Glacier identification method and device, storage medium, and computer equipment

Abstract

The present application discloses a glacier identification method and device, storage medium, and computer equipment. The method includes: obtaining remote sensing images to be processed in different periods of a glacier zone to be identified within a preset time period; performing cloud removal processing on each remote sensing image to be processed to obtain a target remote sensing image, and calculating a normalized snow index of each pixel in each target remote sensing image; generating a snow frequency map of the glacier zone within a preset time period according to the normalized snow index of each pixel in each target remote sensing image and a preset snow frequency threshold; determining a glacier type distinction threshold corresponding to the snow frequency map of the glacier zone based on the maximum inter-class variance method, and determining the glacier type of each glacier to be identified in the glacier zone to be identified based on the pixel value of each pixel in the snow frequency map of the glacier zone and the glacier type distinction threshold, wherein the glacier type includes seasonal glaciers and permanent glaciers. By automatically determining the glacier type distinction threshold, the efficiency of glacier identification is improved.

Description

Technical Field

[0001] The present application relates to the field of glacier remote sensing technology, and in particular to a glacier identification method and device, a storage medium, and a computer device. Background Art

[0002] Glaciers are important freshwater resources and climate change indicators on Earth, and have a profound impact on ecosystems, climate balance, and human society. Existing glacier identification methods include NDSI (Normalized Difference Snow Index), deep learning, and other methods. Deep learning glacier identification faces challenges such as large data requirements, complex training, insufficient data quality and diversity, poor model interpretability, and poor regional adaptability and environmental change adaptability. The NDSI method requires manual determination of thresholds through histograms, which is time-consuming and labor-intensive. Summary of the invention

[0003] In view of this, the present application provides a glacier identification method and device, storage medium, and computer equipment, which obtain remote sensing images to be processed in different periods of a glacier zone to be identified within a preset time period, wherein the glacier zone contains multiple glaciers to be identified; perform cloud removal on each remote sensing image to be processed to obtain a target remote sensing image, and calculate the normalized snow index of each pixel in each target remote sensing image; generate a snow frequency map of the glacier zone within a preset time period based on the normalized snow index of each pixel in each target remote sensing image and a preset snow frequency threshold; determine the glacier type distinction threshold corresponding to the snow frequency map of the glacier zone based on the maximum inter-class variance method, and determine the glacier type of each glacier to be identified in the glacier zone to be identified based on the pixel value of each pixel in the snow frequency map of the glacier zone and the glacier type distinction threshold, wherein the glacier type includes seasonal glaciers and permanent glaciers. By automatically determining the glacier type distinction threshold, it is possible to identify whether the glacier to be identified is a seasonal glacier or a permanent glacier, thereby improving the efficiency of glacier identification.

[0004] According to one aspect of the present application, a glacier identification method is provided, the method comprising:

[0005] Acquire remote sensing images to be processed at different periods of a glacier zone to be identified within a preset time period, wherein the glacier zone to be identified includes a plurality of glaciers to be identified;

[0006] Performing cloud removal processing on each of the remote sensing images to be processed to obtain a target remote sensing image, and calculating a normalized snow cover index of each pixel point in each of the target remote sensing images;

[0007] Generate a snow frequency map of the glacier zone within the preset time period according to the normalized snow index of each pixel point in each target remote sensing image and a preset snow frequency threshold;

[0008] Based on the maximum inter-class variance method, the glacier type distinction threshold corresponding to the snow frequency map of the glacier zone is determined, and based on the pixel value of each pixel point in the snow frequency map of the glacier zone and the glacier type distinction threshold, the glacier type of each glacier to be identified in the glacier zone to be identified is determined, wherein the glacier types include seasonal glaciers and permanent glaciers.

[0009] Optionally, calculating the normalized snow index of each pixel in each target remote sensing image includes:

[0010] For any target remote sensing image, any pixel point is selected in the target remote sensing image, and the normalized snow index of the selected pixel point is calculated according to the normalized snow index calculation formula, the green band surface reflectivity and the short-wave infrared band surface reflectivity of the selected pixel point, wherein the normalized snow index calculation formula is:

[0011]

[0012] NDSI i,j is the normalized snow cover index of the jth pixel of the i-th target remote sensing image, ρ GREEN,i,j is the green band surface reflectance of the jth pixel point of the i-th target remote sensing image, ρ MIR,i,j is the surface reflectivity of the shortwave infrared band of the jth pixel point of the i-th target remote sensing image.

[0013] Optionally, the determining of the glacier type distinction threshold corresponding to the glacier zone snow frequency map based on the maximum inter-class variance method includes:

[0014] The inter-class variance calculation formula is used to calculate the inter-class variance corresponding to each pixel in the snow frequency map of the glacier zone, and the pixel value of the pixel corresponding to the maximum inter-class variance is determined as the glacier type distinction threshold, wherein the inter-class variance calculation formula is:

[0015] σ k 2 =P k,季 (m 季 -m G ) 2 +P k,永 (m 永 -m G ) 2 ,

[0016] σ k is the inter-class variance of the kth pixel in the snow frequency map of the glacier zone, P k,季 is the probability that pixel k belongs to seasonal glacier, m 季 is the average grayscale of seasonal glaciers, m Gis the average grayscale of the snow frequency map in the glacier zone, P k,永 is the probability that pixel k belongs to a permanent glacier, m 永 is the average grayscale of permanent glaciers.

[0017] Optionally, before calculating the inter-class variance corresponding to each pixel point in the snow frequency map in the glacier zone according to the inter-class variance calculation formula, the method further includes:

[0018] Calculate the occurrence probability of various pixel values ​​in the glacier zone snow frequency map, select any pixel value, and determine seasonal glaciers and permanent glaciers in the glacier zone snow frequency map based on the comparison between other pixel values ​​and the selected pixel value;

[0019] Select a seasonal glacier or a permanent glacier as the target glacier, calculate the sum of the occurrence probabilities corresponding to various pixel values ​​in the selected target glacier, obtain the probability that the pixel belongs to the target glacier, and calculate the average grayscale of the target glacier based on the glacier average grayscale calculation formula, where the glacier average grayscale calculation formula is:

[0020]

[0021] m t,目 is the average grayscale of the target glacier when the t-th pixel value is selected, P(mesh / t) is the probability that the t-th pixel value belongs to the target glacier, P(t) is the probability of the t-th pixel value appearing in the snow frequency map of the glacier zone, and P(mesh) is the sum of the occurrence probabilities of various pixel values ​​classified as the target glacier.

[0022] Optionally, the determining the glacier type of each glacier to be identified in the glacier zone to be identified based on the pixel value of each pixel point in the snow frequency map of the glacier zone and the glacier type distinction threshold comprises:

[0023] In the snow frequency map of the glacier zone, the initial edge of each glacier to be identified in the glacier zone to be identified is obtained based on an edge detection algorithm, and the initial edge of the glacier is morphologically expanded to obtain a smooth edge of the glacier;

[0024] In the snow frequency map of the glacier zone containing the smooth edge of the glacier, the glacier edge judgment threshold is determined based on the maximum inter-class variance method, and the fine edge of the glacier corresponding to each glacier to be identified is determined based on the pixel value of each pixel point corresponding to the smooth edge of the glacier and the glacier edge judgment threshold;

[0025] For any glacier to be identified whose fine edge has been determined in the glacier zone to be identified, the glacier type of the glacier to be identified is determined based on the pixel value of each pixel point corresponding to the glacier to be identified and the glacier type distinction threshold.

[0026] Optionally, generating a snow frequency map of the glacier zone within the preset time period according to a normalized snow index of each pixel point in each target remote sensing image and a preset snow frequency threshold comprises:

[0027] For any target remote sensing image, any pixel point is selected in the target remote sensing image, and when the normalized snow accumulation index of the selected pixel point is greater than or equal to the preset snow accumulation frequency threshold, the snow accumulation frequency of the selected pixel point is increased by one counting unit until a final counting result of the snow accumulation frequency of the selected pixel point calculated based on each target remote sensing image within a preset time period is obtained;

[0028] According to the final counting result of the snow accumulation frequency in the preset time period corresponding to each pixel point, a snow accumulation frequency map of the glacier area in the preset time period is generated.

[0029] Optionally, the determining the glacier type of each glacier to be identified in the glacier zone to be identified based on the pixel value of each pixel point in the snow frequency map of the glacier zone and the glacier type distinction threshold comprises:

[0030] Determine the seasonal glacier pixel assignment values ​​for seasonal glaciers and the permanent glacier pixel assignment values ​​for permanent glaciers;

[0031] In the snow frequency map of the glacier zone, for any pixel point corresponding to the glacier to be identified, when the pixel value of the pixel point is greater than or equal to the glacier type distinction threshold, the pixel value of the pixel point is assigned a seasonal glacier pixel assignment value; when the pixel value of the pixel point is less than the glacier type distinction threshold, the pixel value of the pixel point is assigned a permanent glacier pixel assignment value;

[0032] Seasonal glaciers are determined based on the displayed area where values ​​are assigned to seasonal glacier pixels, and permanent glaciers are determined based on the displayed area where values ​​are assigned to permanent glacier pixels.

[0033] According to another aspect of the present application, a glacier identification device is provided, the device comprising:

[0034] A remote sensing image acquisition module, used to acquire remote sensing images to be processed at different periods of a glacier zone to be identified within a preset time period, wherein the glacier zone to be identified includes a plurality of glaciers to be identified;

[0035] A snow index calculation module is used to perform cloud removal processing on each of the remote sensing images to be processed to obtain a target remote sensing image, and calculate a normalized snow index of each pixel in each of the target remote sensing images;

[0036] A snow frequency calculation module, used to generate a snow frequency map of the glacier zone within the preset time period according to the normalized snow index of each pixel point in each target remote sensing image and a preset snow frequency threshold;

[0037] The glacier type identification module is used to determine the glacier type distinction threshold corresponding to the snow frequency map of the glacier zone based on the maximum inter-class variance method, and determine the glacier type of each glacier to be identified in the glacier zone to be identified based on the pixel value of each pixel point in the snow frequency map of the glacier zone and the glacier type distinction threshold, wherein the glacier type includes seasonal glaciers and permanent glaciers.

[0038] Optionally, the snow index calculation module is further used to:

[0039] For any target remote sensing image, any pixel point is selected in the target remote sensing image, and the normalized snow index of the selected pixel point is calculated according to the normalized snow index calculation formula, the green band surface reflectivity and the short-wave infrared band surface reflectivity of the selected pixel point, wherein the normalized snow index calculation formula is:

[0040]

[0041] NDSI i,j is the normalized snow cover index of the jth pixel of the i-th target remote sensing image, ρ GREEN,i,j is the green band surface reflectance of the jth pixel point of the i-th target remote sensing image, ρ MIR,i,j is the surface reflectivity of the shortwave infrared band of the jth pixel point of the i-th target remote sensing image.

[0042] Optionally, the device further comprises: a threshold automatic determination module, configured to:

[0043] The inter-class variance calculation formula is used to calculate the inter-class variance corresponding to each pixel in the snow frequency map of the glacier zone, and the pixel value of the pixel corresponding to the maximum inter-class variance is determined as the glacier type distinction threshold, wherein the inter-class variance calculation formula is:

[0044] σ k 2 =P k,季 (m 季 -m G ) 2 +P k,永 (m 永 -m G ) 2 ,

[0045] σ k is the inter-class variance of the kth pixel in the snow frequency map of the glacier zone, P k,季 is the probability that pixel k belongs to seasonal glacier, m 季 is the average grayscale of seasonal glaciers, m G is the average grayscale of the snow frequency map in the glacier zone, Pk,永 is the probability that pixel k belongs to a permanent glacier, m 永 is the average grayscale of permanent glaciers.

[0046] Optionally, the threshold automatic determination module is further used to:

[0047] Calculate the occurrence probability of various pixel values ​​in the glacier zone snow frequency map, select any pixel value, and determine seasonal glaciers and permanent glaciers in the glacier zone snow frequency map based on the comparison between other pixel values ​​and the selected pixel value;

[0048] Select a seasonal glacier or a permanent glacier as the target glacier, calculate the sum of the occurrence probabilities corresponding to various pixel values ​​in the selected target glacier, obtain the probability that the pixel belongs to the target glacier, and calculate the average grayscale of the target glacier based on the glacier average grayscale calculation formula, where the glacier average grayscale calculation formula is:

[0049]

[0050] m t,目 is the average grayscale of the target glacier when the t-th pixel value is selected, P(mesh / t) is the probability that the t-th pixel value belongs to the target glacier, P(t) is the probability of the t-th pixel value appearing in the snow frequency map of the glacier zone, and P(mesh) is the sum of the occurrence probabilities of various pixel values ​​classified as the target glacier.

[0051] Optionally, the glacier type identification module is further used to:

[0052] In the snow frequency map of the glacier zone, the initial edge of each glacier to be identified in the glacier zone to be identified is obtained based on an edge detection algorithm, and the initial edge of the glacier is morphologically expanded to obtain a smooth edge of the glacier;

[0053] In the snow frequency map of the glacier zone containing the smooth edge of the glacier, the glacier edge judgment threshold is determined based on the maximum inter-class variance method, and the fine edge of the glacier corresponding to each glacier to be identified is determined based on the pixel value of each pixel point corresponding to the smooth edge of the glacier and the glacier edge judgment threshold;

[0054] For any glacier to be identified whose fine edge has been determined in the glacier zone to be identified, the glacier type of the glacier to be identified is determined based on the pixel value of each pixel point corresponding to the glacier to be identified and the glacier type distinction threshold.

[0055] Optionally, the snow accumulation frequency calculation module is further used to:

[0056] For any target remote sensing image, any pixel point is selected in the target remote sensing image, and when the normalized snow accumulation index of the selected pixel point is greater than or equal to the preset snow accumulation frequency threshold, the snow accumulation frequency of the selected pixel point is increased by one counting unit until a final counting result of the snow accumulation frequency of the selected pixel point calculated based on each target remote sensing image within a preset time period is obtained;

[0057] According to the final counting result of the snow accumulation frequency in the preset time period corresponding to each pixel point, a snow accumulation frequency map of the glacier area in the preset time period is generated.

[0058] Optionally, the glacier type identification module is further used to:

[0059] Determine the seasonal glacier pixel assignment values ​​for seasonal glaciers and the permanent glacier pixel assignment values ​​for permanent glaciers;

[0060] In the snow frequency map of the glacier zone, for any pixel point corresponding to the glacier to be identified, when the pixel value of the pixel point is greater than or equal to the glacier type distinction threshold, the pixel value of the pixel point is assigned a seasonal glacier pixel assignment value; when the pixel value of the pixel point is less than the glacier type distinction threshold, the pixel value of the pixel point is assigned a permanent glacier pixel assignment value;

[0061] Seasonal glaciers are determined based on the displayed area where values ​​are assigned to seasonal glacier pixels, and permanent glaciers are determined based on the displayed area where values ​​are assigned to permanent glacier pixels.

[0062] According to another aspect of the present application, a storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the above-mentioned glacier identification method is implemented.

[0063] According to another aspect of the present application, a computer device is provided, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor implements the above-mentioned glacier identification method when executing the program.

[0064] By means of the above technical scheme, the present application provides a glacier identification method and device, storage medium, and computer equipment, which obtain remote sensing images to be processed at different periods in a preset time period of a glacier zone to be identified, wherein the glacier zone to be identified contains multiple glaciers to be identified; perform cloud removal on each remote sensing image to be processed to obtain a target remote sensing image, and calculate the normalized snow index of each pixel in each target remote sensing image; generate a snow frequency map of the glacier zone within a preset time period according to the normalized snow index of each pixel in each target remote sensing image and a preset snow frequency threshold; determine the glacier type distinction threshold corresponding to the snow frequency map of the glacier zone based on the maximum inter-class variance method, and determine the glacier type of each glacier to be identified in the glacier zone to be identified based on the pixel value of each pixel in the snow frequency map of the glacier zone and the glacier type distinction threshold, wherein the glacier type includes seasonal glaciers and permanent glaciers. By automatically determining the glacier type distinction threshold, it is possible to identify whether the glacier to be identified is a seasonal glacier or a permanent glacier, thereby improving the efficiency of glacier identification.

[0065] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0067] Figure 1 A schematic diagram of a process of a glacier identification method provided in an embodiment of the present application is shown;

[0068] Figure 2 A schematic diagram of a process flow of another glacier identification method provided in an embodiment of the present application is shown;

[0069] Figure 3 A schematic diagram of a process flow of another glacier identification method provided in an embodiment of the present application is shown;

[0070] Figure 4 A schematic diagram of the structure of a glacier identification device provided in an embodiment of the present application is shown;

[0071] Figure 5 A schematic structural diagram of another glacier identification device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0072] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0073] In this embodiment, a glacier identification method is provided. Figure 1 As shown, the method includes:

[0074] Step 101, obtaining remote sensing images to be processed at different periods of a glacier zone to be identified within a preset time period, wherein the glacier zone to be identified includes a plurality of glaciers to be identified.

[0075] Step 102, performing cloud removal processing on each of the remote sensing images to be processed to obtain a target remote sensing image, and calculating a normalized snow accumulation index of each pixel in each of the target remote sensing images.

[0076] In the above embodiment of the present application, remote sensing images to be processed at different periods in a preset time period are obtained for the glacier area to be identified. For example, the glacier area to be identified is the Qinghai-Tibet Plateau area. The aforementioned Qinghai-Tibet Plateau area contains multiple glaciers to be identified. Specifically, remote sensing images to be processed in the Qinghai-Tibet Plateau area for nearly one year can be obtained in GEE (Google Earth Engine, a remote sensing big data analysis platform based on Google cloud service infrastructure), and then each remote sensing image to be processed is de-clouded to obtain the target remote sensing image. Then, for each target remote sensing image, the normalized snow accumulation index of each pixel in the aforementioned target remote sensing image is calculated to prepare for subsequent glacier identification.

[0077] Step 103, generating a snow frequency map of the glacier zone within the preset time period according to the normalized snow index of each pixel point in each of the target remote sensing images and a preset snow frequency threshold.

[0078] Next, the number of pixels corresponding to the normalized snow index that meets the preset snow frequency threshold is counted, so as to generate a glacier zone snow frequency map reflecting the "snow frequency" of the glacier to be identified. In particular, the preset snow frequency threshold can be set to 0.4.

[0079] Step 104, based on the maximum inter-class variance method, determine the glacier type distinction threshold corresponding to the snow frequency map of the glacier zone, and determine the glacier type of each glacier to be identified in the glacier zone based on the pixel value of each pixel point in the snow frequency map of the glacier zone and the glacier type distinction threshold, wherein the glacier type includes seasonal glaciers and permanent glaciers.

[0080] Next, in the snow frequency map of the glacier zone, the glacier type distinction threshold corresponding to the snow frequency map of the glacier zone is determined based on the maximum inter-class variance method, and the glacier type of each glacier to be identified in the glacier zone to be identified is determined based on the pixel value of each pixel point in the snow frequency map of the glacier zone and the glacier type distinction threshold, where the glacier type includes seasonal glaciers and permanent glaciers. The maximum inter-class variance method can automatically determine the "optimal" glacier type distinction threshold, and combined with the "snow frequency" information, it can more accurately identify whether the glacier is a seasonal glacier or a permanent glacier, which increases the recognition accuracy of the glacier and improves the efficiency of glacier recognition.

[0081] By applying the technical solution of this embodiment, remote sensing images to be processed at different periods in the glacier zone to be identified within a preset time period are obtained; each remote sensing image to be processed is declouded to obtain a target remote sensing image, and the normalized snow index of each pixel in each target remote sensing image is calculated; a snow frequency map of the glacier zone within a preset time period is generated based on the normalized snow index of each pixel in each target remote sensing image and a preset snow frequency threshold; based on the maximum inter-class variance method, the glacier type distinction threshold corresponding to the snow frequency map of the glacier zone is determined, and the glacier type of each glacier to be identified in the glacier zone to be identified is determined based on the pixel value of each pixel in the snow frequency map of the glacier zone and the glacier type distinction threshold. By automatically determining the glacier type distinction threshold, it is possible to identify whether the glacier to be identified is a seasonal glacier or a permanent glacier, thereby improving the efficiency of glacier identification.

[0082] Furthermore, as a refinement and extension of the specific implementation of the above embodiment, in order to fully illustrate the specific implementation process of this embodiment, another glacier identification method is provided, such as Figure 2 As shown, the method includes:

[0083] Step 201, obtaining remote sensing images to be processed at different periods within a preset time period of a glacier zone to be identified, performing cloud removal processing on each of the remote sensing images to be processed to obtain a target remote sensing image, wherein the glacier zone to be identified includes multiple glaciers to be identified.

[0084] In the above-mentioned embodiment of the present application, remote sensing images to be processed of the glacier zone to be identified at different periods within a preset time period are obtained. Specifically, remote sensing images to be processed of the glacier zone to be identified in the past year can be obtained, and each remote sensing image to be processed is subjected to cloud removal processing to obtain a target remote sensing image, so as to prevent the "cloud" image in the remote sensing image from affecting the subsequent identification of glaciers.

[0085] Step 202, for any target remote sensing image, select any pixel point in the target remote sensing image, and calculate the normalized snow index of the selected pixel point according to the normalized snow index calculation formula, the green band surface reflectance and the short-wave infrared band surface reflectance of the selected pixel point, wherein the normalized snow index calculation formula is:

[0086]

[0087] NDSI i,j is the normalized snow cover index of the jth pixel of the i-th target remote sensing image, ρ GREEN,i,j is the green band surface reflectance of the jth pixel point of the i-th target remote sensing image, ρ MIR,i,j is the surface reflectivity of the shortwave infrared band of the jth pixel point of the i-th target remote sensing image.

[0088] Next, in the target remote sensing image after declouding, for any target remote sensing image, any pixel point is selected in the aforementioned target remote sensing image, and the normalized snow index of the selected pixel point is calculated according to the normalized snow index calculation formula, the green band surface reflectivity of the selected pixel point, and the shortwave infrared band surface reflectivity. Therefore, the normalized snow index of each pixel point in the target remote sensing image is obtained.

[0089] Step 203, generating a snow frequency map of the glacier zone within the preset time period according to the normalized snow index of each pixel point in each of the target remote sensing images and a preset snow frequency threshold.

[0090] Next, based on the normalized snow index of each pixel point in each target remote sensing image and the preset snow frequency threshold, a snow frequency map of the glacier zone within the preset time period is generated. The snow frequency map of the glacier zone can reflect the snow frequency of multiple glaciers to be identified in the glacier zone to be identified, each of which has a preset time period. The snow frequency can be used to identify whether the glacier is a seasonal glacier or a permanent glacier, which can be used to prepare for the subsequent identification of glacier types. In particular, the selection of the preset snow frequency threshold can be set to 0.4.

[0091] Step 204, calculate the occurrence probability of various pixel values ​​in the glacier zone snow frequency map, select any pixel value, and determine seasonal glaciers and permanent glaciers in the glacier zone snow frequency map based on the comparison between other pixel values ​​and the selected pixel value.

[0092] Step 205, select a seasonal glacier or a permanent glacier as a target glacier, calculate the sum of the occurrence probabilities corresponding to various pixel values ​​in the selected target glacier, obtain the probability that the pixel belongs to the target glacier, and calculate the average grayscale of the target glacier based on the glacier average grayscale calculation formula, wherein the glacier average grayscale calculation formula is:

[0093]

[0094] m t,目 is the average grayscale of the target glacier when the t-th pixel value is selected, P(mesh / t) is the probability that the t-th pixel value belongs to the target glacier, P(t) is the probability of the t-th pixel value appearing in the snow frequency map of the glacier zone, and P(mesh) is the sum of the occurrence probabilities of various pixel values ​​classified as the target glacier.

[0095] Step 206, calculating the inter-class variance corresponding to each pixel point in the snow frequency map of the glacier zone according to the inter-class variance calculation formula, and determining the pixel value of the pixel point corresponding to the maximum inter-class variance as the glacier type distinction threshold, wherein the inter-class variance calculation formula is:

[0096] σ k 2 =P k,季 (m 季 -m G ) 2 +P k,永 (m 永 -m G ) 2 ,

[0097] σ k is the inter-class variance of the kth pixel in the snow frequency map of the glacier zone, P k,季 is the probability that pixel k belongs to seasonal glacier, m 季 is the average grayscale of seasonal glaciers, m G is the average grayscale of the snow frequency map in the glacier zone, P k,永 is the probability that pixel k belongs to a permanent glacier, m 永 is the average grayscale of permanent glaciers.

[0098] Step 207, based on the pixel value of each pixel point in the snow frequency map of the glacier zone and the glacier type distinction threshold, determine the glacier type of each glacier to be identified in the glacier zone to be identified, wherein the glacier type includes seasonal glaciers and permanent glaciers.

[0099] Next, in the obtained glacier zone snow frequency map, the occurrence probability of each pixel value in the glacier zone snow frequency map is calculated. The pixel value range is 0 to 255. The calculated occurrence probability of each pixel value in 0 to 255 is shown in Table 1:

[0100] Table 1

[0101] Pixel Value 0 1 2 3 ... 251 252 253 254 255 The probability of a pixel value occurring 0.03 0.01 0.02 0.0 ... 0.01 0.03 0.0 0.0 0.0

[0102] Select any pixel value (for example, select pixel value 3), and then determine seasonal glaciers and permanent glaciers in the snow frequency map of the glacier zone based on the comparison between other pixel values ​​(other pixel values ​​except 3 in the range of 0 to 255) and the selected pixel value (3). Specifically, it can be set to classify various pixel values ​​greater than or equal to 3 as seasonal glaciers, and various pixel values ​​less than 3 as permanent glaciers. At this time, pixel values ​​of 0, 1, and 2 are classified as perpetual glaciers, and the remaining pixel values ​​of 3 to 255 are classified as seasonal glaciers. For this reason, seasonal glaciers and permanent glaciers are determined in the snow frequency map of the glacier zone based on the comparison between other pixel values ​​and the selected pixel value.

[0103] Then, a seasonal glacier can be selected as the target glacier. At this time, the sum of the occurrence probabilities corresponding to various pixel values ​​(3 to 255) in the seasonal glacier is calculated, that is, the probability that the pixel value belongs to the seasonal glacier (target glacier) is obtained. In particular, if a permanent glacier is selected as the target glacier, the probability that the pixel value belongs to the permanent glacier is the sum of the occurrence probabilities corresponding to the pixel values ​​0, 1, and 2, that is, 0.03+0.01+0.02=0.06.

[0104] Then, the average grayscale of seasonal glaciers and permanent glaciers is calculated respectively by the glacier average grayscale calculation formula. For example, when the t-th pixel value is 3, P(mesh / t) is the probability that the pixel value 3 belongs to the seasonal glacier. Since the pixel value 3 is classified as a seasonal glacier and the seasonal glacier is the target glacier, the probability P(mesh / t) that the pixel value 3 belongs to the seasonal glacier is 1. P(t) is the probability of occurrence of the pixel value 3 in the snow frequency map of the glacier zone. It can be seen from Table 1 that it is 0. P(mesh) is the sum of the occurrence probabilities of various pixel values ​​contained in the target glacier, i.e., the seasonal glacier. It can be seen from Table 1 that it is the sum of the occurrence probabilities of each pixel value from pixel value 3 to pixel value 255.

[0105] Next, the inter-class variance corresponding to each pixel is calculated according to the inter-class variance calculation formula, and the pixel value of the pixel corresponding to the maximum inter-class variance is determined as the glacier type distinction threshold.

[0106] Finally, based on the pixel value of each pixel point in the snow frequency map of the glacier zone and the glacier type distinction threshold, the glacier type of each glacier to be identified in the glacier zone to be identified is determined, where the glacier type includes seasonal glaciers and permanent glaciers.

[0107] Specifically, the glacier type distinction threshold can be used to perform binarization processing on the snow frequency map of the glacier zone, and the processed snow frequency map of the glacier zone can clearly distinguish between seasonal glaciers and permanent glaciers. In the process of binarization processing, for example, pixel values ​​greater than or equal to the glacier type distinction threshold can be classified as seasonal glaciers, and pixel values ​​less than the glacier type distinction threshold can be classified as permanent glaciers. At the same time, the pixel values ​​classified as seasonal glaciers are all adjusted to 0 (darkest), and the pixel values ​​classified as permanent glaciers are all adjusted to 255 (brightest). Therefore, seasonal glaciers and permanent glaciers can be clearly distinguished.

[0108] By applying the technical solution of this embodiment, remote sensing images to be processed at different periods of the glacier zone to be identified within a preset time period are obtained for declouding to obtain the target remote sensing image. For any target remote sensing image, the normalized snow index of the pixel point is calculated according to the normalized snow index calculation formula, and the snow frequency map of the glacier zone is obtained based on the normalized snow index of each pixel point and the preset snow frequency threshold. In the snow frequency map of the glacier zone, the occurrence probability of each pixel value in the snow frequency map of the glacier zone is calculated, and any pixel value is selected. Based on the comparison between other pixel values ​​and the selected pixel value, seasonal glaciers and permanent glaciers are determined in the snow frequency map of the glacier zone. The seasonal glacier or permanent glacier is selected as the target glacier, and the sum of the occurrence probabilities corresponding to each pixel value in the selected target glacier is calculated to obtain the probability that the pixel belongs to the target glacier, and the average grayscale of the target glacier is calculated based on the glacier average grayscale calculation formula, and the inter-class variance corresponding to each pixel point is calculated according to the inter-class variance calculation formula, and the pixel value of the pixel point corresponding to the maximum inter-class variance is determined as the glacier type distinction threshold. Based on the pixel value of each pixel point in the snow frequency map of the glacier zone and the glacier type distinction threshold, the glacier type of each glacier to be identified in the glacier zone to be identified is determined. By automatically determining the glacier type distinction threshold, it is possible to identify whether the glacier to be identified belongs to a seasonal glacier or a permanent glacier, thereby improving the efficiency of glacier identification.

[0109] Further, as a refinement and extension of the specific implementation of the above embodiment, in order to fully illustrate the specific implementation process of this embodiment, another glacier identification method is provided, such as Figure 3 As shown, the method includes:

[0110] Step 301, obtaining remote sensing images to be processed at different periods within a preset time period of the glacier zone to be identified, performing cloud removal processing on each of the remote sensing images to be processed to obtain a target remote sensing image, and calculating a normalized snow index of each pixel point in each of the target remote sensing images, wherein the glacier zone to be identified contains multiple glaciers to be identified.

[0111] In the above embodiment of the present application, remote sensing images to be processed at different periods of the glacier area to be identified within a preset time period are obtained, each remote sensing image to be processed is declouded to obtain a target remote sensing image, and the normalized snow accumulation index of each pixel point in each target remote sensing image is calculated so that the snow accumulation frequency of "each pixel point" can be determined by the normalized snow accumulation index.

[0112] Step 302, for any target remote sensing image, select any pixel point in the target remote sensing image, and when the normalized snow accumulation index of the selected pixel point is greater than or equal to the preset snow accumulation frequency threshold, increase the snow accumulation frequency of the selected pixel point by one counting unit until a final counting result of the snow accumulation frequency of the selected pixel point calculated based on each target remote sensing image within a preset time period is obtained.

[0113] Step 303, generating a snow frequency map of the glacier zone within the preset time period according to the final counting result of the snow frequency within the preset time period corresponding to each pixel point.

[0114] Next, for any target remote sensing image, any pixel point is selected in the target remote sensing image, and when the normalized snow accumulation index of the pixel point is greater than or equal to the preset snow accumulation frequency threshold, the snow accumulation frequency of the pixel point is increased by one counting unit. For example, a total of 365 target remote sensing images are currently acquired, among which, in a total of 225 target remote sensing images, the normalized snow accumulation index of the pixel point at the same position is greater than or equal to the preset snow accumulation frequency threshold. In the finally formed glacier zone snow accumulation frequency map, the final counting result of the snow accumulation frequency of the pixel point at the aforementioned position is "225", and accordingly, the pixel point is highlighted and displayed, and the ratio of the highlighted display can be a first ratio of 225 to 365, and the pixel value calculated by the total pixel value range of 255 is displayed (157), and the calculation result can be rounded off. Finally, according to the final counting result of the snow accumulation frequency in the preset time period corresponding to each pixel point, the glacier zone snow accumulation frequency map in the preset time period is generated, and the snow accumulation frequency of each pixel point in the preset time period can be displayed through the glacier zone snow accumulation frequency map.

[0115] Step 304, based on the maximum inter-class variance method, determine the glacier type differentiation threshold corresponding to the snow frequency map of the glacier zone, in the snow frequency map of the glacier zone, obtain the initial glacier edge of each glacier to be identified in the glacier zone to be identified based on the edge detection algorithm, perform morphological expansion on the initial glacier edge, and obtain the smooth edge of the glacier.

[0116] Step 305, in the snow frequency map of the glacier zone including the smooth edge of the glacier, determine the glacier edge judgment threshold based on the maximum inter-class variance method, and determine the glacier fine edge corresponding to each glacier to be identified based on the pixel value of each pixel point corresponding to the smooth edge of the glacier and the glacier edge judgment threshold.

[0117] Next, in the snow frequency map of the glacier zone, based on the maximum inter-class variance method, the glacier type discrimination threshold corresponding to the snow frequency map of the glacier zone is determined. In order to further accurately determine the "specific location" of the glacier to be identified and improve the accuracy of identification, in the snow frequency map of the glacier zone, the initial edge of the glacier of each glacier to be identified in the glacier zone to be identified is obtained based on the edge detection algorithm, and the initial edge of the glacier is morphologically expanded to obtain the smooth edge of the glacier. Next, in the snow frequency map of the glacier zone containing the smooth edge of the glacier, the glacier edge judgment threshold is determined based on the maximum inter-class variance method, and the fine edge of the glacier corresponding to each glacier to be identified is determined based on the pixel value of each pixel point corresponding to the smooth edge of the glacier and the glacier edge judgment threshold. To this end, by first "roughly" identifying the glacier edge, then determining the glacier edge judgment threshold, and further identifying the final fine edge of the glacier, the edge recognition accuracy of the glacier to be identified can be improved.

[0118] Step 306, determining the seasonal glacier pixel assignment value for the seasonal glacier and the permanent glacier pixel assignment value for the permanent glacier.

[0119] Step 307, in the snow frequency map of the glacier zone, for any glacier to be identified in the glacier zone with a determined fine edge of the glacier, when the pixel value of the pixel point corresponding to the glacier to be identified is greater than or equal to the glacier type distinction threshold, the pixel value of the pixel point is assigned a seasonal glacier pixel assignment value.

[0120] Step 308: When the pixel value of the pixel point corresponding to the glacier to be identified is less than the glacier type distinction threshold, the pixel value of the pixel point is assigned a permanent glacier pixel assignment value.

[0121] Step 309, seasonal glaciers are determined based on the display area where the values ​​of the seasonal glacier pixels are assigned, and permanent glaciers are determined based on the display area where the values ​​of the permanent glacier pixels are assigned.

[0122] Finally, the seasonal glacier pixel assignment value (for example, it can be 0) of the seasonal glacier and the permanent glacier pixel assignment value (for example, it can be 255) of the permanent glacier are determined. In the glacier zone snow frequency map, for any glacier to be identified in the glacier zone to be identified whose fine edge has been determined, when the pixel value of the pixel point corresponding to the glacier to be identified is greater than or equal to the glacier type distinction threshold, the pixel value of the pixel point is assigned the seasonal glacier pixel assignment value. When the pixel value of the pixel point corresponding to the glacier to be identified is less than the glacier type distinction threshold, the pixel value of the pixel point is assigned the permanent glacier pixel assignment value. At this point, in the glacier zone snow frequency map, seasonal glaciers can be determined based on the display area of ​​seasonal glacier pixel assignment values, and permanent glaciers can be determined based on the display area of ​​permanent glacier pixel assignment values. By clearly distinguishing seasonal glaciers from permanent glaciers, the glacier type can be identified more clearly and quickly, thereby improving the recognition efficiency.

[0123] By applying the technical solution of this embodiment, remote sensing images to be processed at different periods in a preset time period of the glacier zone to be identified are obtained for declouding to obtain the target remote sensing image, the normalized snow index of each pixel in the target remote sensing image is calculated, and the snow frequency map of the glacier zone is obtained. Based on the maximum inter-class variance method, the glacier type distinction threshold corresponding to the snow frequency map of the glacier zone is determined, and the initial glacier edge of each glacier to be identified in the glacier zone to be identified is obtained based on the edge detection algorithm, and the initial glacier edge is morphologically expanded to obtain the smooth edge of the glacier. In the snow frequency map of the glacier zone containing the smooth edge of the glacier, the glacier edge judgment threshold is determined based on the maximum inter-class variance method, and the fine edge of the glacier corresponding to each glacier to be identified is determined based on the pixel value of each pixel corresponding to the smooth edge of the glacier and the glacier edge judgment threshold. The seasonal glacier pixel assignment value of the seasonal glacier and the permanent glacier pixel assignment value of the permanent glacier are determined. In the snow frequency map of the glacier zone, for any glacier to be identified in the glacier zone to be identified whose fine edge has been determined, when the pixel value of the pixel of the glacier to be identified is greater than or equal to the glacier type distinction threshold, the pixel value of the pixel is assigned the seasonal glacier pixel assignment value. When the pixel value of the pixel point of the glacier to be identified is less than the glacier type distinction threshold, the pixel value of the pixel point is assigned the permanent glacier pixel assignment value. By automatically determining the glacier type distinction threshold, it is possible to identify whether the glacier to be identified is a seasonal glacier or a permanent glacier, thereby improving the efficiency of glacier identification.

[0124] Further, as Figure 1 The specific implementation of the method, the embodiment of the present application provides a glacier identification device, such as Figure 4 As shown, the device comprises:

[0125] The remote sensing image acquisition module 401 is used to acquire remote sensing images to be processed at different periods of the glacier zone to be identified within a preset time period, wherein the glacier zone to be identified includes a plurality of glaciers to be identified;

[0126] The snow index calculation module 402 is used to perform cloud removal processing on each of the remote sensing images to be processed to obtain a target remote sensing image, and calculate the normalized snow index of each pixel in each of the target remote sensing images;

[0127] The snow frequency calculation module 403 is used to generate a snow frequency map of the glacier zone within the preset time period according to the normalized snow index of each pixel point in each target remote sensing image and a preset snow frequency threshold;

[0128] The glacier type identification module 404 is used to determine the glacier type distinction threshold corresponding to the snow frequency map of the glacier zone based on the maximum inter-class variance method, and determine the glacier type of each glacier to be identified in the glacier zone to be identified based on the pixel value of each pixel point in the snow frequency map of the glacier zone and the glacier type distinction threshold, wherein the glacier type includes seasonal glaciers and permanent glaciers.

[0129] Optionally, the snow index calculation module 402 is further used to:

[0130] For any target remote sensing image, any pixel point is selected in the target remote sensing image, and the normalized snow index of the selected pixel point is calculated according to the normalized snow index calculation formula, the green band surface reflectivity and the short-wave infrared band surface reflectivity of the selected pixel point, wherein the normalized snow index calculation formula is:

[0131]

[0132] NDSI i,j is the normalized snow cover index of the jth pixel of the i-th target remote sensing image, ρ GREEN,i,j is the green band surface reflectance of the jth pixel point of the i-th target remote sensing image, ρ MIR,i,j is the surface reflectivity of the shortwave infrared band of the jth pixel point of the i-th target remote sensing image.

[0133] Optionally, the glacier type identification module 404 is further configured to:

[0134] In the snow frequency map of the glacier zone, the initial edge of each glacier to be identified in the glacier zone to be identified is obtained based on an edge detection algorithm, and the initial edge of the glacier is morphologically expanded to obtain a smooth edge of the glacier;

[0135] In the snow frequency map of the glacier zone containing the smooth edge of the glacier, the glacier edge judgment threshold is determined based on the maximum inter-class variance method, and the fine edge of the glacier corresponding to each glacier to be identified is determined based on the pixel value of each pixel point corresponding to the smooth edge of the glacier and the glacier edge judgment threshold;

[0136] For any glacier to be identified whose fine edge has been determined in the glacier zone to be identified, the glacier type of the glacier to be identified is determined based on the pixel value of each pixel point corresponding to the glacier to be identified and the glacier type distinction threshold.

[0137] Optionally, the snow accumulation frequency calculation module 403 is further used to:

[0138] For any target remote sensing image, any pixel point is selected in the target remote sensing image, and when the normalized snow accumulation index of the selected pixel point is greater than or equal to the preset snow accumulation frequency threshold, the snow accumulation frequency of the selected pixel point is increased by one counting unit until a final counting result of the snow accumulation frequency of the selected pixel point calculated based on each target remote sensing image within a preset time period is obtained;

[0139] According to the final counting result of the snow accumulation frequency in the preset time period corresponding to each pixel point, a snow accumulation frequency map of the glacier area in the preset time period is generated.

[0140] Optionally, the glacier type identification module 404 is further configured to:

[0141] For any target remote sensing image, any pixel point is selected in the target remote sensing image, and when the normalized snow accumulation index of the selected pixel point is greater than or equal to the preset snow accumulation frequency threshold, the snow accumulation frequency of the selected pixel point is increased by one counting unit until a final counting result of the snow accumulation frequency of the selected pixel point calculated based on each target remote sensing image within a preset time period is obtained;

[0142] According to the final counting result of the snow accumulation frequency in the preset time period corresponding to each pixel point, a snow accumulation frequency map of the glacier area in the preset time period is generated.

[0143] Furthermore, the present application embodiment provides another glacier identification device, such as Figure 5 As shown, the device comprises:

[0144] The remote sensing image acquisition module 501 is used to acquire remote sensing images to be processed at different periods of the glacier zone to be identified within a preset time period, wherein the glacier zone to be identified includes a plurality of glaciers to be identified;

[0145] The snow index calculation module 502 is used to perform cloud removal processing on each of the remote sensing images to be processed to obtain a target remote sensing image, and calculate the normalized snow index of each pixel in each of the target remote sensing images;

[0146] The snow frequency calculation module 503 is used to generate a snow frequency map of the glacier zone within the preset time period according to the normalized snow index of each pixel point in each target remote sensing image and a preset snow frequency threshold;

[0147] The glacier type identification module 504 is used to determine the glacier type distinction threshold corresponding to the snow frequency map of the glacier zone based on the maximum inter-class variance method, and determine the glacier type of each glacier to be identified in the glacier zone to be identified based on the pixel value of each pixel point in the snow frequency map of the glacier zone and the glacier type distinction threshold, wherein the glacier type includes seasonal glaciers and permanent glaciers;

[0148] The threshold automatic determination module 505 calculates the inter-class variance corresponding to each pixel point in the snow frequency map of the glacier zone according to the inter-class variance calculation formula, and determines the pixel value of the pixel point corresponding to the maximum inter-class variance as the glacier type distinction threshold, wherein the inter-class variance calculation formula is:

[0149] σ k 2 =P k,季 (m 季 -m G ) 2 +P k,永 (m 永 -m G ) 2 ,

[0150] σ k is the inter-class variance of the kth pixel in the snow frequency map of the glacier zone, P k,季 is the probability that pixel k belongs to seasonal glacier, m 季 is the average grayscale of seasonal glaciers, m G is the average grayscale of the snow frequency map in the glacier zone, P k,永 is the probability that pixel k belongs to a permanent glacier, m 永 is the average grayscale of permanent glaciers.

[0151] Optionally, the snow index calculation module 502 is further used to:

[0152] For any target remote sensing image, any pixel point is selected in the target remote sensing image, and the normalized snow index of the selected pixel point is calculated according to the normalized snow index calculation formula, the green band surface reflectivity and the short-wave infrared band surface reflectivity of the selected pixel point, wherein the normalized snow index calculation formula is:

[0153]

[0154] NDSI i,jis the normalized snow cover index of the jth pixel of the i-th target remote sensing image, ρ GREEN,i,j is the green band surface reflectance of the jth pixel point of the i-th target remote sensing image, ρ MIR,i,j is the surface reflectivity of the shortwave infrared band of the jth pixel point of the i-th target remote sensing image.

[0155] Optionally, the threshold automatic determination module 505 is further used to:

[0156] Calculate the occurrence probability of various pixel values ​​in the glacier zone snow frequency map, select any pixel value, and determine seasonal glaciers and permanent glaciers in the glacier zone snow frequency map based on the comparison between other pixel values ​​and the selected pixel value;

[0157] Select a seasonal glacier or a permanent glacier as the target glacier, calculate the sum of the occurrence probabilities corresponding to various pixel values ​​in the selected target glacier, obtain the probability that the pixel belongs to the target glacier, and calculate the average grayscale of the target glacier based on the glacier average grayscale calculation formula, where the glacier average grayscale calculation formula is:

[0158]

[0159] m t,目 is the average grayscale of the target glacier when the t-th pixel value is selected, P(mesh / t) is the probability that the t-th pixel value belongs to the target glacier, P(t) is the probability of the t-th pixel value appearing in the snow frequency map of the glacier zone, and P(mesh) is the sum of the occurrence probabilities of various pixel values ​​classified as the target glacier.

[0160] Optionally, the glacier type identification module 504 is further configured to:

[0161] In the snow frequency map of the glacier zone, the initial edge of each glacier to be identified in the glacier zone to be identified is obtained based on an edge detection algorithm, and the initial edge of the glacier is morphologically expanded to obtain a smooth edge of the glacier;

[0162] In the snow frequency map of the glacier zone containing the smooth edge of the glacier, the glacier edge judgment threshold is determined based on the maximum inter-class variance method, and the fine edge of the glacier corresponding to each glacier to be identified is determined based on the pixel value of each pixel point corresponding to the smooth edge of the glacier and the glacier edge judgment threshold;

[0163] For any glacier to be identified whose fine edge has been determined in the glacier zone to be identified, the glacier type of the glacier to be identified is determined based on the pixel value of each pixel point corresponding to the glacier to be identified and the glacier type distinction threshold.

[0164] Optionally, the snow accumulation frequency calculation module 503 is further used to:

[0165] For any target remote sensing image, any pixel point is selected in the target remote sensing image, and when the normalized snow accumulation index of the selected pixel point is greater than or equal to the preset snow accumulation frequency threshold, the snow accumulation frequency of the selected pixel point is increased by one counting unit until a final counting result of the snow accumulation frequency of the selected pixel point calculated based on each target remote sensing image within a preset time period is obtained;

[0166] According to the final counting result of the snow accumulation frequency in the preset time period corresponding to each pixel point, a snow accumulation frequency map of the glacier area in the preset time period is generated.

[0167] Optionally, the glacier type identification module 504 is further configured to:

[0168] For any target remote sensing image, any pixel point is selected in the target remote sensing image, and when the normalized snow accumulation index of the selected pixel point is greater than or equal to the preset snow accumulation frequency threshold, the snow accumulation frequency of the selected pixel point is increased by one counting unit until a final counting result of the snow accumulation frequency of the selected pixel point calculated based on each target remote sensing image within a preset time period is obtained;

[0169] According to the final counting result of the snow accumulation frequency in the preset time period corresponding to each pixel point, a snow accumulation frequency map of the glacier area in the preset time period is generated.

[0170] It should be noted that for other corresponding descriptions of the functional units involved in the glacier identification device provided in the embodiment of the present application, reference can be made to Figures 1 to 3 The corresponding description in the method will not be repeated here.

[0171] Based on the above Figures 1 to 3 The method shown in the embodiment of the present application is accordingly provided with a storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned Figures 1 to 3 The glacier identification method shown.

[0172] Based on this understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each implementation scenario of the present application.

[0173] Based on the above Figures 1 to 3 The method shown, and Figure 4 , Figure 5In order to achieve the above-mentioned purpose, the embodiment of the present application further provides a computer device, which can be a personal computer, a server, a network device, etc. The computer device includes a storage medium and a processor; the storage medium is used to store a computer program; the processor is used to execute the computer program to achieve the above-mentioned Figures 1 to 3 The glacier identification method shown.

[0174] Optionally, the computer device may further include a user interface, a network interface, a camera, a radio frequency (RF) circuit, a sensor, an audio circuit, a WI-FI module, etc. The user interface may include a display, an input unit such as a keyboard, etc., and the optional user interface may also include a USB interface, a card reader interface, etc. The network interface may optionally include a standard wired interface, a wireless interface (such as a Bluetooth interface, a WI-FI interface), etc.

[0175] Those skilled in the art will appreciate that the computer device structure provided in this embodiment does not limit the computer device, and may include more or fewer components, or a combination of certain components, or different component arrangements.

[0176] The storage medium may also include an operating system and a network communication module. The operating system is a program that manages and saves the hardware and software resources of the computer device, and supports the operation of information processing programs and other software and / or programs. The network communication module is used to realize communication between the components inside the storage medium, and communication with other hardware and software in the physical device.

[0177] Through the description of the above implementation methods, the technicians in this field can clearly understand that the present application can be implemented by means of software plus necessary general hardware platforms, or can be implemented by hardware to obtain remote sensing images to be processed at different periods in the glacier zone to be identified within a preset time period; perform cloud removal on each remote sensing image to be processed to obtain the target remote sensing image, and calculate the normalized snow index of each pixel in each target remote sensing image; generate a snow frequency map of the glacier zone within a preset time period according to the normalized snow index of each pixel in each target remote sensing image and the preset snow frequency threshold; in the snow frequency map of the glacier zone, based on the maximum inter-class variance method, determine the glacier type distinction threshold corresponding to the snow frequency map of the glacier zone, and determine the glacier type of each glacier to be identified in the glacier zone to be identified based on the pixel value of each pixel in the snow frequency map of the glacier zone and the glacier type distinction threshold, wherein the glacier type includes seasonal glaciers and permanent glaciers. By automatically determining the glacier type distinction threshold, it is possible to identify whether the glacier to be identified is a seasonal glacier or a permanent glacier, thereby improving the efficiency of glacier identification.

[0178] Those skilled in the art will appreciate that the accompanying drawings are only schematic diagrams of a preferred implementation scenario, and the modules or processes in the accompanying drawings are not necessarily necessary for implementing the present application. Those skilled in the art will appreciate that the modules in the devices in the implementation scenario can be distributed in the devices of the implementation scenario according to the description of the implementation scenario, or can be changed accordingly and located in one or more devices different from the present implementation scenario. The modules of the above-mentioned implementation scenario can be combined into one module, or can be further split into multiple submodules.

[0179] The above serial numbers of this application are only for description and do not represent the advantages and disadvantages of the implementation scenarios. The above disclosure is only a few specific implementation scenarios of this application, but this application is not limited to them, and any changes that can be thought of by technicians in this field should fall within the scope of protection of this application.

Claims

1. A glacier identification method, characterized in that: The method comprises: Acquire remote sensing images to be processed at different periods of a glacier zone to be identified within a preset time period, wherein the glacier zone to be identified includes a plurality of glaciers to be identified; Performing cloud removal processing on each of the remote sensing images to be processed to obtain a target remote sensing image, and calculating a normalized snow cover index of each pixel point in each of the target remote sensing images; Generate a snow frequency map of the glacier zone within the preset time period according to the normalized snow index of each pixel point in each target remote sensing image and a preset snow frequency threshold; The inter-class variance calculation formula is used to calculate the inter-class variance corresponding to each pixel in the snow frequency map of the glacier zone, and the pixel value of the pixel corresponding to the maximum inter-class variance is determined as the glacier type distinction threshold, wherein the inter-class variance calculation formula is: ， is the inter-class variance of the k-th pixel in the snow frequency map of the glacier zone, is the probability that pixel k belongs to a seasonal glacier, is the average grayscale of seasonal glaciers, is the average grayscale of the snow frequency map in the glacier zone, is the probability that pixel k belongs to a permanent glacier, is the average grayscale of permanent glaciers; Based on the pixel value of each pixel point in the snow frequency map of the glacier zone and the glacier type distinction threshold, the glacier type of each glacier to be identified in the glacier zone to be identified is determined, wherein the glacier type includes seasonal glaciers and permanent glaciers.

2. The method according to claim 1, characterized in that The step of calculating the normalized snow cover index of each pixel in each target remote sensing image comprises: For any target remote sensing image, any pixel point is selected in the target remote sensing image, and the normalized snow index of the selected pixel point is calculated according to the normalized snow index calculation formula, the green band surface reflectivity and the short-wave infrared band surface reflectivity of the selected pixel point, wherein the normalized snow index calculation formula is: ， is the normalized snow cover index of the jth pixel of the i-th target remote sensing image, is the green band surface reflectance of the jth pixel point of the i-th target remote sensing image, is the surface reflectivity of the shortwave infrared band of the jth pixel point of the i-th target remote sensing image.

3. The method according to claim 1, characterized in that Before calculating the inter-class variance corresponding to each pixel point in the snow frequency map of the glacier zone according to the inter-class variance calculation formula, the method further includes: Calculate the occurrence probability of various pixel values ​​in the glacier zone snow frequency map, select any pixel value, and determine seasonal glaciers and permanent glaciers in the glacier zone snow frequency map based on the comparison between other pixel values ​​and the selected pixel value; Select a seasonal glacier or a permanent glacier as the target glacier, calculate the sum of the occurrence probabilities corresponding to various pixel values ​​in the selected target glacier, obtain the probability that the pixel belongs to the target glacier, and calculate the average grayscale of the target glacier based on the glacier average grayscale calculation formula, where the glacier average grayscale calculation formula is: ， is the average grayscale of the target glacier when the t-th pixel value is selected, is the probability that the t-th pixel value belongs to the target glacier, is the probability of occurrence of the t-th pixel value in the snow frequency map of the glacier zone, is the sum of the occurrence probabilities of various pixel values ​​classified as the target glacier.

4. The method according to claim 1, characterized in that: The step of determining the glacier type of each glacier to be identified in the glacier zone to be identified based on the pixel value of each pixel point in the snow frequency map of the glacier zone and the glacier type distinction threshold comprises: In the snow frequency map of the glacier zone, the initial edge of each glacier to be identified in the glacier zone to be identified is obtained based on an edge detection algorithm, and the initial edge of the glacier is morphologically expanded to obtain a smooth edge of the glacier; In the snow frequency map of the glacier zone containing the smooth edge of the glacier, the glacier edge judgment threshold is determined based on the maximum inter-class variance method, and the fine edge of the glacier corresponding to each glacier to be identified is determined based on the pixel value of each pixel point corresponding to the smooth edge of the glacier and the glacier edge judgment threshold; For any glacier to be identified whose fine edge has been determined in the glacier zone to be identified, the glacier type of the glacier to be identified is determined based on the pixel value of each pixel point corresponding to the glacier to be identified and the glacier type distinction threshold.

5. The method according to claim 1, characterized in that: The generating of the snow frequency map of the glacier zone within the preset time period according to the normalized snow index of each pixel point in each target remote sensing image and the preset snow frequency threshold comprises: For any target remote sensing image, any pixel point is selected in the target remote sensing image, and when the normalized snow accumulation index of the selected pixel point is greater than or equal to the preset snow accumulation frequency threshold, the snow accumulation frequency of the selected pixel point is increased by one counting unit until a final counting result of the snow accumulation frequency of the selected pixel point calculated based on each target remote sensing image within a preset time period is obtained; According to the final counting result of the snow accumulation frequency in the preset time period corresponding to each pixel point, a snow accumulation frequency map of the glacier area in the preset time period is generated.

6. The method according to any one of claims 1 to 5, characterized in that The step of determining the glacier type of each glacier to be identified in the glacier zone to be identified based on the pixel value of each pixel point in the snow frequency map of the glacier zone and the glacier type distinction threshold comprises: Determine the seasonal glacier pixel assignment values ​​for seasonal glaciers and the permanent glacier pixel assignment values ​​for permanent glaciers; In the snow frequency map of the glacier zone, for any pixel point corresponding to the glacier to be identified, when the pixel value of the pixel point is greater than or equal to the glacier type distinction threshold, the pixel value of the pixel point is assigned a seasonal glacier pixel assignment value; when the pixel value of the pixel point is less than the glacier type distinction threshold, the pixel value of the pixel point is assigned a permanent glacier pixel assignment value; Seasonal glaciers are determined based on the displayed area where values ​​are assigned to seasonal glacier pixels, and permanent glaciers are determined based on the displayed area where values ​​are assigned to permanent glacier pixels.

7. A glacier identification device, characterized in that: The device comprises: A remote sensing image acquisition module, used to acquire remote sensing images to be processed at different periods of a glacier zone to be identified within a preset time period, wherein the glacier zone to be identified includes a plurality of glaciers to be identified; A snow index calculation module is used to perform cloud removal processing on each of the remote sensing images to be processed to obtain a target remote sensing image, and calculate a normalized snow index of each pixel in each of the target remote sensing images; A snow frequency calculation module, used to generate a snow frequency map of the glacier zone within the preset time period according to the normalized snow index of each pixel point in each target remote sensing image and a preset snow frequency threshold; The glacier type identification module is used to calculate the inter-class variance corresponding to each pixel point in the snow frequency map of the glacier zone according to the inter-class variance calculation formula, determine the pixel value of the pixel point corresponding to the maximum inter-class variance as the glacier type distinction threshold, and determine the glacier type of each glacier to be identified in the glacier zone to be identified based on the pixel value of each pixel point in the snow frequency map of the glacier zone and the glacier type distinction threshold, wherein the glacier type includes seasonal glaciers and permanent glaciers, and the inter-class variance calculation formula is: ， is the inter-class variance of the k-th pixel in the snow frequency map of the glacier zone, is the probability that pixel k belongs to a seasonal glacier, is the average grayscale of seasonal glaciers, is the average grayscale of the snow frequency map in the glacier zone, is the probability that pixel k belongs to a permanent glacier, is the average grayscale of permanent glaciers.

8. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for identifying glaciers according to any one of claims 1 to 6 is implemented.

9. A computer device comprising a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, characterized in that: When the processor executes the computer program, the method for identifying glaciers according to any one of claims 1 to 6 is implemented.

Images