Product fusion method, system and equipment based on generalized vegetation coverage and storage medium

Abstract

The invention discloses a generalized vegetation coverage-based product fusion method, system and device, and a storage medium. The method comprises the steps of obtaining a satellite remote sensing image with a near-infrared band, a red band, a green band and a blue band; calculating a characteristic ratio index according to the near-infrared band, the red band, the green band and the blue band; constructing a linear transformation function according to the characteristic ratio index; setting a virtual minimum value vmin = kxmin of the characteristic ratio index; according to the linear transformation function and the virtual minimum value, constructing a generalized vegetation coverage-based power function; setting vmin + = vmin + rho; determining the characteristic power of the generalized vegetation coverage-based power function according to the condition that the generalized vegetation coverage-based power function value corresponding to the vmin + is equal to the vmin +; according to the principle of product enhancement, the red wave band and the green wave band are respectively enhanced by utilizing a generalized vegetation coverage-based power function with determined characteristic power. The standardization degree of the product enhancement processing process and the consistency of the processing result are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of remote sensing image fusion, and in particular to a product fusion method, system, computer device and computer-readable storage medium based on generalized vegetation cover. Background Art

[0002] With the development of multi-platform, multi-sensor, multi-weather, multi-temporal and multi-resolution remote sensing technologies, images with different spatial resolutions, temporal resolutions, spectral resolutions, etc. are becoming increasingly rich. In the past two decades or so, as a new direction in remote sensing image processing, remote sensing image spatio-temporal fusion has witnessed rapid development of various fusion technologies and achieved a series of new results. However, there is little research on spectral inter-fusion of multi-spectral images, mainly focusing on true color image simulation or vegetation enhancement processing.

[0003] With the popularization of remote sensing applications in various industries, visible light multispectral satellite remote sensing true color images have the excellent property of "what you see is what you get", and have become one of the most widely used remote sensing image types. However, there are inherent defects such as unnatural and untrue vegetation colors, which restrict their application effectiveness. How to effectively enhance the vegetation characteristics of visible light satellite remote sensing true color images is the key and difficult point in the processing of visible light satellite remote sensing true color images. Relevant researchers have made fruitful explorations, laying a theoretical and technical foundation for further solving related problems. Chen Chun et al. based on the primary products of satellite remote sensing, corrected the Rayleigh scattering of remote sensing data to make the color signal image close to the ground true color image (Chen Chun et al., Extraction and reproduction of color signals from remote sensing information sources, Science of Surveying and Mapping, January 2006, Vol. 31, No. 1; Han Xiuzhen et al., Research on the synthesis method and application of true color images of FY-3D satellite, Journal of Marine Meteorology, May 2019, Vol. 39, No. 2). You Jing et al. used the white balance method and color correction based on colorimetry to improve the vegetation characteristics of true color images and obtained more realistic true color images (You Jing et al., A white balance method for processing color multispectral images, Journal of Atmospheric and Environmental Optics, July 2012, Vol. 7, No. 4; Huang Honglian et al., True color synthesis of multispectral remote sensing images based on artificial targets, Infrared and Laser Engineering, November 2016, Vol. 45, No. 11). Fan Xuyan et al. based on the products after the secondary processing of remote sensing images, obtained relatively good true color images through the enhancement processing of the green band. In the early stage, mainly the overall weighted combination operation scheme of the green band and the near-infrared band was used to obtain a new green band (Fan Xuyan et al., A method for simulating true color fusion of remote sensing images based on principal component analysis, Journal of Surveying and Mapping Science and Technology, August 2006, Vol. 23, No. 4; Wang Haiyan et al., Discussion on the transformation and fusion methods of ALOS natural color images, Surveying and Mapping Technology Equipment, 2012, Vol. 14, No. 1; Shi Yuanli et al., Analysis of the applicability of high-resolution satellite remote sensing images for mapping, Bulletin of Surveying and Mapping, 2017, No. 12); later, it gradually developed into using the normalized difference vegetation index as a classification function to classify and weight the vegetation pixels of the image to obtain a new green band (Zhang Wei et al., A method for true color synthesis of multispectral images based on vegetation index, Geomatics & Spatial Information Technology, December 2010, Vol. 33, No. 6); recently, it has developed into using the normalized difference vegetation index to segment and perform Contourlet fusion on the green band and the near-infrared band to obtain a new green band image (Ding Huimei, Research on improving the naturalness of the color of multispectral remote sensing images using near-infrared, Master's thesis, 2016).

[0004] The evaluation technology for the vegetation enhancement effect of true-color images includes visual evaluation and quantitative evaluation. Visual evaluation: As we all know, although the original remote sensing true-color images are consistent with the ground features in terms of colors of water bodies, bare land, etc., the colors in the vegetation areas are dull and lack clarity in levels. Generally, true-color images need to enhance the vegetation to obtain true-color images that are consistent with the ground colors in the above-mentioned types of features. That is, the vegetation is based on green, and different categories and coverage degrees of vegetation present various greens with different shades and intensities; the water bodies are based on blue, and except for presenting green, yellow, black, etc. due to different components such as vegetation coverage on the water surface, high-concentration sediment, high pollution, etc., the main body is various blues with different shades and intensities; other bare lands such as rocks, bare soil, roads, residential areas, etc. are consistent with the rich colors on the ground, presenting various colors such as gray, black, white, red, orange, yellow, green, cyan, blue, purple, etc. Visually select typical feature categories such as water areas, bare land, vegetation, etc., and qualitatively compare the color changes between the original true-color images and the enhanced true-color images, and the effect of vegetation enhancement in true-color images can be visually evaluated. Quantitative evaluation: Quantitative evaluation is, to a certain extent, the quantification of the indicators in visual evaluation. For the enhancement of the vegetation characteristics of true-color images, not only the color of the vegetation in true-color images needs to be improved, but also the richness and clarity of the levels, details, etc. of the enhanced true-color images need to be ensured. Generally, the reconstructed images can be quantitatively evaluated from two aspects: One is the quantitative description and comparison of the vegetation color enhancement effect. Among the models describing color spaces such as RGB, CMYK, IHS, CIELab, etc., it is generally considered that the RGB three-primary color model is suitable for computer screen displays, etc., the CMYK printing model is suitable for color image printing outputs, and the IHS, CIELab color space models conform to the human eye visual perception mode in terms of color description. Based on this understanding, the general method used in quantitatively evaluating the effect of true-color images is: convert the remote sensing images described in the RGB three-primary color space into images described in the IHS or CIELab color space, read the chromaticity, saturation, intensity, etc. of the vegetation features before and after enhancement in these color spaces, and analyze their change trends and characteristics. The second is the statistics and comparison of the quality indicators of the enhanced true-color images. Generally speaking, the quality of image processing can be evaluated from three aspects: First, the information richness of the overall enhanced image and the vegetation area, which can be measured by entropy and joint entropy; Second, the color richness and brightness of the overall enhanced image and the vegetation area, which can be measured by the statistical features of the bands - maximum value, minimum value, mean value, variance, and the correlation indicators between bands - correlation coefficient, covariance, etc.; Third, the levels (edges), details (textures) and clarity of the overall enhanced image and the vegetation area can be measured by gradient, average gradient, etc. By comparing the differences in the indicators of the overall enhanced image and the vegetation area before and after enhancement, the change directions of spectral (gray level, hue) information, edge (level, difference) information, and texture (detail) information can be analyzed.

[0005] In view of the disadvantages of satellite remote sensing true color images, such as dull vegetation and other ground objects and unnatural colors, multi-index, multi-transformation, multi-mode, and multi-parameter natural color image vegetation feature enhancement methods have been developed, effectively improving the quality of true color images and enhancing the visual resolution and computer analysis resolution of true color images. The rich enhancement methods provide rich choices for different image processing technicians to carry out personalized processing of specific images. However, in the application of large-scale true color remote sensing image enhancement, the processing results are often required to have high consistency to facilitate the mapping and classification applications of image processing results. Obviously, there is a huge contradiction between personalized processing and the requirements of standardization and consistency. Summary of the Invention

[0006] The present invention provides a product fusion method, system, computer device, and computer-readable storage medium based on generalized vegetation cover, aiming to solve the contradiction between the personalized processing of a single image and the requirements of standardization and consistency processing of large-scale images in true color image enhancement. By determining a characteristic curve in the power function space based on generalized vegetation cover for true color image enhancement (a power function based on generalized vegetation cover for determining the characteristic power), the automatic calculation of the characteristic power is realized, reducing the artificial dependence on the selection of power transformation parameters and improving the standardization degree of the product enhancement processing process and the consistency of the processing results.

[0007] The first object of the present invention is to provide a product fusion method based on generalized vegetation cover.

[0008] The second object of the present invention is to provide a product fusion system based on generalized vegetation cover.

[0009] The third object of the present invention is to provide a computer device.

[0010] The fourth object of the present invention is to provide a computer-readable storage medium.

[0011] The first object of the present invention can be achieved by adopting the following technical solutions:

[0012] A product fusion method based on generalized vegetation cover, the method comprising:

[0013] Obtaining a satellite remote sensing image having near-infrared, red, green, and blue bands;

[0014] Performing spectral fusion between the near-infrared band and one or more of the red, green, and blue bands, using the fused band combination as the numerator and the corresponding pre-fusion band combination as the denominator to calculate the characteristic ratio index;

[0015] Constructing a linear transformation function according to the characteristic ratio index;

[0016] Set the virtual minimum value v of the characteristic ratio index min = kx min ; According to the linear transformation function and the virtual minimum value, construct a power function based on the generalized vegetation coverage; where, k ∈ [0, 1] is a given value, and x min is the minimum value of the characteristic ratio index x;

[0017] Set v min+ = v min + ρ; From the power function value based on the generalized vegetation coverage corresponding to v min+ being equal to v min+ determine the characteristic power of the power function based on the generalized vegetation coverage; where, ρ is a given value greater than 0 and close to 0;

[0018] According to the principle of multiplicative enhancement, use the power function based on the generalized vegetation coverage that determines the characteristic power to enhance the red band and the green band respectively; Synthesize the enhanced red band and green band with the blue band to obtain the true color image after vegetation enhancement;

[0019] Among them, the linear transformation function satisfies the conditions: h(c) = 1, and: when x > c, then h(x) > 1, when x < c, then h(x) < 1; c is the characteristic threshold of the characteristic ratio index x, and s and t are both constants.

[0020] Furthermore, s = 1, t = 1 - c; or, s = c, t = 0; or, s = 1, t = 0.

[0021] Furthermore, the constructing a power function based on the generalized vegetation coverage according to the linear transformation function and the virtual minimum value includes:

[0022] Construct the generalized vegetation coverage based on h(x) according to the linear transformation function and the virtual minimum value as:

[0023]

[0024] According to the generalized vegetation coverage based on h(x), the power function based on the generalized vegetation coverage is obtained as:

[0025]

[0026] Among them, x max is the maximum value of the characteristic ratio index x; n is the characteristic power to be determined, n > 0.

[0027] Furthermore, the constructing a power function based on the generalized vegetation coverage according to the linear transformation function and the virtual minimum value includes:

[0028] The normalized index NdhI corresponding to the linear transformation function h(x) is:

[0029]

[0030] Based on the normalized index NdhI and the virtual minimum value, the generalized vegetation coverage based on NdhI is constructed as:

[0031]

[0032] Based on the generalized vegetation coverage based on NdhI, the power function based on the generalized vegetation coverage is obtained as:

[0033]

[0034] where x max is the maximum value of the characteristic ratio index x; n is the characteristic power to be determined, and n > 0.

[0035] Furthermore, constructing the power function based on the generalized vegetation coverage according to the linear transformation function includes:

[0036] Based on the linear transformation function and the virtual minimum value, the generalized vegetation coverage based on h(x) is constructed as:

[0037]

[0038] The normalized index NdhI corresponding to the linear transformation function h(x) is:

[0039]

[0040] Based on the normalized index NdhI and the virtual minimum value, the generalized vegetation coverage based on NdhI is constructed as:

[0041]

[0042] Based on the generalized vegetation coverage based on h(x) and the generalized vegetation coverage based on NdhI, the power function based on the product of the two generalized vegetation coverages is obtained as:

[0043]

[0044] where x max is the maximum value of the characteristic ratio index x; n is the characteristic power to be determined, and n > 0.

[0045] Furthermore, constructing the power function based on the generalized vegetation coverage according to the linear transformation function and the virtual minimum value includes:

[0046] Based on the linear transformation function and the virtual minimum value, the generalized vegetation coverage based on h(x) is constructed as:

[0047]

[0048] The normalized exponential NdhI corresponding to the linear transformation function h(x) is as follows:

[0049]

[0050] Based on the normalized exponential NdhI and the virtual minimum value, the generalized vegetation coverage based on NdhI is constructed as:

[0051]

[0052] Based on the generalized vegetation coverage based on h(x) and the generalized vegetation coverage based on NdhI, the power function based on the average value of the two generalized vegetation coverages is obtained as:

[0053]

[0054] where x max is the maximum value of the characteristic ratio index x; n is the characteristic power to be determined, and n > 0.

[0055] Furthermore, according to the principle of multiplicative enhancement, the power function based on the generalized vegetation coverage for determining the characteristic power is used to enhance the red band and the green band respectively, including:

[0056] R′ = Z ρ (x)R

[0057] G′ = Z ρ (x)G

[0058] where R′ and G′ are the enhanced red band and green band respectively, and Z ρ (x) is the power function based on the generalized vegetation coverage for determining the characteristic power, and R and G are the red band and the green band respectively.

[0059] Furthermore, the characteristic threshold c ∈ [x min , x max or the characteristic threshold c ∈ [x w , x v ; where x max is the maximum value of the characteristic ratio index x, and x w , x v are the thresholds of pure water body and pure vegetation determined by human-computer interaction respectively.

[0060] The second object of the present invention can be achieved by adopting the following technical solutions:

[0061] A multiplicative fusion system based on generalized vegetation coverage, the system includes:

[0062] An acquisition module, configured to acquire satellite remote sensing images with near-infrared band, red band, green band and blue band;

[0063] A calculation module, configured to perform spectral fusion between the near-infrared band and one or more of the red band, green band and blue band, use the fused band combination as the numerator, and the corresponding band combination before fusion as the denominator to calculate the characteristic ratio index;

[0064] A first construction module, configured to construct a linear transformation function according to the characteristic ratio index;

[0065] A second construction module, configured to set the virtual minimum value v of the characteristic ratio index min = kx min ; construct a power function based on the generalized vegetation coverage according to the linear transformation function and the virtual minimum value; where k ∈ [0, 1] is a given value, and x min is the minimum value of the characteristic ratio index x;

[0066] A determination module, configured to set v min+ = v min + ρ; determine the characteristic power of the power function based on the generalized vegetation coverage from the fact that the power function value based on the generalized vegetation coverage corresponding to v min+ is equal to v min+ ; where ρ is a given value greater than 0 and close to 0;

[0067] A fusion module, configured to enhance the red band and the green band respectively by using the power function based on the generalized vegetation coverage with the determined characteristic power according to the principle of product enhancement; synthesize the enhanced red band and green band with the blue band to obtain a true color image with enhanced vegetation;

[0068] where the linear transformation function satisfies the conditions: h(c) = 1, and: when x > c, h(x) > 1, when x < c, h(x) < 1; c is the characteristic threshold of the characteristic ratio index x, and s and t are both constants.

[0069] The third object of the present invention can be achieved by adopting the following technical solution:

[0070] A computer device, including a processor and a memory for storing the executable program of the processor. When the processor executes the program stored in the memory, the product fusion method based on the generalized vegetation coverage described above is implemented.

[0071] The fourth object of the present invention can be achieved by adopting the following technical solution:

[0072] A computer-readable storage medium, storing a program, which when executed by a processor, implements the product fusion method based on the generalized vegetation coverage described above.

[0073] The present invention has the following beneficial effects compared with the prior art:

[0074] By setting the virtual minimum value v of the characteristic ratio index in the present invention min is less than or equal to the minimum value of the characteristic ratio index, ρ is greater than 0 and close to 0, v min+ = v min + ρ corresponding to the power function value based on the generalized vegetation coverage is equal to v min+ so as to determine the characteristic power and the corresponding characteristic power curve (power function) of the generalized vegetation coverage, realize the automatic calculation of the characteristic power, reduce the human dependence, improve the standardization degree of the product enhancement processing process and the consistency of the processing results, effectively improve the vegetation chroma and hierarchical characteristics on the original true color composite image, and greatly improve the overall effect of the true color image. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on the structures shown in these drawings without creative efforts.

[0076] Figure 1 is a simple flowchart of the product fusion method based on the generalized vegetation coverage in Embodiment 1 of the present invention;

[0077] Figure 2 is a detailed flowchart of the product fusion method based on the generalized vegetation coverage in Embodiment 1 of the present invention;

[0078] Figure 3 is a true color combined color image before vegetation feature enhancement in Embodiment 1 of the present invention;

[0079] Figure 4 is a characteristic ratio index image in Embodiment 1 of the present invention;

[0080] Figure 5 is an enhanced image of the vegetation coverage based on h(x) in Embodiment 1 of the present invention;

[0081] Figure 6 is an enhanced image of the vegetation coverage based on NDhI in Embodiment 1 of the present invention (without transformation);

[0082] Figure 7 is an enhanced image of the product of two vegetation coverages in Embodiment 1 of the present invention (without transformation);

[0083] Figure 8Enhanced image map (without transformation) based on the average value of two vegetation coverages in Embodiment 1 of the present invention;

[0084] Figure 9 Enhanced image map (exponential translation transformation) of vegetation coverage based on NDhI in Embodiment 1 of the present invention;

[0085] Figure 10 Enhanced image map (exponential translation transformation) based on the product of two vegetation coverages in Embodiment 1 of the present invention;

[0086] Figure 11 Enhanced image map (exponential translation transformation) based on the average value of two vegetation coverages in Embodiment 1 of the present invention;

[0087] Figure 12 Enhanced image map (exponential scaling transformation) of vegetation coverage based on NDhI in Embodiment 1 of the present invention;

[0088] Figure 13 Enhanced image map (exponential scaling transformation) based on the product of two vegetation coverages in Embodiment 1 of the present invention;

[0089] Figure 14 Enhanced image map (exponential scaling transformation) based on the average value of two vegetation coverages in Embodiment 1 of the present invention;

[0090] Figure 15 Structural block diagram of the product fusion system based on the generalized vegetation coverage in Embodiment 2 of the present invention;

[0091] Figure 16 Structural block diagram of the computer device in Embodiment 3 of the present invention. Detailed implementation manners

[0092] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention. It should be understood that the specific embodiments described are only used to explain the present application and are not used to limit the present application.

[0093] Embodiment 1:

[0094] As Figure 1 , 2 shown, this embodiment provides a product fusion method based on the generalized vegetation coverage, including the following steps:

[0095] S101. Input satellite remote sensing images.

[0096] Input satellite remote sensing images with near-infrared (NIR), red (R), green (G), and blue (B) bands.

[0097] S102. Calculate the characteristic ratio index based on the satellite remote sensing images.

[0098] Let b i be one of the bands in the true color bands, and the bands participating in the fusion are NIR and m true color bands. The number of bands in the fused result is m, and the fused result is denoted as b i '.

[0099] (1) For the pseudo-Brovey fusion, the fusion result is denoted as:

[0100]

[0101] where m = 1, 2, 3, i = 1,..., m, and b i , b j are the true color bands participating in the fusion.

[0102] (2) For the Schmidt-Gram, PCA fusion, or Wavelet fusion, the fusion result is denoted as:

[0103] b i ' = Fusion(NIR, b 1 ,..., b m )

[0104] where m = 1, 2, 3, i = 1,..., m, and b 1, ,..., b m are the true color bands participating in the fusion. Fusion is the GS (Gram-Schmidt), PCA fusion, or WL (Wavelet) fusion method.

[0105] The characteristic ratio index is:

[0106]

[0107] where n = 1,..., m.

[0108] S103. Determine the characteristic threshold and construct a linear transformation function based on the characteristic ratio index.

[0109] (1) Determine the characteristic threshold of the characteristic ratio index.

[0110] Calculate the minimum value, maximum value, and average value of the characteristic ratio index x, denoted as x min , x max , x m , then the characteristic threshold c of the characteristic ratio index belongs to [x min , xmax ; or the threshold value x of the pure water body is determined through human-computer interaction w , the threshold value x of the pure bare ground b , the threshold value x of the pure vegetation v , then the characteristic threshold value c ∈ [x w , x v ; by default, c = x is taken m .

[0111] (2) Construct a linear transformation function.

[0112] Let s and t be constants. A linear transformation of the characteristic ratio index can be obtained as follows:

[0113]

[0114] h(x) has similar mathematical properties and physical functions to x. Like x, it can also be used as a basic parameter for vegetation feature enhancement.

[0115] For example, if h(x) is multiplied by the original image, and it is required to enhance at positions greater than the threshold value c, remain unchanged at positions less than the threshold value c, and the image remains continuous, the conditions to be satisfied are: h(c) = 1, and when x > c, h(x) > 1; when x < c, h(x) < 1.

[0116] Two typical linear transformations that satisfy the above conditions are:

[0117] (2-1) During translation transformation, s = 1, t = 1 - c;

[0118] (2-2) During scaling transformation, s = c, t = 0.

[0119] In particular, when s = 1, t = 0, it is an identity transformation.

[0120] Then the normalized index corresponding to h(x) is:

[0121]

[0122] During translation transformation, st - s = -c, st + s = 2 - c; during scaling transformation, st - s = -c, st + s = c.

[0123] In particular, during identity transformation, st - s = -1, st + s = 1.

[0124] NDhI has similar mathematical properties and physical functions to h(x). Like h(x), it can also be used as a basic parameter for vegetation feature enhancement.

[0125] S104. Construct a power function space based on the generalized vegetation cover according to the linear transformation function.

[0126] (1) Let the characteristic ratio index \(x\) have a virtual minimum value \(v\). min \(\in[0,x min \), and based on the characteristic ratio index transformation function \(h(x)\) and the virtual minimum value \(v\), min construct the generalized vegetation coverage as:

[0127]

[0128] Its power function space is:

[0129]

[0130] \(H(v min ) = 0

[0131] \(H(x max ) = 1

[0132] where \(n>0\).

[0133] It can be seen that whether it is a scaling transformation or a translation transformation, the power function of the vegetation coverage has the same mathematical properties as when there is no transformation.

[0134] Similarly:

[0135] (2) The generalized vegetation coverage based on NDhI is:

[0136]

[0137] Its power function space is:

[0138]

[0139] \(N(v min ) = 0

[0140] \(N(x max ) = 1

[0141] where \(n>0\).

[0142] (3) The power function space based on the product of the two generalized vegetation coverages is:

[0143]

[0144] \(P(v min ) = H(v min )N(v min ) = 0

[0145] \(P(x max ) = H(x max )N(x max ) = 1

[0146] where \(n>0\).

[0147] (4) The power function space based on the average value of two generalized vegetation coverages is:

[0148]

[0149] The power function space is:

[0150]

[0151] where n > 0.

[0152] H(x), N(x), P(x), M(x) have excellent characteristics and are increasing functions in the interval x ∈ [v min , x max . The function value is 0 at the virtual minimum x = V min , and the function value is 1 at the maximum value of the characteristic ratio index x = x max . The function value at the characteristic threshold depends on an independent parameter - the power n. When used as a vegetation feature enhancement function, it can be used as an enhancement factor for product fusion.

[0153] S105. Determine the characteristic power of the power function space based on the generalized vegetation coverage and its corresponding product enhancement scheme.

[0154] Let the red, green, and blue bands of the true color image be R, G, and B respectively.

[0155] The general expression for product enhancement is:

[0156] R′ = Z(x)R

[0157] G′ = Z(x)G

[0158] B′ = B

[0159] where Z(x) is the power function space based on the generalized vegetation coverage.

[0160] (1) The product enhancement scheme for the generalized vegetation coverage based on h(x).

[0161] Let ρ > 0 be close to 0, the virtual minimum be v min = kx min , k ∈ [0, 1], v min+ = v min + ρ, when H(v min+ ) = v min+ , let:

[0162]

[0163] Then there is the characteristic power:

[0164]

[0165] Furthermore, the characteristic curve is as follows:

[0166]

[0167] Then there is a characteristic scheme of product enhancement:

[0168] R′ = H ρ (x)R

[0169] G′ = H ρ (x)G

[0170] B′ = B

[0171] (2) Product enhancement scheme for vegetation coverage based on NDhI.

[0172] Let ρ > 0 be close to 0, and the virtual minimum value be v min = kx min , k ∈ [0, 1], v min+ = v min + ρ, N(v min+ ) = v min+ When, let:

[0173]

[0174] Then there is a characteristic power:

[0175]

[0176] Furthermore, the characteristic curve is as follows:

[0177]

[0178] Then there is a characteristic scheme of product enhancement:

[0179] R′ = N ρ (x)R

[0180] G′ = N ρ (x)G

[0181] B′ = B

[0182] (3) Product enhancement scheme based on the product of two vegetation coverages.

[0183] Let ρ > 0 be close to 0, and the virtual minimum value be v min = kx min , k ∈ [0, 1], v min+ = v min + ρ, P(v min+ ) = v min+ When, let:

[0184]

[0185] There is a characteristic power:

[0186]

[0187] Furthermore, the characteristic curve is:

[0188]

[0189] There is a characteristic scheme with product enhancement:

[0190] R′ = P ρ (x)R

[0191] G′ = P ρ (x)G

[0192] B′ = B

[0193] (4) Product enhancement scheme based on the average of two vegetation coverages.

[0194] Let ρ > 0 be close to 0, and the virtual minimum value be v min = kx min , k ∈ [0, 1], v min+ = v min + ρ, M(v min+ ) = v min+ When, let:

[0195]

[0196] There is a characteristic power:

[0197]

[0198] Furthermore, the characteristic curve is:

[0199]

[0200] There is a characteristic scheme with product enhancement:

[0201] R′ = M ρ (x)R

[0202] G′ = M ρ (x)G

[0203] B′ = B

[0204] S106. Synthesize the true color image after vegetation enhancement and save it.

[0205] Use the R′, G′, and B bands to synthesize a color image corresponding to the red, green, and blue channels of the color image, and store the enhanced true color image.

[0206] The enhancement algorithm provided in this embodiment is based on the inherent characteristics of satellite image data, with strong data adaptability; the feature ratio index, feature threshold, generalized vegetation coverage, generalized vegetation coverage power function, and its characteristic curve have clear physical meanings. The parameters for feature enhancement are determined by calculation, reducing human dependence, having a clear processing target, reliable quality, and being easy to apply.

[0207] This embodiment will illustrate the above method with specific application examples:

[0208] To achieve the purpose of enhancing the vegetation characteristics of satellite remote sensing true color images, this embodiment mainly uses ENVI remote sensing image processing software to achieve it.

[0209] Step 1: Input the remote sensing image.

[0210] Open a multispectral remote sensing image with near-infrared band (NIR), red band (R), green band (G), and blue band (B). Figure 3 It is a true color composite color image before vegetation feature enhancement (the effect image stretched by 0.5% according to the default settings of envi).

[0211] Step 2: Calculate the feature ratio index and determine its threshold.

[0212] Take the feature ratio index obtained by quasi-GS fusion as an example.

[0213]

[0214] GSR, GSG, and GSB are the results of the fusion of NIR with R, G, and B respectively. The calculation results are as Figure 4 (the effect image stretched by 0.1% according to the default settings of envi).

[0215] Step 3: Determine the product fusion scheme.

[0216] Table 1 shows the statistical feature results of the feature ratio index, and Tables 2-4 are the calculation scheme result tables.

[0217] Table 1 Statistical feature table of the feature ratio index x

[0218]

[0219] Take the vegetation feature threshold c = 1.022385 of the feature ratio index x, and the linear transformation function of the feature ratio is:

[0220]

[0221] (1) When there is no transformation, s = 1, t = 0 (st = 0, st + s = 1)

[0222] (2) During the translation transformation, s = 1, t = 1 - c = -0.022385 (st = -0.022385, st + s = 0.977615)

[0223] (3) During the scaling transformation, s = c = 1.022385, t = 0 (st = 0, st + s = 1.022385)

[0224] Table 2 Results Table of Calculation Schemes (No Transformation)

[0225]

[0226] Table 3 Results Table of Calculation Schemes (Exponential Translation Transformation)

[0227]

[0228] Table 4 Results Table of Calculation Schemes (Exponential Scaling Transformation)

[0229]

[0230] Among them, b1 is the characteristic ratio index x, and b2 is the red band R or the green band G.

[0231] Step 4: Fusion Results and Their Effects.

[0232] It can be obtained from Tables 2 to 4 that there are 12 calculation schemes corresponding to 3 linear transformations, and the product enhancement effects of the vegetation coverages based on h(x) corresponding to the 3 linear transformations are the same. The results calculated using ENVI can be referred to Figures 5 to 14 .

[0233] From the visual perspective, the enhancement results obtained by the above schemes are similar, all effectively enhancing the vegetation information and better maintaining the characteristics of the non-vegetation information. Therefore, the statistical characteristics of one scheme are used to illustrate the effect of this method. For example, the enhancement results of the generalized vegetation coverage power function of the normalized index corresponding to the characteristic ratio index translation transformation are used to illustrate, as shown in Statistical Tables 5 to 8.

[0234] Table 5 Comparative Analysis Table of Statistical Characteristics of RGB Modes of Enhanced True Color Images and Original True Color Images

[0235]

[0236] Table 6 Comparative Analysis Table of Statistical Characteristics of Vegetation Regions of RGB Modes of Enhanced True Color Images and Original True Color Images

[0237]

[0238] Table 7 Comparative Analysis Table of Statistical Characteristics of Non-Vegetation Regions of RGB Modes of Enhanced True Color Images and Original True Color Images

[0239]

[0240] Table 8 Comparative Analysis Table of Classification Statistical Features of Enhanced True Color Image and Original True Color Image in HLS Mode

[0241]

[0242] According to Tables 5 - 8, the following conclusions can be drawn:

[0243] (1) The characteristics of green vegetation in the true color image are effectively enhanced, improving the visual separability of vegetation and the computer analysis ability.

[0244] By enhancing the true color composite image point by point using the method provided in this embodiment, the vegetation color, texture, and hierarchy are comprehensively improved, effectively improving the visual resolution of vegetation information in the true color image and the computer analysis ability, and enhancing the vegetation analysis ability and effect of the true color image mode.

[0245] (2) The overall characteristics of the true color image are greatly improved, expanding its application scope and potential.

[0246] While enhancing the vegetation characteristics in the true color image, the method provided in this embodiment basically maintains the characteristics of water bodies and exposed ground objects such as soil, rocks, and buildings in the true color image, greatly improving the overall visual characteristics and effect of the true color image. At the same time, the correlation between the bands of the true color image is reduced, and the color, texture, and hierarchy of the image are more abundant.

[0247] Those skilled in the art can understand that all or part of the steps in the method of the above embodiment can be completed by instructing relevant hardware through a program, and the corresponding program can be stored in a computer-readable storage medium.

[0248] It should be noted that although the method operations of the above embodiment are described in a specific order in the drawings, this does not require or imply that these operations must be performed in that specific order, or that all the shown operations must be performed to achieve the desired result. On the contrary, the described steps can be executed in a changed order. Additionally or alternatively, some steps can be omitted, multiple steps can be combined into one step for execution, and / or one step can be decomposed into multiple steps for execution.

[0249] Embodiment 2:

[0250] As Figure 15 shown, this embodiment provides a product fusion system based on generalized vegetation cover, which includes an acquisition module 1501, a calculation module 1502, a first construction module 1503, a second construction module 1504, a determination module 1505, and a fusion module 1506, where:

[0251] An acquisition module 1501, configured to acquire satellite remote sensing images with near-infrared, red, green, and blue bands;

[0252] A calculation module 1502, configured to perform spectral fusion between the near-infrared band and one or more of the red, green, and blue bands, use the fused band combination as the numerator, and use the corresponding unfused band combination as the denominator to calculate a characteristic ratio index;

[0253] A first construction module 1503, configured to construct a linear transformation function according to the characteristic ratio index;

[0254] A second construction module 1504, configured to set a virtual minimum value v of the characteristic ratio index min = kx min ; construct a power function based on the generalized vegetation cover according to the linear transformation function and the virtual minimum value; where k ∈ [0, 1] is a given value, and x min is the minimum value of the characteristic ratio index x;

[0255] A determination module 1505, configured to set v min+ = v min + ρ; determine the characteristic power of the power function based on the generalized vegetation cover from the fact that the power function value based on the generalized vegetation cover corresponding to v min+ is equal to v min+ ; where ρ is a given value greater than 0 and close to 0;

[0256] A fusion module 1506, configured to enhance the red and green bands respectively by using the power function based on the generalized vegetation cover with the determined characteristic power according to the principle of product enhancement; synthesize the enhanced red and green bands with the blue band to obtain a true color image with enhanced vegetation;

[0257] where the linear transformation function satisfies the conditions: h(c) = 1, and: if x > c, then h(x) > 1, if x < c, then h(x) < 1; c is the characteristic threshold of the characteristic ratio index x, and s and t are both constants.

[0258] For the specific implementation of each module in this embodiment, reference may be made to Embodiment 1 above, which will not be elaborated here one by one; it should be noted that the system provided in this embodiment only takes the above division of functional modules as an example. In practical applications, the above functions may be allocated to different functional modules according to needs, that is, the internal structure is divided into different functional modules to complete all or part of the functions described above.

[0259] Embodiment 3:

[0260] This embodiment provides a computer device, which may be a computer, such as Figure 16 As shown, it includes a processor 1602, a memory, an input device 1603, a display 1604, and a network interface 1605 connected through a system bus 1601. The processor is used to provide computing and control capabilities. The memory includes a non-volatile storage medium 1606 and an internal memory 1607. The non-volatile storage medium 1606 stores an operating system, a computer program, and a database. The internal memory 1607 provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. When the processor 1602 executes the computer program stored in the memory, it implements the product fusion method based on the generalized vegetation coverage in the above-mentioned Embodiment 1.

[0261] Embodiment 4:

[0262] This embodiment provides a computer-readable storage medium that stores a computer program. When the computer program is executed by a processor, it implements the product fusion method based on the generalized vegetation coverage in the above-mentioned Embodiment 1.

[0263] It should be noted that the computer-readable storage medium in this embodiment may be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0264] In summary, the present invention mainly aims at the inherent defects of the vegetation characteristics of satellite remote sensing images with near-infrared, red, green, and blue bands. Based on the internal relationship of remote sensing band data, an enhancement formula based on the power function of the generalized vegetation coverage characteristics is constructed, effectively improving the vegetation chromaticity and hierarchical characteristics on the original true-color composite image, and greatly improving the overall effect of the true-color image. This method has a clear physical meaning and a wide range of application objects. The enhanced image has distinct colors, rich information, and is easy for visual and automatic classification. Especially in the context of the rapid development of current high-resolution satellite remote sensing, it has a huge promoting effect on promoting the popularization and application of domestic high-resolution images in various industries at home and abroad.

[0265] In the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include at least one of such features. In the description of the present invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise specifically defined.

[0266] As described above, only the preferred embodiments of the present invention for patents are provided, but the protection scope of the present invention for patents is not limited thereto. Any person skilled in the art within the scope disclosed by the present invention for patents, according to the technical solution and inventive concept of the present invention for patents, makes equivalent substitutions or changes, and all belong to the protection scope of the present invention for patents.

Claims

1. A product fusion method based on generalized vegetation cover, characterized in that: The method comprises: Obtain satellite remote sensing images with near-infrared band, red band, green band and blue band; The near-infrared band is spectrally fused with one or more of the red band, green band and blue band, and the characteristic ratio index is calculated by taking the fused band combination as the numerator and the corresponding band combination before fusion as the denominator; According to the characteristic ratio index, a linear transformation function is constructed; Let the virtual minimum value v of the characteristic ratio index be min =kx min ; According to the linear transformation function and the virtual minimum, a power function based on generalized vegetation cover is constructed; where k∈[0,1] is a given value, x min is the minimum value of the characteristic ratio index x; Let v min+ =v min +ρ; by v min+ The corresponding power function value based on generalized vegetation cover and v min+ Equal, determine the characteristic power of the power function based on generalized vegetation cover; where ρ is a given value greater than 0 and close to 0; According to the principle of product enhancement, the red band and the green band are enhanced respectively by using the power function based on the generalized vegetation cover with a determined characteristic power; the enhanced red band and green band are synthesized with the blue band to obtain the true color image after vegetation enhancement; Among them, the linear transformation function satisfies the conditions: h(c) = 1, and: if x > c, then h(x) > 1; if x < c, then h(x) < 1; c is the characteristic threshold of the characteristic ratio index x, and both s and t are constants.

2. The product fusion method according to claim 1, characterized in that: s=1, t=1-c; or, s=c, t=0; or, s=1, t=0.

3. The product fusion method according to any one of claims 1 and 2, characterized in that: The method of constructing a power function based on generalized vegetation cover according to the linear transformation function and the virtual minimum value includes: According to the linear transformation function and the virtual minimum, the generalized vegetation cover based on h(x) is constructed as: According to the generalized vegetation cover based on h(x), the power function based on the generalized vegetation cover is obtained as follows: Among them, x max is the maximum value of the characteristic ratio exponent x; n is the characteristic power to be determined, n>0.

4. The product fusion method according to any one of claims 1 and 2, characterized in that: The method of constructing a power function based on generalized vegetation cover according to the linear transformation function and the virtual minimum value includes: The normalized index NdhI corresponding to the linear transformation function h(x) is: According to the normalized index NdhI and the virtual minimum value, the generalized vegetation cover based on NdhI is constructed as follows: According to the generalized vegetation cover based on NdhI, the power function based on the generalized vegetation cover is obtained as follows: Among them, x max is the maximum value of the characteristic ratio exponent x; n is the characteristic power to be determined, n>0.

5. The product fusion method according to any one of claims 1 and 2, characterized in that: The method of constructing a power function based on generalized vegetation cover according to the linear transformation function includes: According to the linear transformation function and the virtual minimum, the generalized vegetation cover based on h(x) is constructed as: The normalized index NdhI corresponding to the linear transformation function h(x) is: According to the normalized index NdhI and the virtual minimum value, the generalized vegetation cover based on NdhI is constructed as follows: According to the generalized vegetation cover based on h(x) and the generalized vegetation cover based on NdhI, the power function based on the product of the two generalized vegetation covers is: Among them, x max is the maximum value of the characteristic ratio index x; n is the characteristic power to be determined, n>0; P(x) is a power function constructed based on generalized vegetation cover.

6. The product fusion method according to any one of claims 1 and 2, characterized in that: The method of constructing a power function based on generalized vegetation cover according to the linear transformation function and the virtual minimum value includes: According to the linear transformation function and the virtual minimum, the generalized vegetation cover based on h(x) is constructed as: The normalized index NdhI corresponding to the linear transformation function h(x) is: According to the normalized index NdhI and the virtual minimum value, the generalized vegetation cover based on NdhI is constructed as follows: According to the generalized vegetation cover based on h(x) and the generalized vegetation cover based on NdhI, the power function based on the average value of the two generalized vegetation covers is obtained as follows: Among them, x max is the maximum value of the characteristic ratio index x; n is the characteristic power to be determined, n>0; M(x) is a power function constructed based on generalized vegetation cover.

7. The product fusion method according to any one of claims 1 and 2, characterized in that: According to the principle of product enhancement, the red band and the green band are enhanced respectively by using a power function based on generalized vegetation cover with a determined characteristic power, including: R ′ =Z ρ (x)R G ′ =Z ρ (x)G Among them, R ′ , G ′ are the enhanced red and green bands, Z ρ (x) is the power function based on generalized vegetation cover to determine the characteristic power, R and G are the red band and green band respectively.

8. The product fusion method according to any one of claims 1 and 2, characterized in that: The feature threshold c∈[x min ,x max ] or feature threshold c∈[x w ,x v ]; where x max is the maximum value of the characteristic ratio index x, x w 、x v They are the thresholds of pure water bodies and pure vegetation determined by human-computer interaction, respectively.

9. A product fusion system based on generalized vegetation cover, characterized in that: The system comprises: An acquisition module is used to acquire satellite remote sensing images with near infrared band, red band, green band and blue band; A calculation module is used to perform spectrum fusion between the near infrared band and one or more bands among the red band, the green band and the blue band, and calculate the characteristic ratio index by taking the band combination after fusion as the numerator and the band combination before fusion as the denominator; A first construction module is used to construct a linear transformation function according to a characteristic ratio index; The second building block is used to set the virtual minimum value v of the characteristic ratio index min =kx min ; According to the linear transformation function and the virtual minimum, a power function based on generalized vegetation cover is constructed; where k∈[0,1] is a given value, x min is the minimum value of the characteristic ratio index x; Determine the module used to set v min+ =v min +ρ; by v min+ The corresponding power function value based on generalized vegetation cover and v min+ Equal, determine the characteristic power of the power function based on generalized vegetation cover; where ρ is a given value greater than 0 and close to 0; The fusion module is used to enhance the red band and the green band respectively according to the principle of product enhancement by using a power function based on generalized vegetation cover with a determined characteristic power; the enhanced red band and green band are synthesized with the blue band to obtain a true color image after vegetation enhancement; Among them, the linear transformation function satisfies the conditions: h(c) = 1, and: if x > c, then h(x) > 1; if x < c, then h(x) < 1; c is the characteristic threshold of the characteristic ratio exponent x, and both s and t are constants.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the product fusion method described in any one of claims 1 to 8 is implemented.

Images