CN117422654A - Remote sensing image color homogenizing method, device, equipment and storage medium - Google Patents
Remote sensing image color homogenizing method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN117422654A CN117422654A CN202311385513.4A CN202311385513A CN117422654A CN 117422654 A CN117422654 A CN 117422654A CN 202311385513 A CN202311385513 A CN 202311385513A CN 117422654 A CN117422654 A CN 117422654A
- Authority
- CN
- China
- Prior art keywords
- remote sensing
- sensing image
- leveled
- target
- image
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 92
- 238000003860 storage Methods 0.000 title claims abstract description 13
- 238000013507 mapping Methods 0.000 claims abstract description 84
- 238000012545 processing Methods 0.000 claims abstract description 59
- 238000012937 correction Methods 0.000 claims description 51
- 230000001186 cumulative effect Effects 0.000 claims description 50
- 230000002159 abnormal effect Effects 0.000 claims description 42
- 230000008569 process Effects 0.000 claims description 36
- 230000002452 interceptive effect Effects 0.000 claims description 15
- 238000005070 sampling Methods 0.000 claims description 15
- 230000005855 radiation Effects 0.000 claims description 14
- 238000006243 chemical reaction Methods 0.000 claims description 9
- 238000007499 fusion processing Methods 0.000 claims description 8
- 230000011218 segmentation Effects 0.000 claims description 8
- 230000005856 abnormality Effects 0.000 claims description 7
- 238000001514 detection method Methods 0.000 claims description 7
- 230000000694 effects Effects 0.000 description 14
- 238000000265 homogenisation Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 238000004891 communication Methods 0.000 description 5
- 238000009826 distribution Methods 0.000 description 4
- 238000005315 distribution function Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012876 topography Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 239000000443 aerosol Substances 0.000 description 1
- 238000012271 agricultural production Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Chemical compound O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4023—Scaling of whole images or parts thereof, e.g. expanding or contracting based on decimating pixels or lines of pixels; based on inserting pixels or lines of pixels
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/001—Texturing; Colouring; Generation of texture or colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration using histogram techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
- Facsimile Image Signal Circuits (AREA)
Abstract
The invention relates to the technical field of remote sensing image processing, and discloses a remote sensing image color homogenizing method, device, equipment and storage medium, wherein the method comprises the following steps: carrying out preset stretching treatment on the remote sensing image to be leveled to obtain a target remote sensing image to be leveled; performing downsampling treatment on the reference image and the target remote sensing image to be leveled to obtain a reference image thumbnail and a remote sensing image thumbnail to be leveled; determining a gray level mapping relation table based on the reference image thumbnail and the remote sensing image thumbnail to be leveled; and carrying out color homogenizing operation on the remote sensing image to be subjected to color homogenizing based on the gray level mapping relation table. According to the invention, the gray level mapping relation table is determined based on the reference image thumbnail and the remote sensing image thumbnail to be leveled, and the remote sensing image to be leveled is leveled based on the gray level mapping relation table, so that the technical problem of low operation efficiency in the prior art that the remote sensing image is leveled by manual and professional leveling processing software is solved.
Description
Technical Field
The present invention relates to the field of remote sensing image processing technologies, and in particular, to a remote sensing image color homogenizing method, device, equipment, and storage medium.
Background
In recent years, remote sensing images are widely applied to the fields of agricultural production, environmental monitoring, natural resource management and the like due to the advantages of higher spatial resolution, temporal resolution, spectral resolution and the like, and gradually become an important data source for human auxiliary production decisions. However, factors such as different satellite sensors, different photographing times, different photographing scenes, etc. make the task of comprehensive analysis of a large-span area a great challenge. Therefore, how to use the light and color homogenizing technology to ensure the color consistency between images as much as possible is a key to solve the problems.
In the prior art, remote sensing images can be subjected to color homogenization through manual and professional color homogenization processing software (such as ENVI, QMosaic, photoshop tools) and the like, and the interactive processing method can reduce the color difference between the images to a certain extent, but has the defects of large workload and high prior knowledge requirement on operators, so that the operation efficiency of the remote sensing image color homogenization method is lower.
The foregoing is provided merely for the purpose of facilitating understanding of the technical solutions of the present invention and is not intended to represent an admission that the foregoing is prior art.
Disclosure of Invention
The invention mainly aims to provide a remote sensing image color homogenizing method, device, equipment and storage medium, and aims to solve the technical problems of low operation efficiency in the prior art when remote sensing images are subjected to color homogenizing through manual and professional color homogenizing processing software.
In order to achieve the above object, the present invention provides a remote sensing image color homogenizing method, which includes:
carrying out preset stretching treatment on the remote sensing image to be leveled to obtain a target remote sensing image to be leveled;
performing downsampling treatment on the reference image and the target remote sensing image to be leveled in a preset sampling mode to obtain a reference image thumbnail and a remote sensing image thumbnail to be leveled;
determining a gray level mapping relation table based on the reference image thumbnail and the remote sensing image thumbnail to be leveled, wherein the gray level mapping relation table stores a corresponding relation between pixel gray values of the reference image and the remote sensing image to be leveled;
and carrying out color homogenizing operation on the remote sensing image to be subjected to color homogenizing based on the gray level mapping relation table.
Optionally, the step of performing preset stretching treatment on the remote sensing image to be leveled to obtain the target remote sensing image to be leveled includes:
carrying out preset stretching treatment on the remote sensing image to be leveled in a preset image stretching mode, and eliminating abnormal pixel values in the remote sensing image to be leveled to obtain a stretched remote sensing image to be leveled;
and performing format conversion on the stretched remote sensing image to be leveled to obtain a target remote sensing image to be leveled.
Optionally, the step of determining a gray level mapping relation table based on the reference image thumbnail and the remote sensing image thumbnail to be leveled includes:
acquiring a reference histogram corresponding to the reference image thumbnail and a histogram corresponding to the remote sensing image thumbnail to be leveled;
respectively carrying out histogram equalization processing on the reference histogram and the histogram to obtain a target reference histogram and a target histogram;
determining a reference cumulative histogram corresponding to the reference image thumbnail according to the target reference histogram;
determining a target cumulative histogram corresponding to the remote sensing image thumbnail to be leveled according to the target histogram;
a gray level mapping table is determined based on the target reference histogram, the target histogram, the reference cumulative histogram, and the target cumulative histogram.
Optionally, the step of determining a gray level mapping relation table based on the target reference histogram, the target histogram, the reference cumulative histogram, and the target cumulative histogram includes:
determining a color level form corresponding to the reference image based on the target reference histogram and the reference cumulative histogram;
Determining a target color level form corresponding to the target remote sensing image to be leveled based on the target histogram and the target cumulative histogram;
and determining a gray level mapping relation table based on the color level table and the target color level table.
Optionally, the step of performing the dodging operation on the remote sensing image to be dodged based on the gray level mapping relation table includes:
searching a mapping gray value corresponding to each pixel gray value in the remote sensing image to be uniformly colored based on the gray level mapping relation table;
and carrying out color homogenizing operation on the remote sensing image to be subjected to color homogenizing according to the mapping gray value.
Optionally, after the step of performing the dodging operation on the remote sensing image to be dodged based on the gray level mapping relationship table, the method further includes:
acquiring a target uniform-color remote sensing image after uniform color, and performing anomaly detection on the target uniform-color remote sensing image;
when abnormality in the target uniform-color remote sensing image is detected, extracting an abnormal region in the target uniform-color remote sensing image based on an interactive segmentation large model;
and determining the remote sensing image in the abnormal region as a new remote sensing image to be leveled, and returning to the step of carrying out preset stretching treatment on the remote sensing image to be leveled to obtain a target remote sensing image to be leveled.
Optionally, before the step of performing the preset stretching treatment on the remote sensing image to be leveled to obtain the target remote sensing image to be leveled, the method further includes:
performing radiation calibration processing on the initial remote sensing image based on a preset radiation calibration mode to obtain a remote sensing image after radiation calibration;
performing atmospheric correction processing on the remote sensing image subjected to radiometric calibration through a preset atmospheric correction model to obtain an atmospheric corrected remote sensing image;
performing geometric correction processing on the remote sensing image subjected to the atmospheric correction to obtain a remote sensing image subjected to the geometric correction;
and carrying out data fusion processing on the remote sensing image subjected to geometric correction to obtain the remote sensing image to be leveled.
In addition, in order to achieve the above purpose, the present invention further provides a remote sensing image color homogenizing device, which includes:
the image stretching processing module is used for carrying out preset stretching processing on the remote sensing image to be leveled to obtain a target remote sensing image to be leveled;
the downsampling processing module is used for downsampling the reference image and the target remote sensing image to be leveled in a preset sampling mode to obtain a reference image thumbnail and a remote sensing image thumbnail to be leveled;
the mapping relation table establishing module is used for determining a gray level mapping relation table based on the reference image thumbnail and the remote sensing image thumbnail to be leveled, and the gray level mapping relation table stores the corresponding relation between pixel gray values of the reference image and the remote sensing image to be leveled;
And the image homogenizing module is used for carrying out homogenizing operation on the remote sensing image to be homogenized based on the gray level mapping relation table.
In addition, in order to achieve the above purpose, the present invention further provides a remote sensing image color homogenizing device, which includes: the remote sensing image color homogenizing device comprises a memory, a processor and a remote sensing image color homogenizing program which is stored in the memory and can run on the processor, wherein the remote sensing image color homogenizing program is configured to realize the steps of the remote sensing image color homogenizing method.
In addition, in order to achieve the above objective, the present invention further provides a storage medium, where a remote sensing image color-homogenizing program is stored, and the remote sensing image color-homogenizing program implements the steps of the remote sensing image color-homogenizing method described above when executed by a processor.
In the invention, a preset stretching treatment is carried out on a remote sensing image to be leveled to obtain a target remote sensing image to be leveled; downsampling the reference image and the target remote sensing image to be leveled in a preset sampling mode to obtain a reference image thumbnail and a remote sensing image thumbnail to be leveled; determining a gray level mapping relation table based on the reference image thumbnail and the remote sensing image thumbnail to be leveled, wherein the gray level mapping relation table stores the corresponding relation between pixel gray values of the reference image and the remote sensing image to be leveled; carrying out color homogenizing operation on the remote sensing image to be subjected to color homogenizing based on the gray level mapping relation table; compared with the prior art, the remote sensing image is subjected to the color matching by using the manual and professional color matching processing software, and the prior knowledge of an operator is high, because the reference image and the target remote sensing image to be subjected to the color matching processing are subjected to the down-sampling processing, the reference image thumbnail and the remote sensing image thumbnail to be subjected to the color matching are obtained, the gray level mapping relation table storing the corresponding relation between the pixel gray values of the reference image and the remote sensing image to be subjected to the color matching is determined based on the reference image thumbnail and the remote sensing image thumbnail to be subjected to the color matching processing, and the remote sensing image to be subjected to the color matching operation based on the gray level mapping relation table, the technical problems that the remote sensing image is subjected to the color matching by using the manual and professional color matching processing software in the prior art, and the operation efficiency is low are solved.
Drawings
FIG. 1 is a schematic structural diagram of a remote sensing image color homogenizing device in a hardware operating environment according to an embodiment of the present invention;
FIG. 2 is a flowchart of a remote sensing image color matching method according to a first embodiment of the present invention;
FIG. 3 is a flowchart of a remote sensing image color matching method according to a second embodiment of the present invention;
FIG. 4 is a schematic diagram of a remote sensing image color matching process according to a second embodiment of the remote sensing image color matching method of the present invention;
FIG. 5 is a flowchart of a remote sensing image color matching method according to a third embodiment of the present invention;
FIG. 6 is a schematic diagram showing a remote sensing image color-homogenizing effect according to a third embodiment of the remote sensing image color-homogenizing method of the present invention;
fig. 7 is a block diagram of a remote sensing image color-homogenizing device according to a first embodiment of the present invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Referring to fig. 1, fig. 1 is a schematic structural diagram of a remote sensing image color homogenizing device in a hardware operation environment according to an embodiment of the present invention.
As shown in fig. 1, the remote sensing image color homogenizing device may include: a processor 1001, such as a central processing unit (Central Processing Unit, CPU), a communication bus 1002, a user interface 1003, a network interface 1004, a memory 1005. Wherein the communication bus 1002 is used to enable connected communication between these components. The user interface 1003 may include a Display, an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may further include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a Wireless interface (e.g., a Wireless-Fidelity (Wi-Fi) interface). The Memory 1005 may be a high-speed random access Memory (Random Access Memory, RAM) or a stable nonvolatile Memory (NVM), such as a disk Memory. The memory 1005 may also optionally be a storage device separate from the processor 1001 described above.
It will be appreciated by those skilled in the art that the configuration shown in fig. 1 is not limiting of the remote sensing image homogenizing apparatus and may include more or fewer components than shown, or may be combined with certain components, or may be arranged in different components.
As shown in fig. 1, the memory 1005 as a storage medium may include an operating system, a network communication module, a user interface module, and a remote sensing image shading program.
In the remote sensing image color homogenizing device shown in fig. 1, the network interface 1004 is mainly used for performing data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the remote sensing image color homogenizing device can be arranged in the remote sensing image color homogenizing device, and the remote sensing image color homogenizing device calls a remote sensing image color homogenizing program stored in the memory 1005 through the processor 1001 and executes the remote sensing image color homogenizing method provided by the embodiment of the invention.
An embodiment of the invention provides a remote sensing image color homogenizing method, referring to fig. 2, fig. 2 is a schematic flow chart of a first embodiment of the remote sensing image color homogenizing method.
In this embodiment, the remote sensing image color homogenizing method includes the following steps:
Step S10: and carrying out preset stretching treatment on the remote sensing image to be leveled to obtain the target remote sensing image to be leveled.
It should be noted that, the execution subject of the method of this embodiment may be a remote sensing image homogenizing device for performing a homogenizing process on a remote sensing image of the earth captured by a satellite, or another remote sensing image homogenizing system that can implement the same or similar functions and includes the remote sensing image homogenizing device. The remote sensing image color homogenizing method provided in this embodiment and the following embodiments will be specifically described with reference to a remote sensing image color homogenizing system (hereinafter referred to as a system).
It should be understood that the remote sensing image to be leveled may be an image shot by a satellite and needing to be leveled. Correspondingly, the target remote sensing image to be leveled can be a remote sensing image obtained after the remote sensing image to be leveled is stretched.
It is understood that the predetermined stretching process may be a process of stretching the remote sensing image to remove extreme pixel values in the image.
In practical application, the embodiment can stretch the remote sensing image to be leveled to obtain the target remote sensing image to be leveled, and the points with abnormal brightness or abnormal darkness in the image can be removed by stretching the remote sensing image to be leveled to eliminate the image caused by extreme values in the image to be leveled to the remote sensing image.
Further, in order to reduce the computational complexity of the remote sensing image color homogenizing process, the step S10 includes: carrying out preset stretching treatment on the remote sensing image to be leveled in a preset image stretching mode, and eliminating abnormal pixel values in the remote sensing image to be leveled to obtain a stretched remote sensing image to be leveled; and performing format conversion on the stretched remote sensing image to be leveled to obtain a target remote sensing image to be leveled.
The preset image stretching method may be a method of stretching a remote sensing image. Specifically, in this embodiment, a gaussian 3 σ stretching manner may be used to stretch the remote sensing image to be uniformly colored.
It should be understood that the abnormal pixel values may be extremely high pixel values and extremely low pixel values in the remote sensing image to be leveled. In practical applications, there are usually some dead pixels (i.e., points corresponding to the above-mentioned abnormal pixel values) with extremely high or extremely low pixel values in the remote sensing image captured by the satellite, and these points can be understood as noise in the image. Because the points with extremely high or extremely low pixel values can be in an extremely bright or extremely dark state in the remote sensing image, the points can cause the overall brightness or darkness of the image of the remote sensing image after the uniform color, and in order to eliminate the influence of the points on the uniform color effect of the remote sensing image to be uniform, the embodiment can stretch and convert the remote sensing image to be uniform to remove extremely high or extremely low noise and prevent the influence on the uniform color effect of the remote sensing image to be uniform.
It can be understood that the format conversion may be converting the original remote sensing image to be leveled into a remote sensing image with a target format, where the target format may be 8 bits.
In a specific implementation, the embodiment may perform stretching treatment on the remote sensing image to be leveled by adopting a gaussian 3 sigma stretching manner, wherein sigma represents a probability in mathematics, a 3 sigma default remote sensing image accords with normal distribution, and the coverage area of 3 sigma may be 99.74% of the whole image, at this time, points corresponding to an extremely high value and an extremely low value of a pixel in the image may be treated by the remaining probability, that is, abnormal pixel values in the remote sensing image to be leveled are removed, so as to obtain the stretched remote sensing image to be leveled, so as to prevent the influence of noise with extremely high or extremely low pixel values on the stretching process of the remote sensing image to be leveled. In order to reduce the calculation complexity of the image homogenizing process, format conversion can be performed on the stretched remote sensing image to be homogenized, and the stretched remote sensing image to be homogenized can be converted into an 8-bit target remote sensing image to be homogenized, namely, the stretched remote sensing image to be homogenized is reduced by 8 times, so that the calculation complexity is reduced.
Further, since the parameters of the satellite hardware may be different when the earth is photographed, and meanwhile, the cloud layer is observed through a long-term atmospheric window in the photographing process, the cloud layer may be affected by water vapor, dust, cloud layer, and the like, in order to eliminate the influence of external factors such as system errors and atmospheric environment on the color homogenizing effect of the remote sensing image, before step S10, the method further includes:
Step S01: and carrying out radiation calibration processing on the initial remote sensing image based on a preset radiation calibration mode to obtain a remote sensing image after radiation calibration.
It should be noted that the initial remote sensing image may be an unprocessed remote sensing image captured by a satellite.
It should be understood that the above-mentioned preset radiation calibration method may be a method of performing radiation calibration processing on a remote sensing image, and in this embodiment, the preset radiation calibration method may be a formula in a satellite file corresponding to a satellite that photographs the remote sensing image. Correspondingly, the radiometric calibration process can be to calibrate the radiometric degree of the remote sensing image so as to realize quantitative remote sensing process. The radiation calibration process may also be referred to as a calibration process, the main purpose of which is to ensure the accuracy of the sensor's acquisition of the remote sensing data.
In practical application, satellites for shooting remote sensing images have corresponding satellite files when the satellites are released, wherein the satellite files generally have a formula and parameters for storing corresponding radiometric calibration of the satellites, and the embodiment can perform radiometric calibration processing on an initial remote sensing image based on the formula and parameters for radiometric calibration to obtain the radiometric calibrated remote sensing image.
Step S02: and carrying out atmosphere correction processing on the remote sensing image subjected to the radiometric calibration through a preset atmosphere correction model to obtain the remote sensing image subjected to the atmosphere correction.
It can be understood that the above-mentioned preset atmospheric correction model is a model for performing atmospheric correction processing on the remote sensing image after radiometric calibration, and in this embodiment, the atmospheric correction processing may be performed on the remote sensing image after radiometric calibration through the 6s atmospheric correction model.
The above-mentioned atmosphere correction process may be a process for eliminating atmospheric interference in the remote sensing image.
Specifically, the embodiment can obtain the altitude and azimuth angle of the satellite for shooting the initial remote sensing image, the thickness of the aerosol corresponding to the high layer of the shot area and the current atmosphere, and perform the atmosphere correction on the remote sensing image after the radiometric calibration according to the 6s atmosphere correction model through the parameters, so as to remove the influence of the atmosphere and obtain the remote sensing image after the atmosphere correction.
Step S03: and performing geometric correction processing on the remote sensing image subjected to the atmospheric correction to obtain a remote sensing image subjected to the geometric correction.
It should be appreciated that the geometric correction process described above may be a process that eliminates the effects of topography relief distortion. In practical application, the embodiment can perform geometric correction processing on the remote sensing image after atmospheric correction through polynomial correction of the PRC, specifically, the embodiment can perform geometric correction processing on the remote sensing image after atmospheric correction according to the PRC file obtained after satellite shooting is completed and local DEM data, so as to remove the influence of relief distortion of topography and obtain the remote sensing image after geometric correction.
Step S04: and carrying out data fusion processing on the remote sensing image subjected to geometric correction to obtain the remote sensing image to be leveled.
It can be appreciated that the above data fusion process may be a process of improving the resolution of the remote sensing image by means of fusion. According to the embodiment, the resolution ratio of the remote sensing image can be improved by carrying out data fusion processing on the remote sensing image subjected to geometric correction, so that the remote sensing image is clearer.
In a specific implementation, in order to eliminate satellite system errors, the embodiment may perform radiometric calibration processing on the initial remote sensing image according to radiometric calibration parameters stored in a satellite file in a satellite for capturing the initial remote sensing image, so as to obtain a radiometric calibrated remote sensing image. Meanwhile, in order to eliminate the influence of the atmospheric environment, the embodiment can perform atmospheric correction processing on the remote sensing image after radiometric calibration through a 6s atmospheric correction model to obtain the remote sensing image after atmospheric correction. In order to eliminate the influence of the relief distortion, the embodiment can perform geometric correction processing on the remote sensing image after the atmospheric correction through polynomial correction of the PRC, so as to obtain the remote sensing image after the geometric correction. In order to improve the definition of the remote sensing image, the embodiment can perform data fusion processing on the remote sensing image after geometric correction to obtain the remote sensing image to be uniformly colored.
Step S20: and carrying out downsampling treatment on the reference image and the target remote sensing image to be leveled in a preset sampling mode to obtain a reference image thumbnail and a remote sensing image thumbnail to be leveled.
It should be noted that the above-mentioned preset sampling manner may be a manner of performing downsampling processing on the remote sensing image. Specifically, the present embodiment may perform downsampling on the remote sensing image by nearest neighbor sampling, where nearest neighbor sampling may be a downsampling method that maps each pixel or vertex in the original image or geometry to a nearest neighbor pixel or vertex in the new image or geometry.
It should be appreciated that the downsampling process described above may be a process of obtaining a thumbnail of the remote sensing image.
It can be appreciated that the reference image may be an image corresponding to the region of the remote sensing image to be leveled, where the reference image in this embodiment may be a reference image with little or one scene in the region of the remote sensing image to be leveled. Specifically, unlike the existing reference template-based method, the embodiment can meet the color uniformity requirement of all images by only needing a small amount of thumbnail images of even one scene reference image, thereby greatly reducing the data constraint of the model input end.
It should be appreciated that the reference image thumbnail described above may be a thumbnail corresponding to the reference image. Correspondingly, the thumbnail of the remote sensing image to be leveled can be a thumbnail corresponding to the remote sensing image to be leveled. According to the scheme, the remote sensing images to be uniformly colored can be uniformly colored based on the thumbnail, so that the characteristic distribution of the images is simplified, the robustness of the model is higher, the problem of chromatic aberration among images with larger space-time span can be solved, and meanwhile, the color uniformity efficiency is improved.
In a specific implementation, the embodiment may perform N times downsampling processing on the reference image and the remote sensing image to be leveled simultaneously by using a nearest neighbor sampling manner, so as to obtain a reference image thumbnail corresponding to the reference image and a remote sensing image thumbnail to be leveled corresponding to the remote sensing image to be leveled, where N may be adjusted according to the image size, and N may be 8 in the embodiment.
Step S30: and determining a gray level mapping relation table based on the reference image thumbnail and the remote sensing image thumbnail to be leveled, wherein the gray level mapping relation table stores a corresponding relation between pixel gray values of the reference image and the remote sensing image to be leveled.
It should be noted that the gray level mapping relationship table may be a mapping table storing a correspondence relationship between a pixel gray level value of the reference image and a pixel gray level value of the remote sensing image to be leveled. The pixel gray value can be the brightness value of each pixel point in the remote sensing image, and the range of the pixel gray value is 0-255.
Step S40: and carrying out color homogenizing operation on the remote sensing image to be subjected to color homogenizing based on the gray level mapping relation table.
Further, the step S40 may include: searching a mapping gray value corresponding to each pixel gray value in the remote sensing image to be uniformly colored based on the gray level mapping relation table; and carrying out color homogenizing operation on the remote sensing image to be subjected to color homogenizing according to the mapping gray value.
It should be understood that the above-mentioned mapped gray value may be a gray value corresponding to a gray level mapping relationship table of a gray value of each pixel in the remote sensing image to be leveled, where the gray value is a gray value of a reference image corresponding to a gray value of each pixel in the remote sensing image to be leveled.
In a specific implementation, the system can search the gray level of the reference image corresponding to the gray level of all the pixel points in the remote sensing image to be leveled in the gray level mapping relation table, if the gray level of a certain pixel point in the remote sensing image to be leveled is 1, the mapping gray level corresponding to the pixel point is found to be 0 in the gray level mapping relation table, at the moment, when the remote sensing image to be leveled is leveled, the gray level of the pixel point in the remote sensing image to be leveled can be changed from 1 to 0, and the gray level of the pixel point in the remote sensing image to be leveled is converted according to the mapping gray level corresponding to all the pixel points in the remote sensing image to be leveled, so as to realize the leveling operation of the remote sensing image to be leveled.
The embodiment discloses a method for carrying out preset stretching treatment on a remote sensing image to be leveled to obtain a target remote sensing image to be leveled; downsampling the reference image and the target remote sensing image to be leveled in a preset sampling mode to obtain a reference image thumbnail and a remote sensing image thumbnail to be leveled; determining a gray level mapping relation table based on the reference image thumbnail and the remote sensing image thumbnail to be leveled, wherein the gray level mapping relation table stores the corresponding relation between pixel gray values of the reference image and the remote sensing image to be leveled; carrying out color homogenizing operation on the remote sensing image to be subjected to color homogenizing based on the gray level mapping relation table; compared with the prior art, the remote sensing image is subjected to the color matching by using the manual and professional color matching software, and the prior knowledge of an operator is required to be high.
Referring to fig. 3, fig. 3 is a flowchart illustrating a remote sensing image color homogenizing method according to a second embodiment of the present invention.
Based on the above first embodiment, in order to pre-establish the gray level mapping relationship table to improve the color uniformity efficiency of the remote sensing image, in this embodiment, the step S20 includes:
step S301: and obtaining a reference histogram corresponding to the reference image thumbnail and a histogram corresponding to the remote sensing image thumbnail to be leveled.
It should be noted that, the histogram may be an image representing the frequency of each gray level in the remote sensing image, similar to a "histogram", and in practical application, statistical information of the image may be obtained through the histogram. Correspondingly, the reference histogram may be a histogram corresponding to the reference image thumbnail; the histogram is the histogram corresponding to the remote sensing image thumbnail to be leveled. In practical applications, the embodiment may calculate histograms for each channel of the reference image thumbnail and the remote sensing image thumbnail to be leveled, specifically, may obtain frequencies of occurrence of pixel values of each pixel point in the image on the current channel, and finally, connect the frequency fold to form a probability distribution function (i.e. the histogram), and may calculate a cumulative distribution function (i.e. cumulative histogram) according to the probability distribution function.
Step S302: and respectively carrying out histogram equalization processing on the reference histogram and the histogram to obtain a target reference histogram and a target histogram.
It should be understood that the histogram equalization process may be a process of changing the histogram to a uniform histogram by a transformation function and then modifying the original image according to the uniform histogram, thereby obtaining a new image having uniform gray distribution.
It can be understood that the target reference histogram may be a histogram obtained by performing histogram equalization processing on the reference histogram; correspondingly, the target histogram may be a histogram obtained by performing a histogram equalization process on the histogram.
Step S303: and determining a reference cumulative histogram corresponding to the reference image thumbnail according to the target reference histogram.
Note that the cumulative histogram may be a graph in which gray levels are represented by the horizontal axis and the number of pixels included in each gray level or below or the ratio of the number of pixels to the total number of pixels is represented by the vertical axis. The reference cumulative histogram may be a cumulative histogram corresponding to the reference image thumbnail.
Step S304: and determining a target cumulative histogram corresponding to the remote sensing image thumbnail to be leveled according to the target histogram.
It should be appreciated that the target cumulative histogram may be a cumulative histogram corresponding to a thumbnail of the remote sensing image to be leveled.
Step S305: a gray level mapping table is determined based on the target reference histogram, the target histogram, the reference cumulative histogram, and the target cumulative histogram.
Specifically, the step S305 may include: determining a color level form corresponding to the reference image based on the target reference histogram and the reference cumulative histogram; determining a target color level form corresponding to the target remote sensing image to be leveled based on the target histogram and the target cumulative histogram; and determining a gray level mapping relation table based on the color level table and the target color level table.
The color level table may be a table storing color (gray level) levels corresponding to each pixel in the reference image. In practical application, since the frequency number of each gray level in the remote sensing image is counted in the histogram, and the number of pixels of each gray level and below or the ratio of the number of pixels to the total number of pixels are counted in the cumulative histogram, the embodiment can determine the color level of each pixel point in the remote sensing image based on the histogram and the cumulative histogram, and establish a color level table.
It should be appreciated that the target color level form described above may be a form storing color levels to be matched. In practical application, the embodiment can respectively establish a reference histogram corresponding to the reference image thumbnail and a histogram corresponding to the remote sensing image thumbnail to be leveled according to the frequency of occurrence of the pixel value of each pixel point in the reference image thumbnail and the remote sensing image thumbnail to be leveled on the current channel, change the reference histogram and the histogram into uniform histograms through a transformation function, and then modify the original image according to the uniform histograms to obtain a target reference histogram and a target histogram with uniform gray level distribution. And finally, a gray level mapping relation table can be constructed based on the color level table and the target color level table.
In a specific implementation, referring to fig. 4, fig. 4 is a schematic diagram of a remote sensing image color homogenizing process in a second embodiment of the remote sensing image color homogenizing method of the present invention. As shown in fig. 4, in this solution, four modules may be provided, such as image preprocessing, stretching, converting, downsampling, abbreviated histogram matching mapping, and abnormal interactive processing. If the remote sensing image needs to be subjected to the dodging operation, firstly, a multi-element satellite remote sensing image (namely an initial remote sensing image) can be input into a first module (namely an image preprocessing module), wherein the model of the satellite can comprise: JL1, GF2, BJ3, GF7, etc., which are not limited in this embodiment. After receiving the initial remote sensing image, the image preprocessing module can respectively perform radiation calibration processing, atmospheric correction processing, geometric correction processing and image fusion processing on the initial remote sensing image to obtain a remote sensing image to be leveled, and outputs the remote sensing image to be leveled to a second module (namely, a stretching conversion downsampling module). And after the stretching conversion downsampling module receives the remote sensing image to be leveled, stretching the remote sensing image to be leveled in a Gaussian 3 sigma stretching mode to obtain a stretched remote sensing image to be leveled, and performing format conversion on the stretched remote sensing image to be leveled to convert the stretched remote sensing image to be leveled into an 8-bit target remote sensing image to be leveled. And then, respectively carrying out downsampling treatment on the reference image and the target image to be leveled through nearest neighbor sampling to obtain a reference image thumbnail and a remote sensing image thumbnail to be leveled, and outputting the reference image thumbnail and the remote sensing image thumbnail to be leveled to a thumbnail histogram matching mapping module. After receiving the reference image thumbnail and the remote sensing image thumbnail to be leveled, the thumbnail histogram matching mapping module can calculate histograms of the reference image thumbnail and the remote sensing image thumbnail to be leveled, calculate a cumulative distribution function, obtain a reference histogram and a reference cumulative histogram corresponding to the reference image thumbnail, a histogram and a target cumulative histogram corresponding to the remote sensing image thumbnail to be leveled, and calculate a gray level mapping relation table based on the reference histogram, the reference cumulative histogram, the histogram and the target cumulative histogram, so as to perform the color leveling operation on the remote sensing image to be leveled through the gray level mapping relation table.
According to the embodiment, the target reference histogram and the target histogram are obtained by carrying out histogram equalization processing on the reference histogram corresponding to the reference image thumbnail and the histogram corresponding to the remote sensing image thumbnail to be leveled, the reference cumulative histogram corresponding to the reference image thumbnail is determined according to the target reference histogram, the target cumulative histogram corresponding to the remote sensing image thumbnail to be leveled is determined according to the target histogram, and the gray level mapping relation table is determined based on the target reference histogram, the target histogram, the reference cumulative histogram and the target cumulative histogram, so that the remote sensing image to be leveled can be leveled directly through the gray level mapping relation table in the follow-up process, and the color leveling efficiency of the remote sensing image is improved.
Referring to fig. 5, fig. 5 is a flowchart of a remote sensing image color homogenizing method according to a third embodiment of the present invention.
Based on the above embodiments, in order to improve the efficiency of the color balancing process for the abnormal situation of the polar end, in this embodiment, after step S40, the method further includes:
step S50: and acquiring a target uniform-color remote sensing image after uniform color, and performing anomaly detection on the target uniform-color remote sensing image.
It should be noted that the target dodging remote sensing image may be a remote sensing image obtained after a dodging operation is performed on a remote sensing image to be dodged.
It should be appreciated that factors such as large span space-time differences, sensor differences, imaging environment differences, ground object coverage types of images to be leveled and reference images, and extreme unbalance in duty ratios may all cause problems of abnormal color leveling, and therefore it is necessary to provide a convenient interactive method to handle such extreme cases while avoiding the occurrence of abnormal situations as much as possible. The processing difficulty of the extreme cases is that the mask where the abnormal region is located is determined, the traditional scheme is that the method for creating the intelligent selection region by utilizing Photoshop is very sensitive to the pixel DN value, and the understanding of high-level semantics is lacking. To solve the defect, the present embodiment may extract a region of homogeneous color anomaly based on the interactive segmentation large model, and re-match the image of the anomaly region with homogeneous color to handle the extremely abnormal homogeneous color condition.
In practical application, the embodiment can perform anomaly detection on the target uniform-color remote sensing image after uniform color so as to detect whether a uniform-color abnormal region exists in the target uniform-color remote sensing image after uniform color.
Step S60: and when detecting that the target uniform-color remote sensing image is abnormal, extracting an abnormal region in the target uniform-color remote sensing image based on the interactive segmentation large model.
It will be appreciated that the above-described large interactive segmentation model may be a model for extracting abnormal regions in a remote sensing image, such as an ISAT model. The ISAT model in this embodiment is encapsulated with a SAM (Segment Anything Model) model.
The abnormal region in the target dodging remote sensing image may be a region with poor dodging effect or a region with a dodging color abnormality, which is not limited in this embodiment.
Step S70: and determining the remote sensing image in the abnormal region as a new remote sensing image to be leveled, and returning to the step of carrying out preset stretching treatment on the remote sensing image to be leveled to obtain a target remote sensing image to be leveled.
It should be understood that, in this embodiment, after the abnormal region in the target uniform-color remote sensing image is extracted, the remote sensing image in the abnormal region may be determined as a new remote sensing image to be uniform-color, and the process returns to the step of performing preset stretching processing on the remote sensing image to be uniform-color to obtain the target remote sensing image to be uniform-color, so as to perform the uniform-color operation again on the new remote sensing image to be uniform-color.
In a specific implementation, as shown in fig. 4, after the matching mapping module of the abbreviated histogram performs the dodging operation on the target dodging remote sensing image, the target dodging remote sensing image after dodging may be output to the fourth module (i.e. the abnormal interactive processing module), after the abnormal interactive processing module receives the target dodging remote sensing image after dodging, it may first determine whether the corresponding KL divergence exceeds the preset threshold, if yes, the abnormal mask may be obtained through the SAM model packaged in the ISAT model, that is, an abnormal region of the dodging abnormality is extracted from the target dodging remote sensing image after dodging, and the preset stretching process is performed on the target dodging remote sensing image to obtain the target dodging remote sensing image, so as to re-perform the dodging operation on the remote sensing image in the abnormal region, to normalize the color, and fuse the dodging image with the original image to obtain the complete remote sensing image to be dodged. Referring to fig. 6, fig. 6 is a schematic diagram showing a color-homogenizing effect of a remote sensing image according to a third embodiment of the remote sensing image color-homogenizing method of the present invention. Fig. 6 shows effect charts based on reference images and used for carrying out color homogenization on images to be homogenized through original image histogram matching, thumbnail histogram matching and SAM interactive matching modes, and as can be seen from the figures, the effect of carrying out color homogenization on the images to be homogenized through thumbnail histogram matching is better than that of the original image histogram matching modes, but if the effects of the images to be homogenized are homogenized through the original image histogram matching and thumbnail histogram matching modes, areas with color homogenization abnormality (such as the left lower part of the original image histogram matching and thumbnail histogram matching effect charts in fig. 6) are likely to appear, so that the embodiment can adopt the SAM interactive matching mode to carry out color homogenization operation again on the areas with color homogenization abnormality in the images to be homogenized, so as to obtain the images with better color homogenization effect.
According to the embodiment, the anomaly detection is carried out on the target uniform-color remote sensing image after uniform color, when the target uniform-color remote sensing image is abnormal, the abnormal area in the target uniform-color remote sensing image is extracted based on the interactive segmentation large model, the remote sensing image in the abnormal area is determined to be a new remote sensing image to be uniform-color, so that the new remote sensing image to be uniform-color is subjected to uniform-color operation again, uniform-color processing under the abnormal condition of the polar end is achieved, and the efficiency and the uniform-color effect of uniform-color processing are improved.
In addition, the embodiment of the invention also provides a storage medium, wherein the storage medium is stored with a remote sensing image homogenizing program, and the remote sensing image homogenizing program realizes the steps of the remote sensing image homogenizing method when being executed by a processor.
Referring to fig. 7, fig. 7 is a block diagram illustrating a remote sensing image color homogenizing device according to a first embodiment of the present invention.
As shown in fig. 7, the remote sensing image color homogenizing device provided by the embodiment of the invention includes:
the image stretching processing module 701 is configured to perform preset stretching processing on the remote sensing image to be leveled, so as to obtain a target remote sensing image to be leveled;
the downsampling processing module 702 is configured to downsample the reference image and the target remote sensing image to be leveled in a preset sampling manner, so as to obtain a thumbnail of the reference image and a thumbnail of the remote sensing image to be leveled;
A mapping relationship table establishing module 703, configured to determine a gray level mapping relationship table based on the reference image thumbnail and the remote sensing image thumbnail to be leveled, where the gray level mapping relationship table stores a correspondence between pixel gray values of the reference image and the remote sensing image to be leveled;
the image homogenizing module 704 is configured to perform a homogenizing operation on the remote sensing image to be homogenized based on the gray level mapping relationship table.
Further, the image stretching processing module 701 is further configured to perform a preset stretching process on the remote sensing image to be leveled in a preset image stretching manner, and reject abnormal pixel values in the remote sensing image to be leveled, so as to obtain a stretched remote sensing image to be leveled; and performing format conversion on the stretched remote sensing image to be leveled to obtain a target remote sensing image to be leveled.
Further, the image homogenizing module 704 is further configured to search a mapped gray value corresponding to the gray value of each pixel in the remote sensing image to be homogenized based on the gray level mapping relationship table; and carrying out color homogenizing operation on the remote sensing image to be subjected to color homogenizing according to the mapping gray value.
Further, the image stretching processing module 701 is further configured to perform a radiometric calibration process on the initial remote sensing image based on a preset radiometric calibration mode, so as to obtain a radiometric calibrated remote sensing image; performing atmospheric correction processing on the remote sensing image subjected to radiometric calibration through a preset atmospheric correction model to obtain an atmospheric corrected remote sensing image; performing geometric correction processing on the remote sensing image subjected to the atmospheric correction to obtain a remote sensing image subjected to the geometric correction; and carrying out data fusion processing on the remote sensing image subjected to geometric correction to obtain the remote sensing image to be leveled.
The remote sensing image color homogenizing device of the embodiment discloses that preset stretching treatment is carried out on a remote sensing image to be homogenized, and a target remote sensing image to be homogenized is obtained; downsampling the reference image and the target remote sensing image to be leveled in a preset sampling mode to obtain a reference image thumbnail and a remote sensing image thumbnail to be leveled; determining a gray level mapping relation table based on the reference image thumbnail and the remote sensing image thumbnail to be leveled, wherein the gray level mapping relation table stores the corresponding relation between pixel gray values of the reference image and the remote sensing image to be leveled; carrying out color homogenizing operation on the remote sensing image to be subjected to color homogenizing based on the gray level mapping relation table; compared with the prior art, the remote sensing image is subjected to the color matching by using the manual and professional color matching software, and the prior knowledge of an operator is required to be high.
Based on the first embodiment of the remote sensing image color homogenizing device of the present invention, a second embodiment of the remote sensing image color homogenizing device of the present invention is provided.
In this embodiment, the mapping relation table establishing module 703 is further configured to obtain a reference histogram corresponding to the reference image thumbnail and a histogram corresponding to the remote sensing image thumbnail to be leveled; respectively carrying out histogram equalization processing on the reference histogram and the histogram to obtain a target reference histogram and a target histogram; determining a reference cumulative histogram corresponding to the reference image thumbnail according to the target reference histogram; determining a target cumulative histogram corresponding to the remote sensing image thumbnail to be leveled according to the target histogram; a gray level mapping table is determined based on the target reference histogram, the target histogram, the reference cumulative histogram, and the target cumulative histogram.
Further, the mapping relation table establishing module 703 is further configured to determine a color level table corresponding to the reference image based on the target reference histogram and the reference cumulative histogram; determining a target color level form corresponding to the target remote sensing image to be leveled based on the target histogram and the target cumulative histogram; and determining a gray level mapping relation table based on the color level table and the target color level table.
According to the embodiment, the target reference histogram and the target histogram are obtained by carrying out histogram equalization processing on the reference histogram corresponding to the reference image thumbnail and the histogram corresponding to the remote sensing image thumbnail to be leveled, the reference cumulative histogram corresponding to the reference image thumbnail is determined according to the target reference histogram, the target cumulative histogram corresponding to the remote sensing image thumbnail to be leveled is determined according to the target histogram, and the gray level mapping relation table is determined based on the target reference histogram, the target histogram, the reference cumulative histogram and the target cumulative histogram, so that the remote sensing image to be leveled can be leveled directly through the gray level mapping relation table in the follow-up process, and the color leveling efficiency of the remote sensing image is improved.
Based on the above device embodiments, a third embodiment of the remote sensing image color homogenizing device is provided.
In this embodiment, the image homogenizing module 704 is further configured to obtain a target homogenized remote sensing image after being homogenized, and perform anomaly detection on the target homogenized remote sensing image; when abnormality in the target uniform-color remote sensing image is detected, extracting an abnormal region in the target uniform-color remote sensing image based on an interactive segmentation large model; and determining the remote sensing image in the abnormal region as a new remote sensing image to be leveled, and returning to the step of carrying out preset stretching treatment on the remote sensing image to be leveled to obtain a target remote sensing image to be leveled.
According to the embodiment, the anomaly detection is carried out on the target uniform-color remote sensing image after uniform color, when the target uniform-color remote sensing image is abnormal, the abnormal area in the target uniform-color remote sensing image is extracted based on the interactive segmentation large model, the remote sensing image in the abnormal area is determined to be a new remote sensing image to be uniform-color, so that the new remote sensing image to be uniform-color is subjected to uniform-color operation again, uniform-color processing under the abnormal condition of the polar end is achieved, and the efficiency and the uniform-color effect of uniform-color processing are improved.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
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.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented 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. read-only memory/random-access memory, 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.
The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the invention, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.
Claims (10)
1. The remote sensing image color homogenizing method is characterized by comprising the following steps of:
Carrying out preset stretching treatment on the remote sensing image to be leveled to obtain a target remote sensing image to be leveled;
performing downsampling treatment on the reference image and the target remote sensing image to be leveled in a preset sampling mode to obtain a reference image thumbnail and a remote sensing image thumbnail to be leveled;
determining a gray level mapping relation table based on the reference image thumbnail and the remote sensing image thumbnail to be leveled, wherein the gray level mapping relation table stores a corresponding relation between pixel gray values of the reference image and the remote sensing image to be leveled;
and carrying out color homogenizing operation on the remote sensing image to be subjected to color homogenizing based on the gray level mapping relation table.
2. The method for homogenizing a remote sensing image according to claim 1, wherein the step of performing a preset stretching process on the remote sensing image to be homogenized to obtain the target remote sensing image to be homogenized comprises the steps of:
carrying out preset stretching treatment on the remote sensing image to be leveled in a preset image stretching mode, and eliminating abnormal pixel values in the remote sensing image to be leveled to obtain a stretched remote sensing image to be leveled;
and performing format conversion on the stretched remote sensing image to be leveled to obtain a target remote sensing image to be leveled.
3. The remote sensing image homogenizing method of claim 1, wherein the step of determining a gray level mapping relation table based on the reference image thumbnail and the remote sensing image thumbnail to be homogenized comprises:
acquiring a reference histogram corresponding to the reference image thumbnail and a histogram corresponding to the remote sensing image thumbnail to be leveled;
respectively carrying out histogram equalization processing on the reference histogram and the histogram to obtain a target reference histogram and a target histogram;
determining a reference cumulative histogram corresponding to the reference image thumbnail according to the target reference histogram;
determining a target cumulative histogram corresponding to the remote sensing image thumbnail to be leveled according to the target histogram;
a gray level mapping table is determined based on the target reference histogram, the target histogram, the reference cumulative histogram, and the target cumulative histogram.
4. The remote sensing image shading method according to claim 3, wherein said step of determining a gray level mapping table based on said target reference histogram, said target histogram, said reference cumulative histogram and said target cumulative histogram comprises:
Determining a color level form corresponding to the reference image based on the target reference histogram and the reference cumulative histogram;
determining a target color level form corresponding to the target remote sensing image to be leveled based on the target histogram and the target cumulative histogram;
and determining a gray level mapping relation table based on the color level table and the target color level table.
5. The method for homogenizing remote sensing images according to claim 1, wherein the step of homogenizing the remote sensing images to be homogenized based on the gray level mapping relation table comprises the following steps:
searching a mapping gray value corresponding to each pixel gray value in the remote sensing image to be uniformly colored based on the gray level mapping relation table;
and carrying out color homogenizing operation on the remote sensing image to be subjected to color homogenizing according to the mapping gray value.
6. The method for homogenizing remote sensing images according to claim 1, wherein after the step of homogenizing the remote sensing images to be homogenized based on the gray level mapping relation table, the method further comprises:
acquiring a target uniform-color remote sensing image after uniform color, and performing anomaly detection on the target uniform-color remote sensing image;
when abnormality in the target uniform-color remote sensing image is detected, extracting an abnormal region in the target uniform-color remote sensing image based on an interactive segmentation large model;
And determining the remote sensing image in the abnormal region as a new remote sensing image to be leveled, and returning to the step of carrying out preset stretching treatment on the remote sensing image to be leveled to obtain a target remote sensing image to be leveled.
7. The method for homogenizing a remote sensing image according to claim 1, wherein before the step of performing a preset stretching process on the remote sensing image to be homogenized to obtain the target remote sensing image to be homogenized, the method further comprises:
performing radiation calibration processing on the initial remote sensing image based on a preset radiation calibration mode to obtain a remote sensing image after radiation calibration;
performing atmospheric correction processing on the remote sensing image subjected to radiometric calibration through a preset atmospheric correction model to obtain an atmospheric corrected remote sensing image;
performing geometric correction processing on the remote sensing image subjected to the atmospheric correction to obtain a remote sensing image subjected to the geometric correction;
and carrying out data fusion processing on the remote sensing image subjected to geometric correction to obtain the remote sensing image to be leveled.
8. A remote sensing image color homogenizing device, the device comprising:
the image stretching processing module is used for carrying out preset stretching processing on the remote sensing image to be leveled to obtain a target remote sensing image to be leveled;
The downsampling processing module is used for downsampling the reference image and the target remote sensing image to be leveled in a preset sampling mode to obtain a reference image thumbnail and a remote sensing image thumbnail to be leveled;
the mapping relation table establishing module is used for determining a gray level mapping relation table based on the reference image thumbnail and the remote sensing image thumbnail to be leveled, and the gray level mapping relation table stores the corresponding relation between pixel gray values of the reference image and the remote sensing image to be leveled;
and the image homogenizing module is used for carrying out homogenizing operation on the remote sensing image to be homogenized based on the gray level mapping relation table.
9. A remote sensing image homogenizing device, the device comprising: a memory, a processor and a remote sensing image shading program stored on the memory and operable on the processor, the remote sensing image shading program being configured to implement the steps of the remote sensing image shading method according to any one of claims 1 to 7.
10. A storage medium, wherein a remote sensing image homogenizing program is stored on the storage medium, and the remote sensing image homogenizing program, when executed by a processor, implements the steps of the remote sensing image homogenizing method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311385513.4A CN117422654B (en) | 2023-10-23 | 2023-10-23 | Remote sensing image color homogenizing method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311385513.4A CN117422654B (en) | 2023-10-23 | 2023-10-23 | Remote sensing image color homogenizing method, device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117422654A true CN117422654A (en) | 2024-01-19 |
| CN117422654B CN117422654B (en) | 2024-06-14 |
Family
ID=89524104
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311385513.4A Active CN117422654B (en) | 2023-10-23 | 2023-10-23 | Remote sensing image color homogenizing method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117422654B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10089554B1 (en) * | 2016-03-30 | 2018-10-02 | Descartes Labs, Inc. | Creating a boundary map |
| CN112907486A (en) * | 2021-03-18 | 2021-06-04 | 国家海洋信息中心 | Remote sensing image color matching method based on deep learning and color mapping |
| CN114298562A (en) * | 2021-12-29 | 2022-04-08 | 神彩科技股份有限公司 | Water environment monitoring and early warning method and device, computer equipment and storage medium |
| CN115760578A (en) * | 2022-11-28 | 2023-03-07 | 国网电力空间技术有限公司 | Image processing method and device, electronic equipment and storage medium |
| CN116152677A (en) * | 2022-11-30 | 2023-05-23 | 深圳市宇驰检测技术股份有限公司 | Remote sensing image intelligent interpretation method, device, equipment and storage medium |
| CN116228604A (en) * | 2023-05-09 | 2023-06-06 | 国家卫星海洋应用中心 | Satellite remote sensing optical image color homogenizing method suitable for polar region |
| CN116823628A (en) * | 2022-03-22 | 2023-09-29 | 佛吉亚歌乐电子(厦门)有限公司 | Image processing method and image processing device |
-
2023
- 2023-10-23 CN CN202311385513.4A patent/CN117422654B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10089554B1 (en) * | 2016-03-30 | 2018-10-02 | Descartes Labs, Inc. | Creating a boundary map |
| CN112907486A (en) * | 2021-03-18 | 2021-06-04 | 国家海洋信息中心 | Remote sensing image color matching method based on deep learning and color mapping |
| CN114298562A (en) * | 2021-12-29 | 2022-04-08 | 神彩科技股份有限公司 | Water environment monitoring and early warning method and device, computer equipment and storage medium |
| CN116823628A (en) * | 2022-03-22 | 2023-09-29 | 佛吉亚歌乐电子(厦门)有限公司 | Image processing method and image processing device |
| CN115760578A (en) * | 2022-11-28 | 2023-03-07 | 国网电力空间技术有限公司 | Image processing method and device, electronic equipment and storage medium |
| CN116152677A (en) * | 2022-11-30 | 2023-05-23 | 深圳市宇驰检测技术股份有限公司 | Remote sensing image intelligent interpretation method, device, equipment and storage medium |
| CN116228604A (en) * | 2023-05-09 | 2023-06-06 | 国家卫星海洋应用中心 | Satellite remote sensing optical image color homogenizing method suitable for polar region |
Non-Patent Citations (1)
| Title |
|---|
| 张荞 等: "基于直方图匹配的多源遥感影像匀色研究", 《地理空间信息》, 31 December 2020 (2020-12-31), pages 54 - 57 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117422654B (en) | 2024-06-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Biologically inspired image enhancement based on Retinex | |
| EP1600893B1 (en) | Radiometric calibration from a single image | |
| US7636098B2 (en) | Salience preserving image fusion | |
| CN112258579A (en) | Image processing method, image processing device, electronic equipment and computer readable storage medium | |
| WO2022000397A1 (en) | Low-illumination image enhancement method and apparatus, and computer device | |
| CN106940887B (en) | GF-4 satellite sequence image cloud and cloud shadow detection method | |
| CN112529807B (en) | Relative radiation correction method and device for satellite image | |
| CN111380836A (en) | Black and odorous water positioning method and device based on remote sensing image and related equipment | |
| CN113905183B (en) | Color difference correction method and device for wide dynamic range image | |
| CN112287904B (en) | Airport target identification method and device based on satellite images | |
| CN112200848B (en) | Depth camera vision enhancement method and system under low-illumination weak-contrast complex environment | |
| Díaz et al. | Enhanced gap fraction extraction from hemispherical photography | |
| CN117422654B (en) | Remote sensing image color homogenizing method, device, equipment and storage medium | |
| CN117095417A (en) | Screen shot form image text recognition method, device, equipment and storage medium | |
| JP4147155B2 (en) | Image processing apparatus and method | |
| CN115760578A (en) | Image processing method and device, electronic equipment and storage medium | |
| WO2021189460A1 (en) | Image processing method and apparatus, and movable platform | |
| JP4125191B2 (en) | Image processing apparatus and method | |
| US11334963B2 (en) | Devices, systems and methods for digital image analysis | |
| CN113505844A (en) | Label generation method, device, equipment, storage medium and program product | |
| Kamal et al. | Resoluting multispectral image using image fusion and CNN model | |
| CN117274584B (en) | Shadow processing method and device for remote sensing image, storage medium and terminal | |
| CN113643219B (en) | Image imaging method and device based on three-light fusion | |
| CN117252789B (en) | Shadow reconstruction method and device for high-resolution remote sensing image and electronic equipment | |
| CN117911281A (en) | CIELCH color space-based double-threshold remote sensing image shadow removal method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |