Method, system and readable storage medium for adaptive enhancement of vegetation features in images

Abstract

The present invention discloses a method, system, and readable storage medium for adaptively enhancing vegetation features in images. The method comprises: inputting a multispectral remote sensing image; calculating a characteristic ratio index of the remote sensing image; determining a characteristic threshold and a characteristic transformation function for the characteristic ratio index; constructing a smooth constraint function space; constructing a symmetric curve of the characteristic ratio index transformation function based on a three-point constraint; constructing a vegetation feature enhancement coefficient function and a fusion expression for vegetation feature enhancement; and synthesizing a true color image after vegetation enhancement. The present invention constructs a three-point constrained characteristic curve as an incremental function for enhancing vegetation features in remote sensing images, automatically selecting enhancement parameters such as the enhancement power and the slope of the symmetric curve based on the image's own data features, thereby improving the automation, normalization, and standardization of the enhancement processing process for large-scale, massive remote sensing images and providing a technical foundation for standardized enhancement of massive remote sensing images.

Description

Technical Field

[0001] The present invention relates to the technical field of remote sensing image information enhancement, and more specifically, to a method and system for adaptively enhancing image vegetation features and a readable storage medium. Background Art

[0002] With the development of multi-platform, multi-sensor, multi-weather, multi-temporal, and multi-resolution remote sensing technologies, remote sensing imagery with varying spatial, temporal, and spectral resolutions is becoming increasingly abundant. Over the past two decades, spatiotemporal fusion of remote sensing imagery has emerged as a new direction in remote sensing image processing, with various fusion techniques rapidly developing and achieving a series of new results. However, research on interspectral fusion enhancement of multispectral imagery is relatively limited, primarily focusing on true color image simulation or vegetation feature enhancement.

[0003] With the widespread application of remote sensing across various industries, multispectral satellite true-color imagery, with its advantageous "what you see is what you get" (WYSIWYG) properties, has become one of the most widely used remote sensing image types. However, inherent flaws in raw true-color imagery, such as unnatural and unrealistic vegetation colors, have limited its effectiveness. Effectively enhancing vegetation features in true-color satellite imagery is both a key and challenging aspect of true-color satellite image processing. Researchers have conducted fruitful research on this topic, laying the theoretical and technical foundation for further solutions to related issues.

[0004] Based on primary satellite remote sensing products, Chen Chun et al. corrected remote sensing data for Rayleigh scattering, making the color signal images closer to true color images on the ground (Chen Chun et al., Extraction and Reproduction of Color Signals from Remote Sensing Sources, Surveying and Mapping Science, January 2006, Vol. 31, No. 1; Han Xiuzhen et al., Research and Application of Synthesis Methods for True Color Images of Fengyun-3D Satellite, Journal of Marine Meteorology, May 2019, Vol. 39, No. 2). You Jing et al. used white balance and color correction based on colorimetry to improve vegetation characteristics in true color images, obtaining 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, November 2016, Vol. 45, No. 11).

[0005] Based on the secondary processed products of remote sensing images, Fan Xuyan and others enhanced the green band of remote sensing images to obtain relatively good true color images. In the early days, the green band and the near-infrared band were mainly used to obtain the new green band by the overall weighted combination operation scheme (Fan Xuyan et al., Remote Sensing Image Simulation True Color Fusion Method 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 ALOS Natural Color Image Transformation and Fusion Method, Surveying and Mapping Technology and Equipment, Vol. 14, No. 1, 2012; Shi Yuanli et al., Analysis on the Applicability of GF-2 Satellite Remote Sensing Imagery for Mapping, Bulletin of Surveying and Mapping, No. 12, 2017); Later, it gradually developed into a method based on The normalized vegetation index is used as a classification function to perform weighted classification processing on the vegetation pixels in the image to obtain a new green band (Zhang Wei et al., True color synthesis method of multispectral images based on vegetation index, Surveying and Spatial Geographic Information, Vol. 33, No. 6, December 2010); the recent development is to use the normalized vegetation index segmentation to 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 color naturalness of multispectral remote sensing images using near-infrared, Master's thesis, 2016).

[0006] Although the above methods can achieve vegetation enhancement in true color images to a certain extent, the selection of feature enhancement parameters in the color image enhancement process mainly relies on manual selection, which leads to insufficient automation, normalization and standardization of color image enhancement. Summary of the Invention

[0007] In view of the above problems, the object of the present invention is to provide a method, system and readable storage medium for adaptively enhancing vegetation features in images.

[0008] A first aspect of the present invention provides a method for adaptively enhancing vegetation features in an image, the method comprising:

[0009] Input multispectral remote sensing images with near-infrared band, red band, green band, and blue band;

[0010] Calculate the characteristic ratio index of remote sensing images;

[0011] According to the characteristic ratio index of the remote sensing image, the characteristic threshold value and characteristic transformation function of the characteristic ratio index are determined;

[0012] According to the characteristic ratio index and its characteristic threshold, a smooth constraint function space is constructed;

[0013] According to the smooth constraint function, a symmetric curve of the characteristic ratio exponential transformation function based on three-point constraints is constructed;

[0014] According to the symmetric curve of the characteristic ratio exponential transformation function based on three-point constraints, the vegetation feature enhancement coefficient function and the fusion expression of vegetation feature enhancement are constructed;

[0015] The true color image after vegetation enhancement is synthesized according to the vegetation feature enhancement coefficient function and the fusion expression of vegetation feature enhancement.

[0016] Preferably, the calculating of the characteristic ratio index of the remote sensing image specifically includes:

[0017] Assume b i is a band of remote sensing image, and the remote sensing image bands involved in the fusion are near infrared band and m true color bands; the number of image bands after fusion is m, and the fusion result is recorded as b' i ;

[0018] The characteristic ratio index is:

[0019]

[0020] Where n = 1,…,m.

[0021] Preferably, determining the characteristic threshold of the characteristic ratio index includes:

[0022] Calculate the minimum value x of the characteristic ratio index x min , maximum value x max and the mean value x m ;

[0023] Determine the threshold x of pure water through human-computer interaction w , the threshold value x of pure exposed ground objects b , the threshold value of pure vegetation x v ; then the feature threshold c∈[x w , x v ]; c=x by default m .

[0024] Preferably, the feature transformation function expression is:

[0025]

[0026] Where n>0, sign is the sign operator, and abs is the absolute value operator;

[0027] s and t are transformation parameters; when s = 1, t = 0, h(x) is the characteristic exponential power transformation; when s = 1, t = 1-c, h(x) is the characteristic exponential translation power transformation; when s = c, t = 0, h(x) is the characteristic exponential scaling power transformation; if c = 1, the three transformations are equivalent, all of which are h(x) = x n .

[0028] Preferably, the smooth constraint function space is constructed according to the feature threshold, specifically:

[0029] Assume that the smooth constraint function u(x) of the characteristic ratio index x is at the maximum value x max The value of is 1, that is:

[0030]

[0031] The smooth constraint function v(x) of the normalized exponential corresponding to the characteristic ratio index is at the maximum value x max The value of is also 1, that is:

[0032]

[0033] g(x)=C s v(x)

[0034] l(x)=C s u(x)

[0035] Among them, g(x) is the point passing through (c,0), (x max ,C s ) is the incremental function of vegetation feature enhancement, l(x) is the function passing through the points (c,0), (x max ,C s ) line, C s >0 is a constant, which is the maximum value of the characteristic ratio index x max The value of the normalized exponential smoothing constraint function v(x) at .

[0036] Preferably, constructing a symmetric curve of a characteristic ratio exponential transformation function based on three-point constraints according to a smooth constraint function includes:

[0037] According to the smooth constraint function, the three points of the three-point constraint are determined as (x min ,0), (c,0) and (x max ,C s );

[0038] The normalized exponential smoothing constraint function and symmetry axis of the characteristic ratio exponential transformation function are:

[0039] g(h(x))=C s v(h(x))

[0040] l(h(x))=kA(h(x))

[0041] in,

[0042] A(h(x))=h(x)-h(c)

[0043]

[0044] Let f(h(x)) be the symmetric curve of the normalized exponential smoothing constraint function g(h(x)) about the straight line l(h(x)), and its expression is:

[0045]

[0046] in,

[0047]

[0048] Let f'(h(x))=0, we can get f(h(x)) has the minimum value at x0, and x0∈(x min ,c), at this time:

[0049]

[0050] Let f(h(x0)) / C s =ε, the power n of the transformation function can be automatically determined according to the image data characteristics, thereby automatically determining the slope k of the symmetric curve. At this time, affected by (x min ,0), (c,0) and (x max ,C s ) and the three-point constrained symmetric curve f(h(x)) is the desired vegetation feature enhancement incremental function.

[0051] Preferably, the construction of the vegetation feature enhancement coefficient function and the fusion expression of vegetation feature enhancement includes:

[0052] The increment function f(h(x)) is normalized and the vegetation feature enhancement coefficient function F(x) is constructed according to the superposition principle:

[0053] F(x)=m·f(h(x)) / C s +1

[0054] m is the greenness adjustment coefficient, which has the function of adjusting the greenness of the image. m>0 is a constant, and the default value is m=1;

[0055] Assuming that the red, green, and blue bands of the remote sensing image are R, G, and B respectively, the red and green bands are enhanced using the above enhancement coefficients, and the image fusion expression for vegetation feature enhancement is as follows:

[0056] R'=F(x)×R

[0057] G'=F(x)×G

[0058] Among them, R' is the red band after vegetation characteristics are enhanced, and G' is the green band after vegetation characteristics are enhanced.

[0059] Preferably, the synthesis of the vegetation enhanced true color image according to the vegetation feature enhancement coefficient function and the vegetation feature enhancement fusion expression is specifically as follows:

[0060] The red band R' after vegetation feature enhancement, the green band G' after vegetation feature enhancement and the blue band B are placed correspondingly in the red, green and blue channels to synthesize the color image, which is the true color image vegetation feature enhanced image.

[0061] Preferably, the method further comprises the following step: storing the enhanced true color image.

[0062] The second aspect of the present invention provides an image vegetation feature adaptive enhancement system, including a memory and a processor, wherein the memory includes an image vegetation feature adaptive enhancement method program, and when the image vegetation feature adaptive enhancement method program is executed by the processor, the steps of an image vegetation feature adaptive enhancement method are implemented.

[0063] The third aspect of the present invention provides a computer-readable storage medium, which includes an image vegetation feature adaptive enhancement method program. When the image vegetation feature adaptive enhancement method program is executed by a processor, the steps of an image vegetation feature adaptive enhancement method are implemented.

[0064] Compared with the prior art, the beneficial effects of the technical solution of the present invention are: the present invention provides a method, system and readable storage medium for adaptive enhancement of vegetation features in images. min ,0), (c,0) and (x max ,C s ) The characteristic curve with three-point constraints is used as the incremental function for vegetation feature enhancement of remote sensing true color images. Based on the image's own data characteristics, the automatic selection of enhancement parameters such as the enhancement power n and the symmetry curve slope k is realized. This reduces the human dependence on the selection of feature enhancement parameters in the previous true color image enhancement, improves the automation, normalization and standardization of the large-scale massive remote sensing image enhancement processing process, and provides a technical foundation for the standardized enhancement of massive remote sensing images. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 This is a flow chart of a method for adaptively enhancing vegetation features in an image as described in Example 1.

[0066] Figure 2 This is a GF6 true color composite map of a certain area on January 18, 2021.

[0067] Figure 3 This is the GF6 characteristic ratio index chart of a certain region on January 18, 2021.

[0068] Figure 4This is a true color composite image of a certain area after GF6 enhancement on January 18, 2021. DETAILED DESCRIPTION

[0069] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other.

[0070] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0071] Example 1

[0072] like Figure 1 As shown, this embodiment discloses a method for adaptively enhancing vegetation features in an image, the method comprising:

[0073] S1: Input remote sensing images with near-infrared band, red band, green band and blue band;

[0074] S2: Calculate the characteristic ratio index of remote sensing images;

[0075] More specifically, the calculation of the characteristic ratio index of the remote sensing image includes:

[0076] Let x be the characteristic ratio index of vegetation feature enhancement. The index value has the characteristic of increasing from water to exposed objects such as rock and soil to vegetation objects, and can perform binary classification on the objects.

[0077] Assume b i is a band of remote sensing image, and the remote sensing image bands involved in the fusion are near infrared band and m true color bands; the number of image bands after fusion is m, and the fusion result is recorded as b' i ;

[0078] The characteristic ratio index is:

[0079]

[0080] Where n = 1, ..., m, m is the number of true color bands involved in the fusion, m = 1, 2, 3. b' i It is the result of spectral fusion between the near-infrared band (NIR) and one or more true color bands. Spectral fusion can use various image band fusion methods such as IHS, Brovey, Gram-Schmidt, PCA, CN, and Wavelet.

[0081] S3: Determine the characteristic threshold and characteristic transformation function of the characteristic ratio index according to the characteristic ratio index of the remote sensing image;

[0082] More specifically, determining the characteristic threshold of the characteristic ratio index includes:

[0083] Calculate the minimum value x of the characteristic ratio index x min , maximum value x max and the mean value x m ;

[0084] Determine the threshold x of pure water through human-computer interaction w , the threshold value x of pure exposed ground objects b , the threshold value of pure vegetation x v ; then the feature threshold c∈[x w , x v ]; c=x by default m , where c is the feature threshold for true color satellite image feature enhancement

[0085] More specifically, the feature transformation function is a function obtained by performing power transformation, translation power transformation, and scaling power transformation on the feature ratio exponent x at the feature threshold c, and its expression is:

[0086]

[0087] Where n>0, sign is the sign operator, and abs is the absolute value operator;

[0088] s and t are transformation parameters; when s = 1, t = 0, h(x) is the characteristic exponential power transformation; when s = 1, t = 1-c, h(x) is the characteristic exponential translation power transformation; when s = c, t = 0, h(x) is the characteristic exponential scaling power transformation; if c = 1, the three transformations are equivalent, all of which are h(x) = x n .

[0089] It should be noted that the characteristic transformation function h(x) shares the same monotonic characteristics as the characteristic ratio index x, namely, increasing along the axis from exposed features such as water to rock and soil to vegetation. It also has an equivalent classification function, namely, the features on either side of the threshold c are divided into two categories with the same characteristics. Using h(x) as the basic parameter for vegetation feature enhancement effectively expands the vegetation enhancement capabilities of the characteristic ratio index.

[0090] S4: Construct a smooth constraint function space based on the characteristic ratio index and its characteristic threshold;

[0091] More specifically, a smooth constraint function space is constructed, specifically:

[0092] Assume that the smooth constraint function u(x) of the characteristic ratio index x is at the maximum value xmax The value of is 1, that is:

[0093]

[0094] The smooth constraint function v(x) of the normalized exponential corresponding to the characteristic ratio index is at the maximum value x max The value of is also 1, that is:

[0095]

[0096] g(x)=C s v(x)

[0097] l(x)=C s u(x)

[0098] Among them, g(x) is the point passing through (c,0), (x max ,C s ) is the incremental function of vegetation feature enhancement, l(x) is the function passing through the points (c,0), (x max ,C s ) line, C s >0 is a constant, which is the maximum value of the characteristic ratio index x max The value of the normalized exponential smoothing constraint function v(x) at .

[0099] S5: Construct a symmetric curve of the characteristic ratio exponential transformation function based on three-point constraints according to the smooth constraint function;

[0100] More specifically, constructing a symmetric curve of a characteristic ratio exponential transformation function based on three-point constraints according to a smooth constraint function includes:

[0101] According to the smooth constraint function, the three points of the three-point constraint are determined as (x min ,0), (c,0) and (x max ,C s );

[0102] The normalized exponential smoothing constraint function and symmetry axis of the characteristic ratio exponential transformation function are:

[0103] g(h(x))=C s v(h(x))

[0104] l(h(x))=kA(h(x))

[0105] in,

[0106] A(h(x))=h(x)-h(c)

[0107]

[0108] Let f(h(x)) be the symmetric curve of the normalized exponential smoothing constraint function g(h(x)) about the straight line l(h(x)), and its expression is:

[0109]

[0110] in,

[0111]

[0112] Since the smooth constraint function g(h(x)) and the straight line l(h(x)) both pass through the points (c,0) and (x max ,C s ), then f(h(x)) must pass through (c,0) and (x max ,C s )Two points.

[0113] Let f(h(x min ))=0, then the symmetric curve f(h(x)) passes through (x min ,0) point, then:

[0114] A(h(x min ))=a3

[0115]

[0116] make:

[0117] T(k n )=a3-[h(x min )-h(c)]

[0118] T'(k n )=a3'

[0119] Then the slope k of the symmetric curve f(h(x)) is a function related to the power n, and its expression is:

[0120]

[0121] Let f'(h(x))=0, we can get f(h(x)) has the minimum value at x0, and x0∈(x min ,c), at this time:

[0122] (a1 2 +1)A 2 (h(x))+(2a1a2-a3)A(h(x))-a1a2a3+a3 2 / 4=0

[0123] a=a1 2 +1

[0124] b=2a1a2-a3

[0125] c=a3 2 / 4-a1a2a3

[0126]

[0127] A(h(x))=h(x)-h(c)

[0128] h(x)=A(h(x))+h(c)

[0129]

[0130] Let f(h(x0)) / C s =ε, the power n of the transformation function can be automatically determined according to the image's own data characteristics, thereby scientifically determining the slope k of the symmetric curve, and automatically selecting the incremental function parameters according to the remote sensing image's own data characteristics, thereby improving the deficiency of image enhancement parameters relying on manual selection and improving the degree of automation, normalization, and standardization of image enhancement. At this time, affected by (x min ,0), (c,0) and (x max ,C s ) and the three-point constrained symmetric curve f(h(x)) is the desired vegetation feature enhancement incremental function.

[0131] S6: Based on the symmetric curve of the characteristic ratio exponential transformation function based on the three-point constraint, the vegetation feature enhancement coefficient function and the fusion expression of vegetation feature enhancement are constructed;

[0132] More specifically, the construction of the vegetation feature enhancement coefficient function and the fusion expression of vegetation feature enhancement specifically includes:

[0133] The increment function f(h(x)) is normalized and the vegetation feature enhancement coefficient function F(x) is constructed according to the superposition principle:

[0134] F(x)=m·f(h(x)) / C s +1

[0135] m is the greenness adjustment coefficient, which has the function of adjusting the greenness of the image. m>0 is a constant, and the default value is m=1;

[0136] Assuming that the red, green, and blue bands of the remote sensing image are R, G, and B respectively, the red and green bands are enhanced using the above enhancement coefficients, and the image fusion expression for vegetation feature enhancement is as follows:

[0137] R'=F(x)×R

[0138] G'=F(x)×G

[0139] Among them, R' is the red band after vegetation characteristics are enhanced, and G' is the green band after vegetation characteristics are enhanced.

[0140] S7: Synthesize the vegetation enhanced true color image based on the vegetation feature enhancement coefficient function and the vegetation feature enhancement fusion expression.

[0141] More specifically, the true color image after vegetation enhancement is synthesized according to the vegetation feature enhancement coefficient function and the fusion expression of vegetation feature enhancement, specifically:

[0142] The red band R' after vegetation feature enhancement, the green band G' after vegetation feature enhancement and the blue band B are placed correspondingly in the red, green and blue channels to synthesize the color image, which is the true color image vegetation feature enhanced image.

[0143] S8: Store the enhanced true color image.

[0144] This embodiment provides a method, system, and readable storage medium for adaptively enhancing vegetation features in images. This embodiment constructs a continuous incremental coefficient function that is zero at the minimum and threshold values ​​of the feature ratio index, has a very small negative value between the minimum and threshold values, and is sufficiently large above the threshold value. This ensures that the enhanced image is smooth at the threshold value and maintains the stability of non-vegetation features while enhancing vegetation features in remote sensing images.

[0145] This embodiment automatically determines the parameters of the increment coefficient function based on the data characteristics of the remote sensing image itself, thereby improving the degree of automation, normalization, and standardization of the processing of massive remote sensing images.

[0146] This embodiment uses remote sensing image data to construct a characteristic curve constrained by three points as an incremental function. This allows for the standardized and automatic selection of parameters such as the incremental function power n and the slope k of the symmetric curve. This allows for standardized processing of large amounts of remote sensing imagery over a wide range, improving the consistency of multi-period and multi-view remote sensing image enhancement. Furthermore, this embodiment significantly enhances the amount of vegetation information in remote sensing images, comprehensively improves vegetation features in remote sensing images, and effectively improves the visual resolution and computer analysis capabilities of true color images.

[0147] The method described in this embodiment significantly improves the overall performance of true color images by directionally enhancing the vegetation feature information of true color images and maintaining the stability of the image features of exposed objects such as water bodies, soil, rocks, and buildings in the images. It also reduces the correlation between images in each band, reduces information redundancy, optimizes the data structure, and enhances the image information entropy and gradient. This makes the vegetation information in true color images richer, the colors, textures, layers, and their characteristics of different objects more diverse and significant, and the overall performance is superior.

[0148] As a specific embodiment, the following is described with reference to a specific example:

[0149] To achieve the purpose of enhancing vegetation features in true color satellite remote sensing images, this embodiment mainly uses ENVI remote sensing image processing software to achieve this, and is further described by a remote sensing image with near infrared band (NIR), red band (R), green band (G), and blue band (B).

[0150] 1. Input remote sensing image.

[0151] Open a multispectral remote sensing image with near infrared band (NIR), red band (R), green band (G), and blue band (B). Figure 2 This is the true color composite color image before vegetation feature enhancement (the effect image is stretched by 0.1% according to the default setting of envi).

[0152] 2. Calculate the characteristic ratio index

[0153] Take the feature ratio index obtained by GS fusion as an example.

[0154]

[0155] GSR, GSG, and GSB are the results obtained by fusing NIR with R, G, and B according to the GS method.

[0156] According to the above formula, the band operation expression (1.0*b1+b2+b3) / (1.0*b4+b5+b6) is established to calculate the characteristic ratio index x. Among them, b1 is the red band obtained by GS method fusion, b2 is the green band obtained by GS method fusion, b3 is the blue band obtained by GS method fusion, b4 is the original red band, b5 is the original green band, and b6 is the original blue band. The calculation results are as follows: Figure 3 . (The effect of 0.1% stretching according to the default setting of envi).

[0157] 3. Determine the feature threshold and translation transformation feature power and characteristic curve

[0158] Use ENVI statistical tools to find the mean value x m , which is used as the feature threshold c, that is, c = x m =1.066517. The maximum value of x max =2.921482, the minimum value of x min =0.258185.

[0159] In this embodiment, the parameters of the power transformation are s=1 and t=0. Let ε=-0.01, then f(h(x)) / C s =ε=-0.01, automatically calculate and solve n=1.390986, then the power transformation function h(x)=x 1.390986 , we calculated k = 0.620061, Cs = 2.076579, the symmetric curve f(h(x)) has a minimum value at x = 0.724515, then the symmetric curve function f(h(x)) of the characteristic ratio exponent x transformation space based on the three-point constraint is:

[0160] f(h(x))=(1*sqrt(((-0.759198)*(x^1.390986-1.093712)+(4.964122))^2+(x^1.390986-1.093712 )^2-(-0.941655)*(x^1.390986-1.093712))-((-0.759198)*(x^1.390986-1.093712)+(4.964122)))

[0161] 4. Red and green band vegetation feature enhancement calculation and true color synthesis.

[0162] Taking the greenness adjustment coefficient m = 1, the image vegetation feature enhancement coefficient is:

[0163] F(h(x))=f(h(x)) / C s +1

[0164] R'=F(x)×R

[0165] G'=F(x)×G

[0166] According to the above formula, enter the calculation expression: uint((((1*sqrt(((-0.759198)*(b1^1.390986-1.093712)+(4.964122))^2+(b1^1.390986-1.093712)^2-(-0.941655)*(b1^1.390986-1.093712))-((-0.759198)*(b1^1.390986-1.093712)+(4.964122))) / 2.07657941)+1)*b2), where b1 is the characteristic ratio index x, and b2 is the red band or green band of the true color image. The enhanced red band, green band and original blue band are combined into a color image synthesized according to the red, green and blue channels to form an enhanced true color image. Figure 4 (The result is a 0.1% stretching result according to the default setting of envi.) The reconstructed image file is saved as the result.

[0167] Tables 1 and 2 show a comparative analysis of the statistical characteristics of the vegetation enhancement color image in this case and the original true color image in RGB color mode, including the full image and the enhanced region. Table 3 shows a comparative analysis of the statistical characteristics of the full image and the enhanced region in HLS color mode.

[0168] Table 1 Comparative analysis of RGB mode statistical features of true color images after vegetation feature enhancement and original true color images

[0169]

[0170]

[0171] Table 2 Comparative analysis of vegetation regional statistical characteristics between the true color image after vegetation feature enhancement and the original true color image in RGB mode

[0172]

[0173] Table 3 Comparative analysis of statistical characteristics of HLS mode between true color images after vegetation feature enhancement and original true color images

[0174]

[0175]

[0176] The adaptive enhancement technology for vegetation features in satellite remote sensing images described in this embodiment is mainly aimed at the inherent defects of vegetation features in satellite remote sensing images with near-infrared, red, green, and blue bands. According to the intrinsic relationship of remote sensing band data, an enhancement formula based on a three-point constrained symmetric curve is constructed, and the parameters of the image enhancement incremental function are automatically determined according to the data characteristics of the image itself, thereby realizing the automation, normalization, and standardization of image enhancement. This technical method has clear physical meaning and a wide range of applications. The parameter selection depends on the data characteristics of the image itself, providing a technical basis for the standardized enhancement of massive images. This technical method effectively improves the chromaticity and hierarchical features of vegetation on the original true color composite image, greatly improving the overall effect of the true color image. The enhanced image has bright colors, rich information, and is easy to visually and automatically classify. Especially in the current context of rapid development of high-resolution satellite remote sensing, it has a huge promoting effect on promoting the promotion and application of domestic high-resolution images in various industries at home and abroad.

[0177] Example 2

[0178] This embodiment discloses a system for adaptively enhancing image vegetation features, including a memory and a processor. The memory includes a program for adaptively enhancing image vegetation features. When the program for adaptively enhancing image vegetation features is executed by the processor, the steps of the method for adaptively enhancing image vegetation features as described in Example 1 are implemented.

[0179] Example 3

[0180] This embodiment discloses a computer-readable storage medium, which includes a method program for adaptively enhancing image vegetation features. When the method program for adaptively enhancing image vegetation features is executed by a processor, the steps of the method for adaptively enhancing image vegetation features as described in Example 1 are implemented.

[0181] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0182] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed across multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0183] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0184] Those skilled in the art will understand that all or part of the steps of the above-mentioned method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above-mentioned method embodiments; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROMs), random access memories (RAMs), magnetic disks or optical disks, and other media that can store program codes.

[0185] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

Claims

1. A method for adaptively enhancing vegetation features in an image, characterized in that: The method comprises: Input multispectral remote sensing images with near-infrared band, red band, green band, and blue band; Calculate the characteristic ratio index of remote sensing images; According to the characteristic ratio index of the remote sensing image, the characteristic threshold value and characteristic transformation function of the characteristic ratio index are determined; According to the characteristic ratio index and its characteristic threshold, a smooth constraint function space is constructed; According to the smooth constraint function, a symmetric curve of the characteristic ratio exponential transformation function based on three-point constraints is constructed; According to the symmetric curve of the characteristic ratio exponential transformation function based on three-point constraints, the vegetation feature enhancement coefficient function and the fusion expression of vegetation feature enhancement are constructed; Synthesize the vegetation enhanced true color image according to the vegetation feature enhancement coefficient function and the vegetation feature enhancement fusion expression; The step of constructing a symmetric curve of a characteristic ratio exponential transformation function based on three-point constraints according to a smooth constraint function specifically includes: According to the smooth constraint function, the three points of the three-point constraint are determined as (x min ,0), (c,0) and (x max ,C s ); The normalized exponential smoothing constraint function and symmetry axis of the characteristic ratio exponential transformation function are: g(h(x))=Csv(h(x)) l(h(x))=kA(h(x)) in, A(h(x))=h(x)-h(c) Let f(h(x)) be the symmetric curve of the normalized exponential smoothing constraint function g(h(x)) about the straight line l(h(x)), and its expression is: in, Let f'(h(x))=0, then f(h(x)) has the minimum value at x0, and x0∈(x min ,c), at this time: Let f(h(x0)) / C s =ε, automatically determine the power n of the transformation function according to the image data characteristics, and thus automatically determine the slope k of the symmetric curve. At this time, affected by (x min ,0), (c,0) and (x max ,C s ) and the three-point constrained symmetric curve f(h(x)) is the vegetation feature enhancement incremental function; The construction of the vegetation feature enhancement coefficient function and the fusion expression of vegetation feature enhancement specifically includes: The increment function f(h(x)) is normalized and the vegetation feature enhancement coefficient function F(x) is constructed according to the superposition principle: F(x)=m·f(h(x)) / C s +1 m is the greenness adjustment coefficient, which has the function of adjusting the greenness of the image. m>0 is a constant, and the default value is m=1; Assuming that the red, green, and blue bands of the remote sensing image are R, G, and B respectively, the red and green bands are enhanced using the above enhancement coefficients, and the image fusion expression for vegetation feature enhancement is as follows: R'=F(x)×R G'=F(x)×G Among them, R' is the red band after vegetation characteristics are enhanced, and G' is the green band after vegetation characteristics are enhanced.

2. The method for adaptively enhancing vegetation features in an image according to claim 1, characterized in that: The calculating of the characteristic ratio index of the remote sensing image specifically includes: Assume b i is a band of remote sensing image, and the remote sensing image bands involved in the fusion are near infrared band and m true color bands; the number of image bands after fusion is m, and the fusion result is recorded as b' i ; The characteristic ratio index is: Where n = 1,…,m.

3. The method for adaptively enhancing vegetation features in an image according to claim 1 or 2, characterized in that: The characteristic threshold value of the characteristic ratio index is determined by: Calculate the minimum value x of the characteristic ratio index x min , maximum value x max and the mean value x m ; Determine the threshold x of pure water through human-computer interaction w , the threshold value x of pure exposed ground objects b , the threshold value of pure vegetation x v ; then the feature threshold c∈[x w , x v ]; c=x by default m .

4. The method for adaptively enhancing vegetation features in an image according to claim 3, characterized in that: The feature transformation function expression is: Where n>0, sign is the sign operator, and abs is the absolute value operator; s and t are transformation parameters; when s = 1, t = 0, h(x) is the characteristic exponential power transformation; when s = 1, t = 1-c, h(x) is the characteristic exponential translation power transformation; when s = c, t = 0, h(x) is the characteristic exponential scaling power transformation; if c = 1, the three transformations are equivalent, all of which are h(x) = x n .

5. The method for adaptively enhancing vegetation features in an image according to claim 4, characterized in that: The smooth constraint function space is constructed according to the characteristic ratio index and its characteristic threshold, specifically: Assume that the smooth constraint function u(x) of the characteristic ratio index x is at the maximum value x max If the value of is 1, then: The smooth constraint function v(x) of the normalized exponential corresponding to the characteristic ratio index is at the maximum value x max The value of is also 1, then: g(x)=C s v(x) l(x)=C s u(x) Among them, g(x) is the point passing through (c,0), (x max ,C s ) is the incremental function of vegetation feature enhancement, l(x) is the function passing through the points (c,0), (x max ,C s ) line, C s >0 is a constant, which is the maximum value of the characteristic ratio index x max The value of the normalized exponential smoothing constraint function v(x) at .

6. The method for adaptively enhancing vegetation features in an image according to claim 1, characterized in that: The true color image after vegetation enhancement is synthesized according to the vegetation feature enhancement coefficient function and the fusion expression of vegetation feature enhancement, specifically: The red band R after vegetation features are enhanced ' , Green band G after vegetation features are enhanced ' The color image synthesized by placing the red, green and blue channels corresponding to the blue band B is a true color image with enhanced vegetation features.

7. An image vegetation feature adaptive enhancement system, characterized in that: The method comprises a memory and a processor, wherein the memory comprises a method program for adaptively enhancing image vegetation features, and when the method program for adaptively enhancing image vegetation features is executed by the processor, the steps of a method for adaptively enhancing image vegetation features as claimed in any one of claims 1 to 6 are implemented.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a method program for adaptively enhancing image vegetation features. When the method program for adaptively enhancing image vegetation features is executed by a processor, the steps of the method for adaptively enhancing image vegetation features according to any one of claims 1 to 6 are implemented.

Images