Unmanned aerial vehicle remote sensing surveying and mapping image processing method
By generating step functions and sharpening kernels, the image noise and quality problems are solved, and the clarity and applicability of the image are improved.
Patent Information
- Application Number
- CN202411472333.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-22
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-10-22
AI Technical Summary
Due to poor wind resistance and insufficient focal length of the digital camera, the drone remote sensing image has problems such as noise, brightness and tone differences, which affects the image quality and may deteriorate during transmission and storage. Preprocessing is required to improve image quality. .
By collecting remote sensing mapping images taken by drones, generating step functions and change curves, building surveying and mapping data sets, calculating the mapping right and non-values, generating sharpening kernels, sharpening the images, and enhancing the edges and details of pixel points.
Effectively reduce image noise and distortion, improve image clarity and quality, and make the image more suitable for further analysis and application.
Smart Images

Figure CN119006337A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of image processing, and in particular to an image processing method for remote sensing and mapping of unmanned aerial vehicles. Background Art
[0002] UAV remote sensing mapping technology has the advantages of small size, strong timeliness, flexible operation and low cost, and is widely used in disaster monitoring and environmental reconnaissance. However, due to the low flying altitude, small size, light weight and poor wind resistance of UAVs, as well as insufficient focal length of digital cameras, the images taken often have problems such as noise, brightness difference and hue difference. In particular, the noise problem will greatly affect the edges and contours of the image. In addition, the image may also degrade during transmission and storage, affecting the image quality. Therefore, before the application of remote sensing images, a series of preprocessing is required to improve the image quality and meet the needs of subsequent analysis and application. Summary of the invention
[0003] In order to solve the above problems, the present invention proposes an image processing method for UAV remote sensing mapping.
[0004] The technical solution of the present invention is: a method for processing images from unmanned aerial vehicle remote sensing and mapping comprises the following steps:
[0005] S1, collecting remote sensing mapping images taken by drones, generating a step function for each row of the remote sensing mapping images, and determining the corresponding change curve;
[0006] S2, using the variation curve to generate a sharpening kernel;
[0007] S3. Use a sharpening kernel to enhance the pixels of the remote sensing image.
[0008] Furthermore, S1 includes the following sub-steps:
[0009] S11, collect remote sensing mapping images taken by drones and build mapping data sets;
[0010] S12, generating a corresponding mapping yes / no value according to the probability that the pixel value of each pixel point belongs to the mapping data set;
[0011] S13, generating a step function for each row according to the mapping yes / no value of each row of pixels;
[0012] S14. Generate a change curve according to the step function value of each row.
[0013] The beneficial effect of the above further scheme is that in the present invention, in S14, the number of rows of the remote sensing image is used as the horizontal coordinate, and the step function value corresponding to each row is used as the value of the vertical coordinate to generate a change curve. The method of S11-S14 can be used to effectively analyze the pixel points of the remote sensing image, determine whether the mapping value of each pixel point is 0 or 1, thereby generating a step function for each row, and the change curve constructed thereby can reflect the pixel value of the remote sensing image.
[0014] Furthermore, S11 includes the following sub-steps:
[0015] S111, constructing a first depth space vector, a second depth space vector and a third depth space vector for the pixel points of the remote sensing image;
[0016] S112, performing a dot product operation on every two depth space vectors of the pixel point, and taking an average of three operation values obtained after the dot product operation as the depth space value of the pixel point;
[0017] S113. The depth space values of all pixel points are used as a surveying and mapping data set.
[0018] In S112, specifically: perform a dot product operation on the first depth space vector and the second depth space vector to obtain a first operation value, perform a dot product operation on the second depth space vector and the third depth space vector to obtain a second operation value, perform a dot product operation on the first depth space vector and the third depth space vector to obtain a third operation value, and perform an average operation on the first operation value, the second operation value, and the third operation value as the depth space value.
[0019] Furthermore, in S111, the red intensity and green intensity of the pixel are spliced as the first depth space vector of the pixel, the red intensity and blue intensity of the pixel are spliced as the second depth space vector of the pixel, and the green intensity and blue intensity of the pixel are spliced as the third depth space vector of the pixel.
[0020] Further, in S12, the pixel point mapping is non-value The calculation formula is:
[0021] ;
[0022] In the formula, Indicates The probability that the pixel value of a pixel belongs to the surveying and mapping dataset.
[0023] Further, in S13, Row step function The expression is:
[0024] ;
[0025] In the formula, Indicates The number of pixels in the row whose value is not 1 is plotted. Indicates The first pixel in the row is the pixel value of a non-valued pixel. Indicates No. The first mapping is the pixel value of the pixel point that is not 1. Indicates No. The first mapping is the pixel value of the pixel point that is not 1. Indicates The number of pixels in the row that are not zero is plotted. Indicates The first pixel in the row is the pixel value of a non-zero pixel. Indicates No. The first mapping is the pixel value of the non-zero pixel point, Indicates No. The first mapping is the pixel value of the non-zero pixel.
[0026] Furthermore, S2 includes the following sub-steps:
[0027] S21, generating the size of the sharpening kernel according to the variation curve;
[0028] S22, calculating the sharpening center value;
[0029] S23. Use the sharpening center value as the value of the sharpening kernel diagonal, set the value above the diagonal to 1, and set the value below the diagonal to -1, to generate a sharpening kernel.
[0030] The beneficial effect of the above further scheme is: in the present invention, excessive sharpening may cause problems such as jagged edges, increased noise or color distortion in the image. Therefore, when performing image sharpening, it is necessary to select appropriate sharpening parameters according to the specific situation. The sharpening kernel is a special matrix that can be used for image sharpening operations. It enhances the edges and details of the image by performing convolution operations on each pixel of the image and its neighborhood, making the image look clearer and more vivid. The present invention sets the sharpening kernel to be a square matrix, so the value of the diagonal on the square matrix is first determined by using the change curve, the upper right value is set to 1, and the lower left value is set to -1.
[0031] Furthermore, in S21, the number of rows and columns of the sharpening kernel is the same;
[0032] The number of rows of the sharpening kernel The calculation formula is:
[0033] ;
[0034] In the formula, Indicates rounding up. It represents the horizontal coordinate of the pixel with the maximum pixel value in the remote sensing image. It represents the vertical coordinate of the pixel with the maximum pixel value in the remote sensing image. It represents the horizontal coordinate of the pixel point with the minimum pixel value in the remote sensing image. It represents the vertical coordinate of the pixel point with the minimum pixel value in the remote sensing image. Indicates the length of the remote sensing image. Indicates the width of the remote sensing image.
[0035] Further, in S22, the sharpening center value The calculation formula is:
[0036] ;
[0037] In the formula, Indicates the change curve A maximum value, Indicates The value corresponding to the horizontal coordinate of the maximum value minus 1, Indicates The value corresponding to the abscissa of the maximum value plus 1, Indicates the number of maximum values of the change curve, Indicates the change curve A maximum value, Indicates The value corresponding to the abscissa of the minimum value minus 1, Indicates The value corresponding to the minimum value plus 1 on the horizontal axis is Indicates the number of minimum values of the variation curve.
[0038] For example, the fourth maximum The horizontal axis of is 3, then It should be the vertical coordinate value corresponding to the horizontal coordinate of 2. It should be the vertical coordinate value corresponding to the horizontal coordinate of 4.
[0039] The beneficial effects of the present invention are as follows: the present invention proposes a method for processing unmanned aerial vehicle remote sensing mapping images, determines a mapping data set by using the red, green and blue components of pixel points in the remote sensing mapping image, determines the step function of each row by using the relationship between each pixel point and the mapping data set, and thereby generates a change curve; generates a specific sharpening kernel by using the characteristic parameters of the change curve, performs image sharpening processing on the image, and the sharpening parameters are determined according to the specific conditions of the image, which can effectively reduce problems such as image distortion caused by over-sharpening; the present invention can enhance the edges and contours in the remote sensing mapping image so that the enhanced remote sensing mapping image looks clearer, which is convenient for users to further analyze and process the image. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] Figure 1 Flowchart of the image processing method for UAV remote sensing mapping. DETAILED DESCRIPTION
[0041] The embodiments of the present invention will be further described below in conjunction with the accompanying drawings.
[0042] like Figure 1 As shown, the present invention provides a method for processing an image of a UAV remote sensing image, comprising the following steps:
[0043] S1, collecting remote sensing mapping images taken by drones, generating a step function for each row of the remote sensing mapping images, and determining the corresponding change curve;
[0044] S2, using the variation curve to generate a sharpening kernel;
[0045] S3. Use a sharpening kernel to enhance the pixels of the remote sensing image.
[0046] In this embodiment of the present invention, S1 includes the following sub-steps:
[0047] S11, collect remote sensing mapping images taken by drones and build mapping data sets;
[0048] S12, generating a corresponding mapping yes / no value according to the probability that the pixel value of each pixel point belongs to the mapping data set;
[0049] S13, generating a step function for each row according to the mapping yes / no value of each row of pixels;
[0050] S14. Generate a change curve according to the step function value of each row.
[0051] In the present invention, in S14, the number of rows of the remote sensing image is used as the horizontal coordinate, and the step function value corresponding to each row is used as the value of the vertical coordinate to generate a change curve. The method of S11-S14 can be used to effectively analyze the pixel points of the remote sensing image, determine whether the mapping value of each pixel point is 0 or 1, and thus generate a step function for each row. The change curve constructed in this way can reflect the pixel value of the remote sensing image.
[0052] In the embodiment of the present invention, S11 includes the following sub-steps:
[0053] S111, constructing a first depth space vector, a second depth space vector and a third depth space vector for the pixel points of the remote sensing image;
[0054] S112, performing a dot product operation on every two depth space vectors of the pixel point, and taking an average of three operation values obtained after the dot product operation as the depth space value of the pixel point;
[0055] S113. The depth space values of all pixel points are used as a surveying and mapping data set.
[0056] In S112, specifically: perform a dot product operation on the first depth space vector and the second depth space vector to obtain a first operation value, perform a dot product operation on the second depth space vector and the third depth space vector to obtain a second operation value, perform a dot product operation on the first depth space vector and the third depth space vector to obtain a third operation value, and perform an average operation on the first operation value, the second operation value, and the third operation value as the depth space value.
[0057] In an embodiment of the present invention, in S111, the red intensity and green intensity of the pixel are spliced as the first depth space vector of the pixel, the red intensity and blue intensity of the pixel are spliced as the second depth space vector of the pixel, and the green intensity and blue intensity of the pixel are spliced as the third depth space vector of the pixel.
[0058] In the embodiment of the present invention, in S12, the mapping of the pixel point is non-value The calculation formula is:
[0059] ;
[0060] In the formula, Indicates The probability that the pixel value of a pixel belongs to the surveying and mapping dataset.
[0061] In the embodiment of the present invention, in S13, Row step function The expression is:
[0062] ;
[0063] In the formula, Indicates The number of pixels in the row whose value is not 1 is plotted. Indicates The first pixel in the row is the pixel value of a non-valued pixel. Indicates No. The first mapping is the pixel value of the pixel point that is not 1. Indicates No. The first mapping is the pixel value of the pixel point that is not 1. Indicates The number of pixels in the row that are not zero is plotted. Indicates The first pixel in the row is the pixel value of a non-zero pixel. Indicates No. The first mapping is the pixel value of the non-zero pixel point, Indicates No. The first mapping is the pixel value of the non-zero pixel.
[0064] In this embodiment of the present invention, S2 includes the following sub-steps:
[0065] S21, generating the size of the sharpening kernel according to the variation curve;
[0066] S22, calculating the sharpening center value;
[0067] S23. Use the sharpening center value as the value of the sharpening kernel diagonal, set the value above the diagonal to 1, and set the value below the diagonal to -1, to generate a sharpening kernel.
[0068] In the present invention, excessive sharpening may cause problems such as jagged edges, increased noise or color distortion in the image. Therefore, when performing image sharpening, it is necessary to select appropriate sharpening parameters according to the specific situation. The sharpening kernel is a special matrix that can be used for image sharpening operations. It enhances the edges and details of the image by performing convolution operations on each pixel of the image and its neighborhood, making the image look clearer and more vivid. The present invention sets the sharpening kernel as a square matrix, so the value of the diagonal on the square matrix is first determined by using the change curve, the upper right value is set to 1, and the lower left value is set to -1.
[0069] In the embodiment of the present invention, in S21, the number of rows and columns of the sharpening kernel are the same;
[0070] The number of rows of the sharpening kernel The calculation formula is:
[0071] ;
[0072] In the formula, Indicates rounding up. It represents the horizontal coordinate of the pixel with the maximum pixel value in the remote sensing image. It represents the vertical coordinate of the pixel with the maximum pixel value in the remote sensing image. It represents the horizontal coordinate of the pixel point with the minimum pixel value in the remote sensing image. It represents the vertical coordinate of the pixel point with the minimum pixel value in the remote sensing image. Indicates the length of the remote sensing image. Indicates the width of the remote sensing image.
[0073] In the embodiment of the present invention, in S22, the sharpening center value The calculation formula is:
[0074] ;
[0075] In the formula, Indicates the change curve A maximum value, Indicates The value corresponding to the horizontal coordinate of the maximum value minus 1, Indicates The value corresponding to the abscissa of the maximum value plus 1, Indicates the number of maximum values of the change curve, Indicates the change curve A maximum value, Indicates The value corresponding to the abscissa of the minimum value minus 1 is Indicates The value corresponding to the minimum value plus 1 on the horizontal axis is Indicates the number of minimum values of the variation curve.
[0076] For example, the fourth maximum The horizontal axis of is 3, then It should be the vertical coordinate value corresponding to the horizontal coordinate of 2. It should be the vertical coordinate value corresponding to the horizontal coordinate of 4.
[0077] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.
Claims
1. A method for processing images from unmanned aerial vehicle remote sensing imaging, characterized in that: The following steps are involved: S1, collecting remote sensing mapping images taken by drones, generating a step function for each row of the remote sensing mapping images, and determining the corresponding change curve; S2, using the change curve to generate a sharpening kernel; S3. Use a sharpening kernel to enhance the pixels of the remote sensing image.
2. The method for processing images from unmanned aerial vehicle remote sensing imaging according to claim 1, characterized in that: The S1 comprises the following sub-steps: S11, collect remote sensing mapping images taken by drones and build mapping data sets; S12, generating a corresponding mapping yes / no value according to the probability that the pixel value of each pixel point belongs to the mapping data set; S13, generating a step function for each row according to the mapping yes / no value of each row of pixels; S14. Generate a change curve according to the step function value of each row.
3. The method for processing images from unmanned aerial vehicle remote sensing imaging according to claim 2, characterized in that: The S11 comprises the following sub-steps: S111, constructing a first depth space vector, a second depth space vector and a third depth space vector for the pixel points of the remote sensing image; S112, performing a dot product operation on every two depth space vectors of the pixel point, and taking an average of three operation values obtained after the dot product operation as the depth space value of the pixel point; S113. The depth space values of all pixel points are used as a surveying and mapping data set.
4. The method for processing images from unmanned aerial vehicle remote sensing imaging according to claim 3, characterized in that: In S111, the red intensity and green intensity of the pixel are spliced as the first depth space vector of the pixel, the red intensity and blue intensity of the pixel are spliced as the second depth space vector of the pixel, and the green intensity and blue intensity of the pixel are spliced as the third depth space vector of the pixel.
5. The method for processing images from unmanned aerial vehicle remote sensing imaging according to claim 2, characterized in that: In the S12, the pixel point mapping is non-value The calculation formula is: ; In the formula, Indicates The probability that the pixel value of a pixel belongs to the surveying and mapping dataset.
6. The method for processing images from unmanned aerial vehicle remote sensing imaging according to claim 2, characterized in that: In the S13, Row step function The expression is: ; In the formula, Indicates The number of pixels in the row whose value is not 1 is plotted. Indicates The first pixel in the row is the pixel value of a non-valued pixel. Indicates No. The first mapping is the pixel value of the pixel point that is not 1. Indicates No. The first mapping is the pixel value of the pixel point that is not 1. Indicates The number of pixels in the row that are not zero is plotted. Indicates The first pixel in the row is the pixel value of a non-zero pixel. Indicates No. The first mapping is the pixel value of the non-zero pixel point, Indicates No. The first mapping is the pixel value of the non-zero pixel.
7. The method for processing images from unmanned aerial vehicle remote sensing imaging according to claim 1, characterized in that: The S2 comprises the following sub-steps: S21, generating the size of the sharpening kernel according to the variation curve; S22, calculating the sharpening center value; S23. Use the sharpening center value as the value of the sharpening kernel diagonal, set the value above the diagonal to 1, and set the value below the diagonal to -1, to generate a sharpening kernel.
8. The method for processing images from unmanned aerial vehicle remote sensing imaging according to claim 7, characterized in that: In S21, the number of rows and columns of the sharpening kernel is the same; The number of rows of the sharpening kernel The calculation formula is: ; In the formula, Indicates rounding up. It represents the horizontal coordinate of the pixel with the maximum pixel value in the remote sensing image. It represents the vertical coordinate of the pixel with the maximum pixel value in the remote sensing image. It represents the horizontal coordinate of the pixel point with the minimum pixel value in the remote sensing image. It represents the vertical coordinate of the pixel point with the minimum pixel value in the remote sensing image. Indicates the length of the remote sensing image. Indicates the width of the remote sensing image.
9. The method for processing images from unmanned aerial vehicle remote sensing imaging according to claim 7, characterized in that: In S22, the center value is sharpened The calculation formula is: ; In the formula, Indicates the change curve A maximum value, Indicates The value corresponding to the horizontal coordinate of the maximum value minus 1, Indicates The value corresponding to the abscissa of the maximum value plus 1, Indicates the number of maximum values of the change curve, Indicates the change curve A maximum value, Indicates The value corresponding to the abscissa of the minimum value minus 1, Indicates The value corresponding to the minimum value plus 1 on the horizontal axis is Indicates the number of minimum values of the variation curve.
Citation Information
Patent Citations
Image enhancement in the mosaic domain
CN101490694A
Sharpened image identification method based on multi-resolution overshoot effect measurement
CN105138984A
Remote sensing surveying and mapping image enhancement method
CN115908186A
Sky-sky-ground landslide monitoring method and system based on multi-source data
CN118171918A
Method and device for deblurring out-of-focus blurred images
US20180276796A1
Cited By
Unmanned aerial vehicle remote sensing surveying and mapping image processing method
CN119648765A
A method for processing UAV remote sensing mapping images
CN119648765B
