels serving as sampling points from a first image, and determining abundance values of end members in the plurality of pixels; acquiring a second image corresponding to the plurality of pixels by using the unmanned vehicle, determining a model based on the ground feature type information, and determining a plurality of ground feature type vectors corresponding to the ground feature type; determining score values corresponding to the plurality of pixels based on the abundance values of the end members in the plurality of pixels and a plurality of feature type vectors corresponding to the feature types; and under the condition that the score value corresponding to the plurality of pixels is larger than a preset score threshold value, the plurality of pixels can be used as sampling points.
According to another aspect of the embodiments of the present disclosure, there is also provided a storage medium including a stored program, wherein the method of any one of the above is performed by a processor when the program is run.
According to another aspect of the embodiments of the present disclosure, there is also provided an apparatus for improving accuracy of determining feature type information, including: the image acquisition module is used for acquiring a first image corresponding to the target area based on a satellite; the abundance value determining module is used for extracting a plurality of pixels serving as sampling points from the first image and determining abundance values of end members in the plurality of pixels; the ground object type vector determining module is used for acquiring second images corresponding to the plurality of pixels by using the unmanned vehicle, determining a model based on ground object type information and determining a plurality of ground object type vectors corresponding to the ground object types; the score value determining module is used for determining score values corresponding to the plurality of pixels based on the abundance values of the end members in the plurality of pixels and a plurality of ground object type vectors corresponding to the ground object types; and the first judging module is used for indicating that the plurality of pixels can be used as sampling points under the condition that the score values corresponding to the plurality of pixels are larger than a preset score threshold value.
According to another aspect of the embodiments of the present disclosure, there is also provided an apparatus for improving accuracy of determining feature type information, including: a processor; and a memory, coupled to the processor, for providing instructions to the processor for processing the steps of: acquiring a first image corresponding to a target area based on a satellite; extracting a plurality of pixels serving as sampling points from a first image, and determining abundance values of end members in the plurality of pixels; acquiring a second image corresponding to the plurality of pixels by using the unmanned vehicle, determining a model based on the ground feature type information, and determining a plurality of ground feature type vectors corresponding to the ground feature type; determining score values corresponding to the plurality of pixels based on the abundance values of the end members in the plurality of pixels and a plurality of feature type vectors corresponding to the feature types; and under the condition that the score value corresponding to the plurality of pixels is larger than a preset score threshold value, the plurality of pixels can be used as sampling points.
The application provides a method for improving accuracy of determining feature type information. First, the processor acquires a first image corresponding to a target area based on a satellite. The first image is a remote sensing image. The processor then extracts a plurality of pixels as sampling points in the first image and determines an abundance value of an end member of the plurality of pixels. Further, the processor acquires a second image corresponding to the plurality of pixels by using the unmanned vehicle, and determines a model based on the feature type information, and determines a plurality of feature type vectors corresponding to the feature type. The processor then determines score values corresponding to the plurality of pixels based on the abundance values of the end members in the plurality of pixels and the plurality of feature type vectors corresponding to the feature types. Finally, the processor indicates that the plurality of pixels can be used as sampling points when the score values corresponding to the plurality of pixels are greater than a preset score threshold.
Because the processor determines the model based on the ground object type information, determines a plurality of ground object type vectors corresponding to the ground object type, and determines the score values corresponding to the plurality of pixels based on the abundance values of the end members in the plurality of pixels and the plurality of ground object type vectors corresponding to the ground object type, the score values in the case that the plurality of pixels are taken as sampling points can be accurately known. Thus, in the case that the score value is larger than the preset score threshold value, the selected plurality of pixels are qualified as sampling points. And then the ground object type information determined based on the plurality of pixels is more accurate.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this application, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure and do not constitute an undue limitation on the disclosure. In the drawings:
FIG. 1A is a schematic diagram of a satellite-based unmanned system according to embodiment 1 of the present application;
fig. 1B is a schematic diagram of a hardware architecture of a satellite according to embodiment 1 of the present application;
fig. 1C is a schematic diagram of a hardware architecture of an unmanned vehicle according to embodiment 1 of the present application;
FIG. 2 is a flow chart of a method for enhancing certain terrain type information according to embodiment 1 of the present application;
FIG. 3 is a schematic illustration of satellite communication with an unmanned vehicle according to embodiment 1 of the present application;
FIG. 4 is a schematic illustration of an unmanned vehicle according to embodiment 1 of the present application;
FIG. 5 is a schematic diagram of a model for determining the type of feature information according to embodiment 1 of the present application
FIG. 6 is a schematic diagram of various wave positions in a satellite coverage area according to embodiment 1 of the present application;
FIG. 7 is a schematic diagram of a satellite scanning various wave positions with multiple beams according to embodiment 1 of the present application;
FIG. 8 is a schematic diagram of an apparatus for improving accuracy of determining terrain type information according to embodiment 2 of the present application; and
fig. 9 is a schematic diagram of an apparatus for improving accuracy of determining feature type information according to embodiment 3 of the present application.
Detailed Description
In order to better understand the technical solutions of the present disclosure, the following description will clearly and completely describe the technical solutions of the embodiments of the present disclosure with reference to the drawings in the embodiments of the present disclosure. It will be apparent that the described embodiments are merely embodiments of a portion, but not all, of the present disclosure. All other embodiments, which can be made by one of ordinary skill in the art without inventive effort, based on the embodiments in this disclosure, shall fall within the scope of the present disclosure.
It should be noted that the terms "first," "second," and the like in the description and claims of the present disclosure and in the foregoing figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the disclosure described herein may be capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
According to the present embodiment, there is provided an embodiment of a method of improving accuracy in determining feature type information, it should be noted that the steps illustrated in the flowchart of the drawings may be performed in a computer system such as a set of computer executable instructions, and that although a logical order is illustrated in the flowchart, in some cases the steps illustrated or described may be performed in an order other than that illustrated herein.
Fig. 1A is a schematic diagram of a satellite-based unmanned system according to an embodiment of the present application. Referring to fig. 1A, the unmanned vehicle 11 travels on a path S within a target area, where at the present time, the unmanned vehicle 11 is located at a P1 position and communicates with a satellite 200, and unmanned is realized under the control of the satellite 200. For example, the unmanned vehicle 11 uploads the photographed image video and the collected sensor data to the satellite 200, so that the satellite 200 transmits an instruction to the unmanned vehicle 11. In addition, and with reference to FIG. 1A, satellite 200 is also in communication with ground station 30 and is connected to a remote navigation server 40 via ground station 30.
Fig. 1B further illustrates a schematic diagram of the hardware architecture of the satellite 200 of fig. 1A. The structure of the satellite 20 shown in fig. 1B may be applied to the satellite 200 shown in fig. 1A. Referring to fig. 1B, the satellite 200 includes an integrated electronic system including: processor, memory, bus management module and communication interface. Wherein the memory is coupled to the processor such that the processor can access the memory, read program instructions stored in the memory, read data from the memory, or write data to the memory. The bus management module is connected to the processor and also to a bus, such as a CAN bus. The processor can communicate with the satellite-borne peripheral connected with the bus through the bus managed by the bus management module. In addition, the processor is also in communication connection with the camera, the star sensor, the measurement and control transponder, the data transmission equipment and other equipment through the communication interface. It will be appreciated by those of ordinary skill in the art that the configuration shown in fig. 1B is merely illustrative and is not intended to limit the configuration of the electronic device described above. For example, satellite 200 may also include more or fewer components than shown in FIG. 1B, or have a different configuration than shown in FIG. 1B.
Fig. 1C further illustrates a schematic diagram of the hardware architecture of the ground station 30 of fig. 1A. Referring to fig. 1C, the ground station 30 may include one or more processors (which may include, but are not limited to, a microprocessor MCU, a processing device such as a programmable logic device FPGA), a memory for storing data, a transmission device for communication functions, and an input/output interface. Wherein the memory, the transmission device and the input/output interface are connected with the processor through a bus. It will be appreciated by those of ordinary skill in the art that the configuration shown in fig. 1C is merely illustrative and is not intended to limit the configuration of the electronic device described above. For example, the ground system may also include more or fewer components than shown in FIG. 1C, or have a different configuration than shown in FIG. 1C.
It should be noted that one or more of the processors and/or other data processing circuits shown in fig. 1B and 1C may be referred to herein generally as a "data processing circuit. The data processing circuit may be embodied in whole or in part in software, hardware, firmware, or any other combination. Furthermore, the data processing circuitry may be a single stand-alone processing module, or incorporated in whole or in part into any of the other elements in the computing device. As referred to in the embodiments of the present disclosure, the data processing circuit acts as a processor control (e.g., selection of the variable resistance termination path to interface with).
The memory shown in fig. 1B and 1C may be used to store software programs and modules of application software, such as a program instruction/data storage device corresponding to a method for improving accuracy of determining feature type information in the embodiments of the present disclosure, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, implements the method for improving accuracy of determining feature type information of the application program. The memory may include high speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid state memory
It should be noted here that, in some alternative embodiments, the apparatus shown in fig. 1B and 1C described above may include hardware elements (including circuits), software elements (including computer code stored on a computer readable medium), or a combination of both hardware elements and software elements. It should be noted that fig. 1B and 1C are only one example of a specific example, and are intended to illustrate the types of components that may be present in the above-described devices.
In the above-described operating environment, according to a first aspect of the present embodiment, there is provided a method of improving accuracy of determining feature type information, the method being implemented by a processor shown in fig. 1C. Fig. 2 shows a schematic flow chart of the method, and referring to fig. 2, the method includes:
S202: acquiring a first image corresponding to a target area based on a satellite;
s204: extracting a plurality of pixels serving as sampling points from a first image, and determining abundance values of end members in the plurality of pixels;
s206: acquiring a second image corresponding to the plurality of pixels by using the unmanned vehicle, determining a model based on the ground feature type information, and determining a plurality of ground feature type vectors corresponding to the ground feature type;
s208: determining score values corresponding to the plurality of pixels based on the abundance values of the end members in the plurality of pixels and a plurality of feature type vectors corresponding to the feature types; and
s210: and under the condition that the score values corresponding to the plurality of pixels are larger than a preset score threshold value, the plurality of pixels can be used as sampling points.
Specifically, the processor may acquire a first image corresponding to the target area, for example, based on a satellite (S302). The first image is a remote sensing image corresponding to the target area.
Then, the processor extracts a plurality of pixels U1 to UL as sampling points in the remote sensing image, and determines abundance values of end members in the plurality of pixels U1 to UL (S304). Specifically, the processor determines spectral features corresponding to the plurality of picture elements U1-UL. The frequency spectrum features are used for reflectivity data of different frequency bands corresponding to the indication pixels. The processor then obtains reference spectral features corresponding to the end members of the different surface feature types. The end member is used for indicating pixels only comprising one ground object type, and the reference spectrum characteristic comprises reflectivity data of different frequency bands corresponding to the end members of different ground object types. And finally, the processor determines the abundance value of the signal attenuated ground object type in the pixel according to the frequency spectrum characteristic and the reference frequency spectrum characteristic corresponding to the end members of different ground object types. The foregoing will be described in detail later, and thus will not be described in detail here.
Further, the processor acquires a second image corresponding to the plurality of pixels using the unmanned vehicle, and determines a plurality of feature type vectors corresponding to the feature type based on the feature type information determination model (S306). Specifically, fig. 3 is a schematic diagram of satellite communication with an unmanned vehicle according to embodiment 1 of the present application. Fig. 4 is a schematic view of an unmanned vehicle according to embodiment 1 of the present application. As shown with reference to fig. 3 and 4, for example, the unmanned vehicle 11, the unmanned vehicle 12, and the unmanned vehicle 13 may communicate with the satellite 200, respectively. And the unmanned vehicles 11 to 13 may be provided with a plurality of cameras, for example, in the present embodiment, the unmanned vehicles 11 to 13 may be provided with 8 cameras Cam1 to Cam8. Thus, the unmanned vehicles 11 to 13 can collect images of the surroundings of the unmanned vehicles 11 to 13 through the cameras Cam1 to Cam8.
Further, for example, the traveling node P1 of the unmanned vehicle 11 corresponds to the pixel U1 in the first image; the driving node P2 of the unmanned vehicle 12 corresponds to the pixel U2 in the first image; the driving node P3 of the unmanned vehicle 13 corresponds to the pixel U3 in the first image; and so on; the driving node PL of the unmanned L corresponds to the picture element UL in the first image. Note that, the travel nodes P1 to PL are obtained by dividing the target area.
Thus, when the satellite 200 needs to determine the feature type information related to the traveling nodes P1 to PL, it is possible to acquire images captured by the respective cameras of the unmanned vehicles 11 to L of the traveling nodes P1 to PL and evaluate the feature type information determined from the remote sensing images using the images. That is, it is determined whether or not the selected pixel as the sampling point is qualified.
Thus, the processor can acquire the second image corresponding to the plurality of picture elements U1 to UL using the unmanned vehicles 11 to L. For example, referring to fig. 4, cameras Cam1 to Cam8 of the unmanned vehicle 11 may capture images Img1 to Img8, respectively. For example, the camera Cam1 captures an image Img1; the camera Cam2 shoots an image Img2; ..; camera Cam7 captures an image Img7 and camera Cam8 captures an image Img8. In this embodiment, img1 to Img8 are images including 3 channels, for example. And will not be described in detail hereinafter. Thus, the satellite 200 acquires images Img1 to Img8 acquired by the unmanned vehicle 11.
Then, the satellite 200 inputs the acquired images Img1 to Img8 to the ground pattern information determination model based on the convolutional neural network. Wherein fig. 5 shows a schematic diagram of a model for determining the type of the surface feature information. Referring to fig. 5, the feature type information determining module includes: a plurality of convolution/pooling layers; fully connected layer and softmax classifier.
The softmax classifier outputs a feature type vector Q corresponding to the feature type 1 ：
Q 1 ＝[q 11 ,q 12 ,q 13 ,q 14 ,q 15 ] T 。
Wherein q 11 Representing the proportion of the tree land object type in the surrounding environment of the driving node P1; q 12 Representing the proportion of the building ground object type in the surrounding environment of the driving node P1; q 13 Representing the proportion of the ground object type in the surrounding environment of the driving node P1; q 14 Representing the proportion of the hilly mountain land object type in the surrounding environment of the driving node P1; q 15 Representing the specific gravity of the water body ground object type in the surrounding environment of the driving node P1.
Thus, the feature type vector Q corresponding to the traveling nodes P1 to PL 1 ～Q L Can be determined by the above-described operation steps.
The processor then based on the abundance value K of the end members in the plurality of pixels U1-UL 1 ～K L And with the groundA plurality of ground object type vectors Q corresponding to object types 1 ～Q L Fractional values corresponding to the plurality of pixels are determined (S308). Specifically, a plurality of feature type vectors Q corresponding to the feature types are determined at the processor 1 ～Q L Can determine the average value of a plurality of feature type vectors corresponding to the feature typeThe calculation formula is as follows:
wherein Q is i Represents any one of a plurality of feature type vectors.
Further, the processor is based on the determined average of the plurality of terrain type vectors And abundance value K of end member in multiple pixels U1-UL 1 ～K L Score values corresponding to the plurality of picture elements U1 to UL are determined. Specifically, first, the abundance value K of the end member in the plurality of pixels U1 to UL is determined 1 ～K L Average value of vectors of multiple ground object types +.>Distance d therebetween. The calculation formula is as follows:
then, the processor uses the distance d as a parameter and determines score values P corresponding to the plurality of pixels by using a logistic regression model.
Finally, the processor judges whether the score value P corresponding to the plurality of pixels is larger than a preset score threshold value P x . When the score value P corresponding to a plurality of pixels is larger than the preset score threshold value P x In the case of (a) representsThe plurality of pixels are qualified as sampling points, and finally, the ground object type information determined based on the plurality of pixels is more accurate; in the case that the score value P corresponding to a plurality of pixels is smaller than the preset score threshold value P x In the case where it is indicated that a plurality of picture elements are not qualified as sampling points, it is necessary to reselect a plurality of picture element points as sampling points.
As described in the background art, the existing statistics of the ground feature type information is generally performed by a statistics person in the field for investigation, and then manually performing statistics or performing statistics on the existing ground feature type information according to the record of the historical ground feature type information. However, whatever method is used for counting the ground object type information, omission is easy to occur, and the accuracy of counting the ground object type information is not high.
For example, the existing technical scheme is that based on a remote sensing image which is acquired by a satellite and corresponds to a target area, a plurality of pixels are extracted from the remote sensing image, and the abundance value of an end member in the plurality of pixels is determined, so that the ground object type information of the target area is determined according to the determined abundance value.
However, the above operations of extracting a plurality of pixels from the remote sensing image and determining the ground object type information of the target area based on the abundance values of the end members in the plurality of pixels may result in that the satellite does not perform quality evaluation on the pixel points in the remote sensing image, so that it cannot be determined whether the selected pixel points are qualified sampling points, and thus the accuracy of the finally counted ground object type information may be low.
In view of this, the present application provides a method for improving accuracy in determining terrain type information. And because the processor determines the model based on the ground object type information, determines a plurality of ground object type vectors corresponding to the ground object types, and determines the score values corresponding to the plurality of pixels based on the abundance values of the end members in the plurality of pixels and the plurality of ground object type vectors corresponding to the ground object types, the score values under the condition that the plurality of pixels serve as sampling points can be accurately known. Thus, in the case that the score value is larger than the preset score threshold value, the selected plurality of pixels are qualified as sampling points. And then the ground object type information determined based on the plurality of pixels is more accurate. Therefore, the technical problem that in the prior art, as the satellite does not evaluate the quality of the pixel points in the remote sensing image, whether the selected pixel points are qualified sampling points cannot be determined, and the accuracy of the finally counted ground feature type information is low is solved.
Optionally, the operation of extracting a plurality of pixels as sampling points in the remote sensing image and determining the abundance value of the end member in the plurality of pixels includes: extracting a plurality of pixels serving as sampling points from the received remote sensing image; determining spectral features corresponding to the pixels, wherein the spectral features are used for indicating reflectivity data of different frequency bands corresponding to the pixels; acquiring reference spectrum characteristics corresponding to end members of different ground object types, wherein the end members are used for indicating that pixels of only one ground object type are included, and the reference spectrum characteristics include reflectivity data of different frequency bands corresponding to the end members of different ground object types; and determining abundance values of the signal attenuated ground object types in the pixels according to the frequency spectrum characteristics and the reference frequency spectrum characteristics corresponding to the end members of the different ground object types.
Specifically, first, the processor extracts a plurality of pixels U1 to UL as sampling points in the received remote sensing image. The processor then determines spectral features corresponding to the picture elements. Fig. 6 is a schematic diagram of each wave position in a satellite coverage area E according to embodiment 1 of the present application. Referring to FIG. 6, for example, satellite 110 includes a plurality of wave positions P within coverage area E 1 ～P n And satellite 110 may transmit to the covered plurality of wave positions P 1 ～P n Satellite communication services are provided.
FIG. 7 is a satellite with m beams B according to an embodiment of the present application 1 ～B m For each wave position P in coverage range 1 ～P n Schematic of the scan. Referring to FIG. 7, at the same time, satellites 110 may respectively transmit in m beams B 1 ～B m At the same time for different wave positions P in the coverage area E 1 ～P n Scanning is performed, where m < n. In the present embodiment, for example, it is possible to define the respective beam pairs within the defined areaThe wave position is scanned. For example, beam B 1 The covered area is S 1 Coverage area S 1 The wave position P corresponds to 1 ～P 3 So that beam B 1 For para-wave position P 1 ～P 3 Scanning is performed and beam B 1 Can be directed to the covered wave position P 1 ～P 3 Providing satellite communication services; beam B 2 The covered area is S 2 Coverage area S 2 The wave position P corresponds to 4 ～P 6 So that beam B 2 For para-wave position P 4 ～P 6 Scanning is performed and beam B 2 Can be directed to the covered wave position P 4 ～P 6 Providing satellite communication services; ... Beam B m The covered area is S m Coverage area S m The wave position P corresponds to n-2 ～P n So that beam B m For para-wave position P n-2 ～P n Scanning is performed and beam B m Can be directed to the covered wave position P n-2 ～P n Satellite communication services are provided. The satellite 110 may thus switch between corresponding wave positions by m beams and provide satellite communication services for corresponding respective wave positions in a time division manner.
Thus, the processor determines, for each pel U1-UL, the spectral features Fd corresponding to pels U1-UL:
Fd＝[fd 1 ,fd 2 ,...,fd u ] T
wherein fd is x (x=1 to U) is the pixel U 1 Reflectivity data of different frequency bands corresponding to the reflection data. The different frequency bands may be, for example, reflectance data of different frequency bands generated by a surface reflectance sensor, or reflectance data of several frequency bands obtained after principal component analysis.
Since in this embodiment the variables to be determined include the abundance values corresponding to the end members of the 5 different land feature types (i.e., tree land feature type, building land feature type, air land feature type, hilly mountain land feature type, and water body feature type). Therefore u.gtoreq.5. In addition, in the following, the end members are subtended by different types of featuresFrequency band of corresponding reference spectrum characteristic and pixel U 1 The frequency bands of the corresponding spectral features are the same. For example, the reference spectrum features corresponding to the end members of different ground object types also comprise U frequency bands and are matched with the pixel U 1 The frequency bands of the spectral features of (a) are the same. And will not be described in detail herein.
And then the processor acquires the reference spectrum characteristics of the end member multi-effect with different ground object types respectively. The reference spectral feature may be predetermined, for example, by measurement. In this embodiment, the reference spectrum features corresponding to the end members of different feature types include, for example:
Reference spectral features F of forest end members corresponding to the type of forest land object 1 ：
F 1 ＝[f 11 ,f 12 ,...,f 1u ]Wherein f 1x And (x=1-u) is the reflectivity data of different frequency bands corresponding to the forest end members.
Reference spectral features F of building end members corresponding to types of building features 2 ：
F 2 ＝[f 21 ,f 22 ,...,f 2u ]Wherein f 2x (x=1 to u) is the reflectivity data of different frequency bands corresponding to the building end members.
Reference spectral features F of air-to-ground end members corresponding to air-to-ground object types 3 ：
F 3 ＝[f 31 ,f 32 ,...,f 3u ]Wherein f 3x (x=1 to u) is the reflectivity data of different frequency bands corresponding to the air-ground end members.
Reference spectral features F of hilly mountain end members corresponding to hilly mountain feature types 4 ：
F 4 ＝[f 41 ,f 42 ,...,f 4u ]Wherein f 4x (x=1 to u) is the reflectivity data of different frequency bands corresponding to the end members of hilly and mountain lands.
Reference spectrum characteristic F of water body end member corresponding to water body ground object type 5 ：
F 5 ＝[f 51 ,f 52 ,...,f 5u ]Wherein f 5x And (x=1-u) is the reflectivity data of different frequency bands corresponding to the water body end members.
The satellite 200 then builds an equation according to the following formula:
k 1 F 1 +k 2 F 2 +k 3 F 3 +k 4 F 4 +k 5 F 5 =fd (equation 3)
Wherein k is 1 The abundance value of the forest end member in the pixel U1; k (k) 2 The abundance value of the building end member in the pixel U1; k (k) 3 Is the abundance value of the air-ground end member in the pixel U1; k (k) 4 The abundance value of the end member in the pixel U1 is the hilly and mountain area; k (k) 5 Is the abundance value of the water body end member in the pixel U1.
Thus, the computing device in satellite 200 will determine the spectral feature Fd of pel U1, the reference spectral feature F of the forest end member 1 Reference spectral features F of building end members 2 Reference spectral features F of air-to-ground end members 3 Reference spectral features F of hilly mountain end members 4 Reference spectral features F of water body end members 5 Substituting the above formula 1 to obtain the abundance value k 1 ～k 5 . Satellite 200 is determining an abundance value k 1 ～k 5 Then, the abundance value k of the end member corresponding to the tree land feature type, the building land feature type and the hilly and mountain land feature type in the pixel U1 can be determined 1 、k 2 And k 4 . Then, satellite 200 will have an abundance value k 1 、k 2 And k 4 The sum is taken as the abundance value of the ground object type of signal attenuation in the pixel U1.
With reference to the same operations as described above, satellite 200 determines the abundance value of the ground object type of signal attenuation in all pixels U2 to UL associated with the target area.
Optionally, the operation of acquiring a second image corresponding to the plurality of pixels with the unmanned vehicle and determining a model based on the feature type information to determine a plurality of feature type vectors corresponding to the feature type includes: inputting second images corresponding to the plurality of pixels acquired by the unmanned vehicle into a ground object type information determining model; and outputting a plurality of terrain type vectors from the terrain type information determination model, wherein the terrain type information determination model is a convolutional neural network model.
Specifically, the processor inputs a second image corresponding to the plurality of pixels acquired by the unmanned vehicle to a preset ground object type information determination model after obtaining the second image. The feature type information determining model is a convolutional neural network model, as shown in fig. 5. Thus, the feature type information determination model can output a feature type vector Q corresponding to the feature type 1 . Wherein Q is 1 ＝[q 11 ,q 12 ,q 13 ,q 14 ,q 15 ] T 。
Wherein q 11 Representing the proportion of the tree land object type in the surrounding environment of the driving node P1; q 12 Representing the proportion of the building ground object type in the surrounding environment of the driving node P1; q 13 Representing the proportion of the ground object type in the surrounding environment of the driving node P1; q 14 Representing the proportion of the hilly mountain land object type in the surrounding environment of the driving node P1; q 15 Representing the specific gravity of the water body ground object type in the surrounding environment of the driving node P1.
Thus, the feature type vector Q corresponding to the traveling nodes P1 to PL 1 ～Q L Can be determined by the above-described operation steps.
Optionally, the method further comprises: based on the plurality of clutter type vectors, an average of clutter type vectors is determined.
Specifically, a plurality of feature type vectors Q corresponding to the feature types are determined at the processor 1 ～Q L Can determine the average value of a plurality of feature type vectors corresponding to the feature typeThe calculation formula is as follows:
wherein the method comprises the steps of，Q i Represents any one of a plurality of feature type vectors.
Optionally, the operation of determining the score value corresponding to the plurality of pixels based on the abundance value of the end member in the plurality of pixels and the plurality of feature type vectors corresponding to the feature type includes: determining distance parameters corresponding to the plurality of pixels based on the abundance values of the end members in the plurality of pixels and the average value of the ground object type vectors; and determining score values corresponding to the plurality of pixels according to a preset logistic regression model and distance parameters corresponding to the plurality of pixels.
Optionally, the method further comprises: and re-extracting the plurality of pixel points serving as sampling points under the condition that the score values corresponding to the plurality of pixel points are smaller than a preset score threshold value. Specifically, in the case where the score values corresponding to the plurality of pixels are smaller than the preset score threshold value, it is explained that the plurality of pixel points as the sampling points are not qualified, and therefore, it is necessary to re-extract the plurality of pixel points as the sampling points.
According to the first aspect of the embodiment, the technical effect that the ground object type information determined based on the plurality of pixel points can be more accurate by selecting the plurality of qualified pixel points serving as sampling points is achieved.
Further, as shown with reference to fig. 1B and 1C, according to a second aspect of the present embodiment, there is provided a storage medium. The storage medium includes a stored program, wherein the method of any one of the above is performed by a processor when the program is run.
Therefore, according to the embodiment, the technical effect that the ground object type information determined based on the plurality of pixel points can be more accurate by selecting the plurality of qualified pixel points serving as sampling points is achieved.
It should be noted that, for simplicity of description, the foregoing method embodiments are all described as a series of acts, but it should be understood by those skilled in the art that the present invention is not limited by the order of acts described, as some steps may be performed in other orders or concurrently in accordance with the present invention. Further, those skilled in the art will also appreciate that the embodiments described in the specification are all preferred embodiments, and that the acts and modules referred to are not necessarily required for the present invention.
From the description of the above embodiments, it will be clear to a person skilled in the art that the method according to the above embodiments may be implemented by means of software plus the necessary general hardware platform, but of course also by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method according to the embodiments of the present invention.
Example 2
Fig. 8 shows an apparatus 800 for improving accuracy of determining feature type information according to the first aspect of the present embodiment, the apparatus 800 corresponding to the method according to the first aspect of embodiment 1. Referring to fig. 8, the apparatus 800 includes: an image acquisition module 810 for acquiring a first image corresponding to a target area based on a satellite; an abundance value determining module 820, configured to extract a plurality of pixels as sampling points in the first image, and determine abundance values of end members in the plurality of pixels; a feature type vector determining module 830, configured to collect second images corresponding to the plurality of pixels using the unmanned vehicle, and determine a plurality of feature type vectors corresponding to the feature type based on the feature type information determining model; a score value determining module 840, configured to determine score values corresponding to the plurality of pixels based on the abundance values of the end members in the plurality of pixels and a plurality of ground object type vectors corresponding to the object types; and a first determining module 850, configured to indicate that the plurality of pixels can be used as sampling points if the score values corresponding to the plurality of pixels are greater than a preset score threshold.
Optionally, the abundance value determination module 820 includes: the extraction module is used for extracting a plurality of pixels serving as sampling points from the received remote sensing image; the spectrum characteristic acquisition module is used for acquiring spectrum characteristics corresponding to the pixels, wherein the spectrum characteristics are used for indicating reflectivity data of different frequency bands corresponding to the pixels; the reference spectrum characteristic acquisition module is used for acquiring reference spectrum characteristics corresponding to end members of different ground object types, wherein the end members are used for indicating to only comprise pixels of one ground object type, and the reference spectrum characteristics comprise reflectivity data of different frequency bands corresponding to the end members of the different ground object types; and the abundance value determining submodule is used for determining the abundance value of the ground object type with signal attenuation in the pixel according to the frequency spectrum characteristic and the reference frequency spectrum characteristic corresponding to the end members of different ground object types.
Optionally, the feature type vector determining module 830 includes: the image input module is used for inputting second images which are acquired by the unmanned vehicle and correspond to the plurality of pixels into the ground object type information determining model; and a vector output module for outputting a plurality of ground object type vectors from the ground object type information determination model, wherein the ground object type information determination model is a convolutional neural network model.
Optionally, the apparatus 800 comprises: based on the plurality of clutter type vectors, an average of clutter type vectors is determined.
Optionally, the score value determination module 840 includes: the distance parameter determining module is used for determining distance parameters corresponding to the pixels based on the abundance values of the end members in the pixels and the average value of the ground object type vectors; and the score value determining module is used for determining score values corresponding to the pixels according to a preset logistic regression model and distance parameters corresponding to the pixels.
Optionally, the apparatus 800 comprises: and the second judging module is used for re-extracting the plurality of pixel points serving as sampling points under the condition that the score values corresponding to the plurality of pixels are smaller than a preset score threshold value.
Therefore, according to the embodiment, the technical effect that the ground object type information determined based on the plurality of pixel points can be more accurate by selecting the plurality of qualified pixel points serving as sampling points is achieved.
Example 3
Fig. 9 shows an apparatus 900 for improving accuracy of determining feature type information according to the first aspect of the present embodiment, the apparatus 900 corresponding to the method according to the first aspect of embodiment 1. Referring to fig. 9, the apparatus 900 includes: a processor 910; and a memory 920 coupled to the processor 910 for providing instructions to the processor 910 for processing the following processing steps: acquiring a first image corresponding to a target area based on a satellite; extracting a plurality of pixels serving as sampling points from a first image, and determining abundance values of end members in the plurality of pixels; acquiring a second image corresponding to the plurality of pixels by using the unmanned vehicle, determining a model based on the ground feature type information, and determining a plurality of ground feature type vectors corresponding to the ground feature type; determining score values corresponding to the plurality of pixels based on the abundance values of the end members in the plurality of pixels and a plurality of feature type vectors corresponding to the feature types; and under the condition that the score value corresponding to the plurality of pixels is larger than a preset score threshold value, the plurality of pixels can be used as sampling points.
Optionally, the operation of extracting a plurality of pixels as sampling points in the remote sensing image and determining the abundance value of the end member in the plurality of pixels includes: extracting a plurality of pixels serving as sampling points from the received first image; determining spectral features corresponding to the pixels, wherein the spectral features are used for indicating reflectivity data of different frequency bands corresponding to the pixels; acquiring reference spectrum characteristics corresponding to end members of different ground object types, wherein the end members are used for indicating that pixels of only one ground object type are included, and the reference spectrum characteristics include reflectivity data of different frequency bands corresponding to the end members of different ground object types; and determining abundance values of the signal attenuated ground object types in the pixels according to the frequency spectrum characteristics and the reference frequency spectrum characteristics corresponding to the end members of the different ground object types.
Optionally, the operation of acquiring a second image corresponding to the plurality of pixels with the unmanned vehicle and determining a model based on the feature type information to determine a plurality of feature type vectors corresponding to the feature type includes: inputting second images corresponding to the plurality of pixels acquired by the unmanned vehicle into a ground object type information determining model; and outputting a plurality of terrain type vectors from the terrain type information determination model, wherein the terrain type information determination model is a convolutional neural network model.
Optionally, the method further comprises: based on the plurality of clutter type vectors, an average of clutter type vectors is determined.
Optionally, the operation of determining the score value corresponding to the plurality of pixels based on the abundance value of the end member in the plurality of pixels and the plurality of feature type vectors corresponding to the feature type includes: determining distance parameters corresponding to the plurality of pixels based on the abundance values of the end members in the plurality of pixels and the average value of the ground object type vectors; and determining score values corresponding to the plurality of pixels according to a preset logistic regression model and distance parameters corresponding to the plurality of pixels.
Optionally, the method further comprises: and re-extracting the plurality of pixel points serving as sampling points under the condition that the score values corresponding to the plurality of pixel points are smaller than a preset score threshold value.
Therefore, according to the embodiment, the technical effect that the ground object type information determined based on the plurality of pixel points can be more accurate by selecting the plurality of qualified pixel points serving as sampling points is achieved.
The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
In the foregoing embodiments of the present invention, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.
In the several embodiments provided in the present application, it should be understood that the disclosed technology content may be implemented in other manners. The above-described embodiments of the apparatus are merely exemplary, and the division of the units, such as the division of the units, is merely a logical function division, and may be implemented in another manner, for example, multiple units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interfaces, units or modules, or may be in electrical or other forms.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The foregoing is merely a preferred embodiment of the present invention and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present invention, which are intended to be comprehended within the scope of the present invention.
Claims (10)
1. A method of improving accuracy in determining terrain type information, comprising:
acquiring a first image corresponding to a target area based on a satellite;
extracting a plurality of pixels serving as sampling points from the first image, and determining abundance values of end members in the plurality of pixels;
acquiring second images corresponding to the plurality of pixels by using an unmanned vehicle, determining a model based on the ground feature type information, and determining a plurality of ground feature type vectors corresponding to the ground feature type;
determining score values corresponding to the plurality of pixels based on the abundance values of the end members in the plurality of pixels and a plurality of feature type vectors corresponding to the feature type; and
and under the condition that the score values corresponding to the plurality of pixels are larger than a preset score threshold value, the plurality of pixels can be used as sampling points.
2. The method of claim 1, wherein the operations of extracting a plurality of picture elements as sampling points in the first image and determining the abundance values of end members in the plurality of picture elements comprise:
Extracting a plurality of pixels serving as sampling points from the received first image;
determining spectral features corresponding to the pixels, wherein the spectral features are used for indicating reflectivity data of different frequency bands corresponding to the pixels;
acquiring reference spectrum characteristics corresponding to end members of different ground object types, wherein the end members are used for indicating pixels only comprising one ground object type, and the reference spectrum characteristics comprise reflectivity data of different frequency bands corresponding to the end members of different ground object types; and
and determining the abundance value of the ground object type of the signal attenuation in the pixel according to the frequency spectrum characteristic and the reference frequency spectrum characteristic corresponding to the end members of different ground object types.
3. The method of claim 1, wherein the operation of capturing a second image corresponding to the plurality of pixels with the unmanned vehicle and determining a model based on the feature type information, determining a plurality of feature type vectors corresponding to the feature type, comprises:
inputting second images corresponding to the plurality of pixels acquired by the unmanned vehicle to the ground object type information determination model; and
and outputting the plurality of feature type vectors from the feature type information determination model, wherein the feature type information determination model is a convolutional neural network model.
4. A method according to claim 3, further comprising: and determining an average value of the ground object type vectors based on the plurality of ground object type vectors.
5. The method of claim 4, wherein determining the fractional value corresponding to the plurality of picture elements based on the abundance value of the end member of the plurality of picture elements and a plurality of feature type vectors corresponding to the feature type comprises:
determining distance parameters corresponding to the plurality of pixels based on the abundance values of the end members in the plurality of pixels and the average value of the ground object type vectors; and
and determining score values corresponding to the pixels according to a preset logistic regression model and distance parameters corresponding to the pixels.
6. The method as recited in claim 5, further comprising: and re-extracting the plurality of pixel points serving as sampling points under the condition that the score values corresponding to the plurality of pixels are smaller than a preset score threshold value.
7. A storage medium comprising a stored program, wherein the method of any one of claims 1 to 6 is performed by a processor when the program is run.
8. An apparatus for improving accuracy in determining a terrain type information, comprising:
the image acquisition module is used for acquiring a first image corresponding to the target area based on a satellite;
the abundance value determining module is used for extracting a plurality of pixels serving as sampling points from the first image and determining abundance values of end members in the plurality of pixels;
the ground object type vector determining module is used for acquiring second images corresponding to the plurality of pixels by using the unmanned vehicle, determining a model based on ground object type information and determining a plurality of ground object type vectors corresponding to the ground object types;
the score value determining module is used for determining score values corresponding to the plurality of pixels based on the abundance values of the end members in the plurality of pixels and a plurality of ground object type vectors corresponding to the ground object types; and
the first judging module is used for indicating that the plurality of pixels can be used as sampling points under the condition that the score values corresponding to the plurality of pixels are larger than a preset score threshold value.
9. The apparatus of claim 8, wherein the abundance value determination module comprises:
the extraction module is used for extracting a plurality of pixels serving as sampling points from the received first image;
The spectrum characteristic acquisition module is used for acquiring spectrum characteristics corresponding to the pixels, wherein the spectrum characteristics are used for indicating reflectivity data of different frequency bands corresponding to the pixels;
the device comprises a reference spectrum characteristic acquisition module, a reference spectrum characteristic acquisition module and a detection module, wherein the reference spectrum characteristic acquisition module is used for acquiring reference spectrum characteristics corresponding to end members of different ground object types, the end members are used for indicating that pixels of one ground object type are included, and the reference spectrum characteristics include reflectivity data of different frequency bands corresponding to the end members of the different ground object types; and
and the abundance value determining submodule is used for determining the abundance value of the ground object type with signal attenuation in the pixel according to the frequency spectrum characteristic and the reference frequency spectrum characteristic corresponding to the end members of different ground object types.
10. An apparatus for improving accuracy in determining a terrain type information, comprising:
a processor; and
a memory, coupled to the processor, for providing instructions to the processor to process the following processing steps:
acquiring a first image corresponding to a target area based on a satellite;
extracting a plurality of pixels serving as sampling points from the first image, and determining abundance values of end members in the plurality of pixels;
Acquiring second images corresponding to the plurality of pixels by using an unmanned vehicle, determining a model based on the ground feature type information, and determining a plurality of ground feature type vectors corresponding to the ground feature type;
determining score values corresponding to the plurality of pixels based on the abundance values of the end members in the plurality of pixels and a plurality of feature type vectors corresponding to the feature type; and
and under the condition that the score values corresponding to the plurality of pixels are larger than a preset score threshold value, the plurality of pixels can be used as sampling points.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311535765.0A CN117576579A (en) | 2023-11-16 | 2023-11-16 | Method, device and storage medium for improving accuracy of determining feature type information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311535765.0A CN117576579A (en) | 2023-11-16 | 2023-11-16 | Method, device and storage medium for improving accuracy of determining feature type information |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117576579A true CN117576579A (en) | 2024-02-20 |
Family
ID=89887523
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311535765.0A Pending CN117576579A (en) | 2023-11-16 | 2023-11-16 | Method, device and storage medium for improving accuracy of determining feature type information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117576579A (en) |
-
2023
- 2023-11-16 CN CN202311535765.0A patent/CN117576579A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111080526B (en) | Method, device, equipment and medium for measuring and calculating farmland area of aerial image | |
| CN112541395A (en) | Target detection and tracking method and device, storage medium and electronic device | |
| KR20210035269A (en) | Object detection using multiple neural networks trained on different image fields | |
| CN113012215B (en) | Space positioning method, system and equipment | |
| CN114611635B (en) | Object identification method and device, storage medium and electronic device | |
| CN111899279A (en) | Method and device for detecting motion speed of target object | |
| CN111291646A (en) | People flow statistical method, device, equipment and storage medium | |
| Awad | Toward robust segmentation results based on fusion methods for very high resolution optical image and lidar data | |
| CN111832515A (en) | Dense pedestrian detection method, medium, terminal and device | |
| CN113743151A (en) | Method and device for detecting road surface sprinkled object and storage medium | |
| CN115424264A (en) | Panorama segmentation method, related device, electronic equipment and storage medium | |
| CN111695627A (en) | Road condition detection method and device, electronic equipment and readable storage medium | |
| CN112633114B (en) | Unmanned aerial vehicle inspection intelligent early warning method and device for building change event | |
| CN114155278A (en) | Target tracking and related model training method, related device, equipment and medium | |
| CN111860431B (en) | Method and device for identifying object in image, storage medium and electronic device | |
| CN117576579A (en) | Method, device and storage medium for improving accuracy of determining feature type information | |
| US10977772B1 (en) | Unmanned aircraft system (UAS) detection and assessment via temporal intensity aliasing | |
| CN116611977A (en) | Method and device for tracing and rescuing path of disjunctor | |
| KR102099816B1 (en) | Method and apparatus for collecting floating population data on realtime road image | |
| CN113469130A (en) | Shielded target detection method and device, storage medium and electronic device | |
| Blasch et al. | Compression induced image quality degradation in terms of NIIRS | |
| CN113378719A (en) | Lane line recognition method and device, computer equipment and storage medium | |
| CN117523419B (en) | Method, device and storage medium for improving accuracy of determining feature type information | |
| CN109241811B (en) | Scene analysis method based on image spiral line and scene target monitoring system using same | |
| CN115994934B (en) | Data time alignment method and device and domain controller |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication |