An automatic shadow detection method and apparatus for high-resolution remote sensing satellite images

Abstract

This invention relates to the fields of high-resolution multispectral imaging technology and shadow region restoration technology for remote sensing satellite images, and particularly to an automatic shadow detection method and apparatus for high-resolution remote sensing satellite images. The method includes: acquiring a multispectral image and a panchromatic image, and performing radiometric calibration on the multispectral and panchromatic images; performing NND image fusion on the radiometrically calibrated multispectral and panchromatic images to obtain a high-resolution multispectral satellite remote sensing image; converting the high-resolution multispectral satellite remote sensing image in RGB color space to a high-resolution multispectral satellite remote sensing image in HSV invariant color space, and performing compression processing; performing shadow detection on the compressed image to obtain a shadow detection result image. The shadow detection region obtained by this invention has high accuracy, and the conversion between different color spaces simplifies the algorithm calculation, avoids image limitations due to different scenes, has a wide range of applications, and is easy to promote and use.

Description

Technical Field

[0001] This invention relates to the fields of high-resolution multispectral imaging technology and remote sensing satellite image shadow region restoration technology, and particularly to an automatic shadow detection method and apparatus for high-resolution remote sensing satellite images. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] Visual observation of shadow regions in high-resolution multispectral remote sensing images is the simplest and most time-consuming method for shadow detection and identification. Building upon visual observation, and with the development of numerous shadow detection algorithms, various attribute-based and model-based methods have been applied to shadow detection in remote sensing images. While these methods can detect shadow regions, their accuracy needs improvement; they are prone to missing small shadow regions and falsely detecting easily confused non-shadow regions. Summary of the Invention

[0004] To address the technical problems mentioned above, this invention provides an automatic shadow detection method and apparatus for high-resolution remote sensing satellite images. It utilizes the characteristics of shadow areas in the HSV invariant color space—high hue, high saturation, and low brightness—and the spectral characteristics of near-infrared band reflectivity being higher than that of visible light band, thereby improving accuracy. This method can improve the problem that small shadow areas are easily overlooked, and that blue, green, and dark areas are easily misclassified as shadows.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] The first aspect of the present invention provides an automatic shadow detection method for high-resolution remote sensing satellite images.

[0007] An automatic shadow detection method for high-resolution remote sensing satellite images includes:

[0008] Acquire multispectral and panchromatic images, and perform radiometric calibration on the multispectral and panchromatic images;

[0009] The radiometrically calibrated multispectral image and the panchromatic image are fused using the Nearest Neighbor Diffusion Pan Sharpening (NND) algorithm to obtain a high-resolution multispectral satellite remote sensing image.

[0010] High-resolution multispectral satellite remote sensing images in RGB color space are converted to high-resolution multispectral satellite remote sensing images in HSV invariant color space and compressed. Shadow detection is then performed on the compressed images to obtain the shadow detection effect image.

[0011] A second aspect of the present invention provides an automatic shadow detection device for high-resolution remote sensing satellite images.

[0012] An automatic shadow detection device for high-resolution remote sensing satellite images includes:

[0013] The acquisition and radiometric calibration module is configured to acquire multispectral and panchromatic images and perform radiometric calibration on the multispectral and panchromatic images.

[0014] The image fusion module is configured to perform NND image fusion on radiometrically calibrated multispectral images and panchromatic images to obtain high-resolution multispectral satellite remote sensing images.

[0015] The spatial conversion and compression module is configured to convert high-resolution multispectral satellite remote sensing images in RGB color space into high-resolution multispectral satellite remote sensing images in HSV invariant color space, perform compression processing, perform shadow detection on the compressed image, and obtain a shadow detection effect image.

[0016] A third aspect of the present invention provides a computer-readable storage medium.

[0017] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the automatic shadow detection method for high-resolution remote sensing satellite imagery as described in the first aspect above.

[0018] A fourth aspect of the present invention provides a computer device.

[0019] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the steps of the automatic shadow detection method for high-resolution remote sensing satellite images as described in the first aspect above.

[0020] Compared with the prior art, the beneficial effects of the present invention are:

[0021] This invention utilizes the differences between shadowed and unshadowed regions in the visible and near-infrared bands, selecting the near-infrared band for shadow detection. Furthermore, it leverages the differences in hue and brightness, saturation and brightness between shadowed and unshadowed regions in the HSV color space to construct a logarithmic shadow index of normalized hue-brightness difference and saturation-brightness difference, enabling automatic shadow detection in high-resolution multispectral remote sensing satellite images and significantly improving the accuracy of shadow detection.

[0022] This invention employs the NND image fusion method, which effectively preserves color, texture, and spectral information in its fusion results.

[0023] This invention uses Gamma filtering to reduce speckle noise in remote sensing images while preserving edge information. Attached Figure Description

[0024] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0025] Figure 1 This is a flowchart of an automatic shadow detection method for high-resolution remote sensing satellite images, as shown in Embodiment 1 of the present invention;

[0026] Figure 2 The original image of the complex scene shown in Embodiment 1 of the present invention;

[0027] Figure 3 This is a diagram illustrating the effect of shadow detection in a complex scene as shown in Embodiment 1 of the present invention;

[0028] Figure 4 The image shown is the original image of a scene with water features, as illustrated in Embodiment 1 of the present invention.

[0029] Figure 5 This is a diagram showing the effect of water shadow detection in Embodiment 1 of the present invention;

[0030] Figure 6 The first embodiment of this invention only shows the original drawings of urban civilization buildings;

[0031] Figure 7 The image shown in Embodiment 1 of this invention illustrates the effect of shadow detection on urban buildings. Detailed Implementation

[0032] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0033] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0034] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0035] It should be noted that the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of methods and systems according to various embodiments of this disclosure. It should be noted that each block in a flowchart or block diagram may represent a module, segment, or portion of code, which may include one or more executable instructions for implementing the logical functions specified in the various embodiments. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the drawings. For example, two consecutively represented blocks may actually be executed substantially in parallel, or they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the flowcharts and / or block diagrams, and combinations of blocks in the flowcharts and / or block diagrams, may be implemented using a dedicated hardware-based system that performs the specified functions or operations, or using a combination of dedicated hardware and computer instructions.

[0036] Example 1

[0037] like Figure 1As shown, this embodiment provides an automatic shadow detection method for high-resolution remote sensing satellite images. This embodiment uses the application of this method to a server as an example for illustration. It is understood that this method can also be applied to terminals, and can also be applied to systems including terminals, servers, and other components, and can be implemented through interaction between the terminal and the server. The server can be an independent physical server, a server cluster composed of multiple physical servers, or a distributed system. It can also be a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network servers, cloud communication, middleware services, domain name services, CDN security services, and big data and artificial intelligence platforms. The terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, etc., but is not limited to these. The terminal and server can be directly or indirectly connected via wired or wireless communication, which is not limited herein. In this embodiment, the method includes the following steps:

[0038] First, radiometric calibration is performed on the multispectral and panchromatic images. Then, atmospheric correction is performed on the radiometrically calibrated multispectral images, which are then fused with the radiometrically calibrated panchromatic images using NND image fusion to obtain a high-resolution multispectral satellite remote sensing image. This fusion algorithm can well preserve color, texture, and spectral information, and is a state-of-the-art fusion algorithm.

[0039] Then, Gamma filtering is selected to eliminate noise in the remote sensing image. Gamma filtering reduces speckle noise in the remote sensing image while preserving edge information.

[0040] The high-resolution multispectral satellite remote sensing image in RGB color space is then converted to HSV invariant color space. The image is then compressed using a new shadow index based on logarithmic normalized mixed property (LNMPSI). The resulting image is then subjected to shadow detection processing, which significantly improves the accuracy of the shadow detection results.

[0041] Choose a suitable threshold to binarize the image. Here, we use the manual thresholding method to determine the optimal threshold result.

[0042] Finally, a morphological cleaning process is performed on the obtained binarized image. The morphological cleaning process includes dilation and erosion operations. A closed-loop morphological cleaning process is selected, which performs dilation first and then erosion.

[0043] The shadow detection area obtained in this embodiment has high accuracy. At the same time, the conversion of different color spaces simplifies the algorithm calculation, avoids the limitation of images by different scenes, has a wide range of applications, and is easy to promote and use.

[0044] Taking the Gaofen-1 satellite remote sensing image as an example, three different scene types of images were selected for the experiment. The reflectance of the incident light diffusion part can be used to represent the difference between the shadow area and the non-shadow area, as shown in Equation (1):

[0045] H d =n d ∫ λ S c (λ)r(λ)h d (λ)dλ (1)

[0046] Among them, H d ∈{R d G d B d The system provides red, green, and blue sensors to respond to the incident light diffusion section, n d It is only related to geometric information, S c (λ)∈{S R (λ),S G (λ),S B (λ)} represents the spectral sensitivity at wavelength λ, r(λ) represents the incident light magnitude, and h d (λ)∈{R d (λ),G d (λ),B d (λ)} represents the surface reflectivity of the red, green, and blue bands.

[0047] In remote sensing applications, sunlight is used as the sole incident light. The integral of the spectral sensitivity is shown in equation (2):

[0048] ∫ λ S c (λ)dλ=C (2)

[0049] Where C is a certain constant.

[0050] In electromagnetic wave theory, surface reflectivity is directly proportional to wavelength. Because λ NIR >λ R >λ G >λ B , where λ NIR , λ R , λ G , λ B These are the wavelengths of the near-infrared band, red band, green band, and blue band, respectively. Therefore, the unequal relationship of surface reflectance in the near-infrared, red, green, and blue bands is shown in equation (3):

[0051] NIR d (λ)>R d (λ)>G d (λ)>B d (λ) (3)

[0052] Among them, NIR d (λ), R d (λ), G d (λ), B d (λ) represents the surface reflectance in the near-infrared band, red band, green band, and blue band, respectively.

[0053] Relationship of tonal components between shadow and non-shadow areas:

[0054]

[0055] Among them, H nonshadow R nonshadow G nonshadow B nonshadow These represent the hue components, red band components, green band components, and blue band components of the non-shaded area, respectively. The hue values ​​of the non-shaded area are converted to the hue values ​​of the shaded area based on the differences between the shaded and non-shaded areas.

[0056]

[0057] Among them, H shadow R represents the hue component of the shaded area. d G d B d These represent the reflectance values ​​of the red, green, and blue bands of the incident light diffusion portion, respectively.

[0058] Therefore, based on equations (4) and (5), it can be deduced that the hue value of the shaded area is greater than the hue value of the unshaded area, as shown in equation (6):

[0059] H shadow >H nonshadow (6)

[0060] Furthermore, the intensity value of the shaded area can also be expressed by equation (7):

[0061]

[0062] Among them, V shadow V represents the intensity value of the shaded area. nonshadow This represents the intensity value of the non-shaded area.

[0063] According to equation (7), it can be deduced that the intensity value of the shaded area is less than that of the unshaded area, as shown in equation (8):

[0064] V shadow <V nonshadow (8)

[0065] Furthermore, the saturation values ​​of the non-shaded areas and the shaded areas are represented by equations (9) and (10), respectively:

[0066]

[0067] Among them, S nonshadow This represents the saturation value of the non-shaded area.

[0068]

[0069] Among them, S shadow This represents the saturation value of the shaded area.

[0070] Based on equations (9) and (10), it can be deduced that the saturation value of the shaded area is greater than that of the unshaded area, as shown in equation (11):

[0071] S shadow >S nonshadow (11)

[0072] The hue and intensity components, and the saturation and intensity components are normalized to the range of [0,1]. Based on equations (6), (8), and (11), the relationship between the hue-intensity difference in the shadow area and the hue-intensity difference in the non-shadow area can be derived. The relationship between the saturation-intensity difference in the shadow area and the saturation-intensity difference in the non-shadow area is shown in equations (12) and (13), respectively.

[0073] H shadow -V shadow >H nonshadow -V nonshadow (12)

[0074] S shadow -V shadow >S nonshadow -V nonshadow (13)

[0075] First, the H and V components, and the S and V components are normalized to the range [-1, 1]. Then, taking full advantage of the higher reflectivity of the shadowed regions in the NIR band, a natural logarithm function is added to compress the data scale to a narrower pixel-level scale, as shown in the following equation:

[0076]

[0077] "+1" is to avoid the case of ln0.

[0078] The method described in this embodiment first involves experimental testing, analysis, and comparison of remote sensing image samples from the Gaofen-1 satellite. Then, the experimentally derived method is applied to remote sensing images from various times and scenarios for testing and verification. These image types include complex terrain scenes, terrain scenes containing water bodies, and simple terrain scenes, such as... Figures 2-7 As shown, the accuracy and effectiveness of shadow detection results obtained from various types of remote sensing images were evaluated to assess its performance. The results demonstrate that the proposed method exhibits excellent shadow detection accuracy and improves upon the problems of small shadow areas being easily overlooked and blue, green, and dark areas being misclassified as shadows. It significantly contributes to subsequent shadow area compensation in remote sensing images and can quickly, conveniently, and effectively detect shadow areas in high-resolution multispectral remote sensing satellite images.

[0079] Example 2

[0080] This embodiment provides an automatic shadow detection device for high-resolution remote sensing satellite images.

[0081] An automatic shadow detection device for high-resolution remote sensing satellite images includes:

[0082] The acquisition and radiometric calibration module is configured to acquire multispectral and panchromatic images and perform radiometric calibration on the multispectral and panchromatic images.

[0083] The image fusion module is configured to perform NND image fusion on radiometrically calibrated multispectral images and panchromatic images to obtain high-resolution multispectral satellite remote sensing images.

[0084] The spatial conversion and compression module is configured to convert high-resolution multispectral satellite remote sensing images in RGB color space into high-resolution multispectral satellite remote sensing images in HSV invariant color space, perform compression processing, perform shadow detection on the compressed image, and obtain a shadow detection effect image.

[0085] It should be noted that the acquisition and radiometric calibration module, image fusion module, and spatial transformation and compression module described above are the same examples and application scenarios implemented in Embodiment 1, but are not limited to the content disclosed in Embodiment 1. It should also be noted that these modules, as part of the system, can be executed in a computer system such as a set of computer-executable instructions.

[0086] This embodiment is primarily based on the theoretical basis of the difference in attributes between shadowed and unshadowed areas, supported by the imaging technology of the Gaofen-1 satellite. The specific construction process utilizes the characteristics of shadowed areas in the HSV invariant color space (high hue, high saturation, low brightness) and the spectral characteristics of near-infrared band reflectivity being higher than that of visible light band. Ultimately, it yields an exponential function utilizing the near-infrared, hue, saturation, and brightness bands. The above description introduces the theoretical basis, basic principles, and significant applications of this shadow detection device. This invention is not limited to any particular field; it establishes a connection between shadowed and unshadowed areas based on their attributes. High-resolution remote sensing satellite images obtained using the Gaofen-1 imaging device can be used for shadow detection. The above implementation examples and specifications only introduce the background and basic principles of this device. Various improvements and corresponding changes can be made without altering the principle of this device; all such improvements and changes are within the scope of protection and can broaden the application range of this device.

[0087] The Gaofen-1 satellite, launched into orbit on April 26, 2013, is capable of imaging 8 orbits per day with a 35° side-swing capability. Designed for a lifespan of 5-8 years, it features multispectral imaging, wide swath, short revisit period, high temporal resolution, and high spatial resolution. The Gaofen-1 satellite carries two 2m resolution panchromatic / 8m resolution multispectral cameras and four 16m resolution multispectral cameras. Through multispectral data fusion, it significantly improves imaging quality and effects.

[0088] Example 3

[0089] This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in the automatic shadow detection method for high-resolution remote sensing satellite images as described in Embodiment 1 above.

[0090] Example 4

[0091] This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps in the automatic shadow detection method for high-resolution remote sensing satellite images as described in Embodiment 1 above.

[0092] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0093] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0094] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0095] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0096] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0097] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An automatic shadow detection method for high-resolution remote sensing satellite images, characterized in that, include: Acquire multispectral and panchromatic images, and perform radiometric calibration on the multispectral and panchromatic images; The radiometrically calibrated multispectral image and panchromatic image are fused using NND image fusion to obtain a high-resolution multispectral satellite remote sensing image. High-resolution multispectral satellite remote sensing images in RGB color space are converted into high-resolution multispectral satellite remote sensing images in HSV invariant color space and compressed. Shadow detection is then performed on the compressed images to obtain the shadow detection effect image. The compression process uses the Shadow Index Formula (LNMPSI) to compress the image. The Shadow Index Formula (LNMPSI) is: In the formula, H Indicates the hue value of the shaded area. V Indicates the intensity value of the shaded area. S This represents the saturation value of the shadow area. "+1" is to avoid the case where ln0 occurs.

2. The automatic shadow detection method for high-resolution remote sensing satellite images according to claim 1, characterized in that, The step of performing NND image fusion on the radiometrically calibrated multispectral image and the panchromatic image specifically includes first performing atmospheric correction on the radiometrically calibrated multispectral image, and then performing NND image fusion with the radiometrically calibrated panchromatic image.

3. The automatic shadow detection method for high-resolution remote sensing satellite images according to claim 1, characterized in that, Before performing spatial transformation, the process also includes: using Gamma filtering to eliminate noise in high-resolution multispectral satellite remote sensing images in the HSV invariant color space.

4. The automatic shadow detection method for high-resolution remote sensing satellite images according to claim 1, characterized in that, After obtaining the shadow effect image, the method further includes: determining the optimal threshold condition, performing binarization processing on the shadow effect image, and obtaining a binarized image.

5. The automatic shadow detection method for high-resolution remote sensing satellite images according to claim 4, characterized in that, After obtaining the binarized image, the process further includes: selecting a closed-loop morphological cleaning process that first performs dilation and then erosion on the binarized image.

6. The automatic shadow detection method for high-resolution remote sensing satellite images according to claim 4, characterized in that, The optimal threshold condition is determined using a manual thresholding method.

7. An automatic shadow detection device for high-resolution remote sensing satellite images, characterized in that, include: The acquisition and radiometric calibration module is configured to acquire multispectral and panchromatic images and perform radiometric calibration on the multispectral and panchromatic images. The image fusion module is configured to perform NND image fusion on radiometrically calibrated multispectral images and panchromatic images to obtain high-resolution multispectral satellite remote sensing images. The spatial conversion and compression module is configured to convert high-resolution multispectral satellite remote sensing images in RGB color space into high-resolution multispectral satellite remote sensing images in HSV invariant color space, compress the images, perform shadow detection on the compressed images, and obtain shadow detection effect images. The compression process uses the Shadow Index Formula (LNMPSI) to compress the image. The Shadow Index Formula (LNMPSI) is: In the formula, H Indicates the hue value of the shaded area. V Indicates the intensity value of the shaded area. S This represents the saturation value of the shadow area. "+1" is to avoid the case where ln0 occurs.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the automatic shadow detection method for high-resolution remote sensing satellite images as described in any one of claims 1-6.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the automatic shadow detection method for high-resolution remote sensing satellite images as described in any one of claims 1-6.

Images