Comprehensive satellite remote sensing image analysis system and method based on deep learning

Abstract

The invention discloses a comprehensive satellite remote sensing image analysis system and method based on deep learning, and belongs to the technical field of remote sensing image analysis, and the system comprises an image acquisition and preprocessing module, a water storage detection module, a surface floating object detection module and an overall state evaluation module. According to the method, the visible light remote sensing image of the current lake is obtained through the satellite and is preprocessed to obtain the preprocessed visible light remote sensing image, and the water storage capacity is obtained based on the preprocessed visible light remote sensing image; three surface floating object state parameters including the floating object local area ratio, the floating object local number and the floating object local dispersion degree are obtained, the corresponding surface floating object state of each subarea of the lake can be accurately represented, subsequent overall state evaluation work is carried out in combination with the water storage capacity, and the evaluation result is more accurate.

Description

Technical Field

[0001] The present invention relates to the technical field of remote sensing image analysis, and in particular to a comprehensive satellite remote sensing image analysis system and method based on deep learning. Background Art

[0002] Satellite remote sensing images are mainly taken or recorded by satellites, capturing different features of the Earth's surface and its reflection or radiation information in different spectral bands. With the continuous advancement of remote sensing technology and the widespread application of satellites, the acquisition of satellite remote sensing images will become more convenient and efficient.

[0003] Satellite remote sensing images can be used to identify a variety of objects and phenomena on the earth's surface. For example, through deep learning and machine learning algorithms, natural objects, man-made objects, special objects or phenomena can be monitored. Natural objects include water bodies, vegetation, soil, and topography. Man-made objects include buildings, roads and transportation facilities, vehicles, and agricultural facilities. Special objects or phenomena include disasters, mineral resources, and climate change. In summary, satellite remote sensing imaging technology has broad application prospects and important value. It can detect and identify a variety of common objects and phenomena, providing vital data support for scientific research, policy making, and business decision-making.

[0004] The existing method of evaluating the overall status of a lake is mainly to obtain relevant parameters through manual detection, and then obtain the lake status monitoring results based on the relevant parameters and manual experience. The above method has problems such as high labor costs and low monitoring efficiency.

[0005] The above problems need to be solved urgently. To this end, an integrated satellite remote sensing image analysis system based on deep learning is proposed. Summary of the invention

[0006] The technical problem to be solved by the present invention is: how to solve the problems of high labor cost and low monitoring efficiency in the existing methods of evaluating the overall status of lakes, and provide a comprehensive satellite remote sensing image analysis system based on deep learning.

[0007] The present invention solves the above technical problems through the following technical solutions, which include an image acquisition and preprocessing module, a water storage capacity detection module, a surface floating object detection module and an overall state assessment module;

[0008] The image acquisition and preprocessing module is used to acquire the visible light remote sensing image of the current lake through a satellite and preprocess it to obtain the preprocessed visible light remote sensing image;

[0009] The water storage capacity detection module is used to detect the water storage capacity of the current lake based on the pre-processed visible light remote sensing image to obtain the water storage capacity A;

[0010] The surface floating object detection module is used to detect the state parameters of surface floating objects in each sub-area of ​​the current lake after preprocessing the visible light remote sensing image, and obtain the local area ratio S of the floating objects. i 、Local number of floating objects C i Local dispersion of floating objects F i , where the subscript i represents the i-th sub-region;

[0011] The overall state evaluation module is used to calculate the overall state score of the lake based on the acquired surface floating object state parameters and water storage capacity of each sub-area of ​​the current lake.

[0012] Furthermore, the image acquisition and preprocessing module includes a satellite shooting unit and a preprocessing unit. The satellite shooting unit is used to capture the visible light remote sensing image of the current lake on the satellite; the preprocessing unit is used to perform noise reduction and image enhancement processing on the visible light remote sensing image to obtain the preprocessed visible light remote sensing image.

[0013] Furthermore, the water storage capacity detection module includes a target detection unit, a contour detection unit and a water storage capacity acquisition unit; the target detection unit is used to use the trained lake target detection model to detect the lake target in the preprocessed visible light remote sensing image to obtain a lake target detection frame; the contour detection unit is used to use the contour detection function in OpenCV to perform outer contour detection on the lake area in the lake target detection frame, obtain the coordinates of each point on the outer contour line of the lake area in the image, and calculate the area of ​​the lake area in the image according to the coordinates of each point in the image, recorded as Sp; the water storage capacity acquisition unit is used to search and compare in a preset area-water level height database according to the area Sp of the lake area in the image, obtain the water level height Hp corresponding to the area Sp of the lake area in the image, and then search and compare in a preset water level height-water storage capacity database according to the water level height Hp, and obtain the water storage capacity corresponding to the water level height Hp, that is, obtain the water storage capacity A of the current lake.

[0014] Furthermore, the surface floating object detection module includes a sub-area division unit, an area ratio calculation unit, a quantity calculation unit and a dispersion calculation unit; the sub-area division unit is used to divide the current lake area into multiple sub-areas; the area ratio calculation unit is used to calculate the local area ratio S of the floating objects in each sub-area according to the divided sub-areas. i The number calculation unit is used to count and record the local number of floating objects in each sub-area according to the divided sub-areas C i The dispersion calculation unit is used to calculate the local dispersion F of floating objects in each sub-area according to the divided sub-areas. i .

[0015] Furthermore, the specific processing process of the surface floating object detection module is as follows:

[0016] Step S31: Based on the coordinates of each point on the outer contour line of the lake area obtained by the contour detection unit, OpenCV is used to calculate the geometric center point Z of the lake area, and a straight line L is drawn through the geometric center point Z in the direction parallel to the long side of the lake target detection frame. z , intersecting the two short sides of the lake target detection frame at points D1 and D2 respectively, calculating the midpoint of line segment ZD1, recorded as point D3, and calculating the midpoint of line segment ZD2, recorded as point D4. Points D1, D3, Z, D4, and D2 are arranged on line segment D1D2 in sequence, and straight lines H are drawn through points D3, Z, and D4 in the direction parallel to the short side of the lake target detection frame. D3 , H Z , H D4 , and record the straight line H D3 , H Z , H D4 The coordinates of the intersection with the outer contour of the lake area, and then divide the current lake area into four sub-areas;

[0017] Step S32: Calculate the local area ratio S of the floating objects in each sub-area according to the divided sub-areas. i , the calculation formula is as follows:

[0018] S i =Pz i / P i ;

[0019] Among them, Pz i represents the sum of the areas of all floating objects in the i-th sub-region in the image, P i represents the area of ​​the i-th sub-region in the image;

[0020] Step S33: According to the divided sub-areas, count and record the local number of floating objects C in each sub-area i ;

[0021] Step S34: Calculate the local dispersion of floating objects in each sub-area according to the divided sub-areas. i , the calculation formula is as follows:

[0022] F i =(Q1+Q2+……Q j-1 +Q j );

[0023] Among them, Q j It represents the distance between the geometric center point of the jth floating object in the i-th sub-region and the geometric center point of the sub-region in the image.

[0024] Furthermore, in steps S32 to S34, target detection is performed on floating objects in the sub-area using the trained floating object detection model.

[0025] Furthermore, in the overall status assessment module, the calculation formula of the overall status score T of the lake is as follows:

[0026] T=S′ wavg +C′ wavg +F′ wavg +A′;

[0027] Among them, S′ wavg is the local area ratio of floating objects in each sub-area S i The weighted average S wavg The result obtained after dimensionless processing; C′ wavg is the local number of floating objects in each sub-area C i The weighted average C wavg The result obtained after dimensionless processing; F′ wavg is the local dispersion of floating objects in each sub-area F i The weighted average F wavg A′ is the result obtained after dimensionless processing; A′ is the result obtained after dimensionless processing of water storage capacity A.

[0028] Furthermore, in the overall lake status score T:

[0029] Weighted average S wavg The calculation formula is as follows:

[0030] S wavg =w1*S1+w2*S2+w3*S3+w4*S4;

[0031] Weighted average C wavg The calculation formula is as follows:

[0032] C wavg =w1*C1+w2*C2+w3*C3+w4*C4;

[0033] Weighted average F wavg The calculation formula is as follows:

[0034] F wavg =w1*F1+w2*F2+w3*F3+w4*F4;

[0035] Among them, w1, w2, w3, and w4 are the weights of the corresponding sub-regions, which are positively correlated with the distance between the geometric center point of the corresponding sub-region and the geometric center point of the lake area. That is, the farther the geometric center point of the sub-region is from the geometric center point of the lake area, the greater the corresponding weight.

[0036] The present invention also provides a comprehensive satellite remote sensing image analysis method based on deep learning, which uses the above-mentioned image analysis system to analyze satellite remote sensing images and then evaluate the overall state of the lake, including the following steps:

[0037] Step S1: acquiring a visible light remote sensing image of the current lake through a satellite and preprocessing the image to obtain a preprocessed visible light remote sensing image;

[0038] Step S2: Based on the pre-processed visible light remote sensing image, the water storage capacity of the current lake is detected to obtain the water storage capacity;

[0039] Step S3: dividing the lake area into sub-areas, and then detecting the state parameters of surface floating objects in each sub-area of ​​the current lake;

[0040] Step S4: Calculate the overall lake status score based on the acquired surface floating object status parameters and water storage capacity of each sub-area of ​​the current lake.

[0041] Compared with the prior art, the present invention has the following advantages: the integrated satellite remote sensing image analysis system and method based on deep learning obtains the visible light remote sensing image of the current lake through the satellite and preprocesses it to obtain the preprocessed visible light remote sensing image, obtains the water storage capacity based on the preprocessed visible light remote sensing image, and then obtains three surface floating object state parameters, namely the local area ratio of floating objects, the local number of floating objects and the local dispersion of floating objects. It can accurately characterize the surface floating object state of each sub-area of ​​the corresponding lake, and conduct subsequent overall state evaluation in combination with the water storage capacity, so that the evaluation result is more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 is a structural schematic diagram of a comprehensive satellite remote sensing image analysis system based on deep learning in an embodiment of the present invention;

[0043] Figure 2 Schematic diagram of the sub-region division in the embodiment of the present invention (taking a near-sighted elliptical lake as an example);

[0044] Figure 3 It is a flowchart of a comprehensive satellite remote sensing image analysis method based on deep learning in an embodiment of the present invention. DETAILED DESCRIPTION

[0045] The following is a detailed description of an embodiment of the present invention. This embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation method and a specific operation process are given, but the protection scope of the present invention is not limited to the following embodiment.

[0046] like Figure 1As shown, this embodiment provides a technical solution: an integrated satellite remote sensing image analysis system based on deep learning, including an image acquisition and preprocessing module, a water storage capacity detection module, a surface floating object detection module and an overall state assessment module;

[0047] The image acquisition and preprocessing module is used to acquire the visible light remote sensing image of the current lake through a satellite and preprocess it to obtain the preprocessed visible light remote sensing image;

[0048] In this embodiment, the image acquisition and preprocessing module includes a satellite shooting unit and a preprocessing unit. The satellite shooting unit is used to capture the visible light remote sensing image of the current lake on the satellite and send it to the preprocessing unit for final preprocessing; the preprocessing unit is used to perform noise reduction and image enhancement processing on the visible light remote sensing image to obtain the preprocessed visible light remote sensing image.

[0049] More specifically, the satellite shooting unit is an ultra-high-definition camera installed on a satellite, which can obtain ultra-high-definition visible light images. In this embodiment, the noise reduction method includes but is not limited to Gaussian filtering noise reduction, median filtering noise reduction, mean filtering noise reduction, etc., which are used to significantly reduce the noise of the image, and the image enhancement method includes but is not limited to histogram equalization, grayscale transformation, etc., which are used to improve the quality of the image.

[0050] The water storage capacity detection module is used to detect the water storage capacity of the current lake and obtain the water storage capacity A.

[0051] In this embodiment, the water storage capacity detection module includes a target detection unit, a contour detection unit and a water storage capacity acquisition unit; the target detection unit is used to use the trained lake target detection model to detect the lake target in the preprocessed visible light remote sensing image to obtain a lake target detection frame; the contour detection unit is used to use the contour detection function in OpenCV to perform outer contour detection on the lake area in the lake target detection frame, obtain the coordinates of each point on the outer contour line of the lake area in the image, and calculate the area of ​​the lake area in the image according to the coordinates of each point in the image, recorded as Sp; the water storage capacity acquisition unit is used to search and compare in a preset area-water level height database according to the area Sp of the lake area in the image, obtain the water level height Hp corresponding to the area Sp of the lake area in the image, and then search and compare in a preset water level height-water storage capacity database according to the water level height Hp to obtain the water storage capacity corresponding to the water level height Hp, that is, obtain the water storage capacity A of the current lake. The present invention associates the area of ​​the lake region in the image with the water level and establishes a corresponding database, so that the water level can be obtained conveniently and quickly, and then the correlation between the water level and the water storage capacity is used to accurately and quickly obtain the current water storage capacity of the lake.

[0052] To be more specific, the lake object detection model is trained based on the SSD network (deep learning network).

[0053] To be more specific, in the water storage capacity acquisition unit, a mapping relationship between the area of ​​the lake area in the image and the corresponding water level height is preset in the area-water level height database. The mapping relationship is obtained by obtaining a large amount of area data of the lake area in the image and manually measuring the water level height data under the corresponding area data in advance, and finally the mapping relationship is obtained by mathematical fitting.

[0054] As more specific, in the water storage acquisition unit, a mapping relationship between the water level height and the corresponding water storage capacity is preset in the water level height-water storage capacity database. The mapping relationship is obtained by simulating the water storage capacity in the early stage. When simulating the water storage capacity in the early stage, the elevation data of the terrain surface of the lake area is obtained by using a digital elevation model. By analyzing the elevation data, the boundary and bottom area of ​​the lake can be determined, and the water storage capacity can be calculated in combination with the water level height. The corresponding mapping relationship can be established by the calculated water storage capacity and the corresponding water level height.

[0055] The surface floating object detection module is used to detect the state parameters of the surface floating objects in each sub-area of ​​the current lake, and obtain the local area ratio S of the floating objects. i 、Local number of floating objects C i Local dispersion of floating objects F i , wherein the subscript i represents the i-th sub-region; in the present invention, through the three surface floating object state parameters of local floating object area ratio, local floating object number and local floating object dispersion, the surface floating object state of each sub-region of the corresponding lake can be accurately characterized, which is conducive to the subsequent overall state evaluation work and makes the evaluation result more accurate.

[0056] In this embodiment, the surface floating object detection module includes a sub-area division unit, an area ratio calculation unit, a quantity calculation unit and a dispersion calculation unit; the sub-area division unit is used to divide the current lake area into multiple sub-areas; the area ratio calculation unit is used to calculate the local area ratio S of the floating objects in each sub-area according to the divided sub-areas. i The number calculation unit is used to count and record the local number of floating objects in each sub-area according to the divided sub-areas C i The dispersion calculation unit is used to calculate the local dispersion F of floating objects in each sub-area according to the divided sub-areas. i .

[0057] To be more specific, when OpenCV is used to calculate the geometric center point Z of the lake area, the concept of polygon centroid is used to calculate the center point; the centroid can be obtained by calculating the weighted average position of all vertices, and then the geometric center point Z is calculated.

[0058] As more specific, the specific processing process of the surface floating object detection module is as follows:

[0059] Step S31: Based on the coordinates of each point on the outer contour line of the lake area obtained by the contour detection unit, OpenCV is used to calculate the geometric center point Z of the lake area, and a straight line L is drawn through the geometric center point Z in the direction parallel to the long side of the lake target detection frame. z , intersecting the two short sides of the lake target detection frame at points D1 and D2 respectively, calculating the midpoint of line segment ZD1, recorded as point D3, and calculating the midpoint of line segment ZD2, recorded as point D4. Points D1, D3, Z, D4, and D2 are arranged on line segment D1D2 in sequence, and straight lines H are drawn through points D3, Z, and D4 in the direction parallel to the short side of the lake target detection frame. D3 , H Z , H D4 , and record the straight line H D3 , H Z , H D4 The coordinates of the intersection with the outer contour of the lake area are used to divide the current lake area into four sub-areas, such as Figure 2 As shown;

[0060] Step S32: Calculate the local area ratio S of the floating objects in each sub-area according to the divided sub-areas. i , the calculation formula is as follows:

[0061] S i =Pz i / P i ;

[0062] Among them, Pz i P represents the sum of the areas of all floating objects in the i-th sub-region in the image. The calculation method is the same as the calculation principle of the area Sp of the lake area in the image. First, the floating objects in the sub-region are detected by the trained floating object detection model, and then the outer contour detection is performed. Finally, the sum of the areas of all floating objects in the i-th sub-region in the image is calculated; P i represents the area of ​​the i-th sub-region in the image;

[0063] It should be noted that the floating object detection model is trained based on the YOLO v3 network;

[0064] Step S33: According to the divided sub-areas, count and record the local number of floating objects C in each sub-areai ,The number of floating objects is obtained by using the floating object detection model to detect the floating objects in the sub-area;

[0065] Step S34: Calculate the local dispersion of floating objects in each sub-area according to the divided sub-areas. i , the calculation formula is as follows:

[0066] F i =(Q1+Q2+……Q j-1 +Q j );

[0067] Among them, Q j represents the distance between the geometric center point of the jth floating object in the i-th sub-region and the geometric center point of the sub-region in the image, F i The larger the value, the more dispersed the floating objects in the sub-area are. The calculation method of the floating object geometric center point and the sub-area geometric center point refers to the calculation method of the lake area geometric center point, which will not be repeated here.

[0068] The overall state evaluation module is used to calculate the overall state score of the lake based on the acquired surface floating object state parameters and water storage capacity of each sub-area of ​​the current lake;

[0069] In this embodiment, the calculation formula of the overall lake status score T is as follows:

[0070] T=S′ wavg +C′ wavg +F′ wavg +A′;

[0071] Among them, S′ wavg is the local area ratio of floating objects in each sub-area S i The weighted average S wavg The result obtained after dimensionless processing; C′ wavg is the local number of floating objects in each sub-area C i The weighted average C wavg The result obtained after dimensionless processing; F′ wavg is the local dispersion of floating objects in each sub-area F i The weighted average F wavg A′ is the result obtained after dimensionless processing; A′ is the result obtained after dimensionless processing of water storage capacity A.

[0072] It should be noted that the dimensionless processing in this embodiment is a normalization processing.

[0073] As more specific:

[0074] Weighted average S wavg The calculation formula is as follows:

[0075] S wavg =w1*S1+w2*S2+w3*S3+w4*S4;

[0076] Weighted average C wavg The calculation formula is as follows:

[0077] C wavg =w1*C1+w2*C2+w3*C3+w4*C4;

[0078] Weighted average F wavg The calculation formula is as follows:

[0079] F wavg =w1*F1+w2*F2+w3*F3+w4*F4;

[0080] Among them, w1, w2, w3, and w4 are the weights of the corresponding sub-regions, which are positively correlated with the distance between the geometric center points of the corresponding sub-regions and the geometric center points of the lake area. That is, the farther the geometric center point of the sub-region is from the geometric center point of the lake area, the greater the corresponding weight, which is conducive to calculating a more accurate overall lake status score.

[0081] like Figure 3 As shown, this embodiment also provides a comprehensive satellite remote sensing image analysis method based on deep learning, which uses the above-mentioned image analysis system to analyze satellite remote sensing images and then evaluate the overall state of the lake, including the following steps:

[0082] Step S1: acquiring a visible light remote sensing image of the current lake through a satellite and preprocessing the image to obtain a preprocessed visible light remote sensing image;

[0083] Step S2: Based on the pre-processed visible light remote sensing image, the water storage capacity of the current lake is detected to obtain the water storage capacity;

[0084] Step S3: dividing the lake area into sub-areas, and then detecting the state parameters of surface floating objects in each sub-area of ​​the current lake;

[0085] Step S4: Calculate the overall lake status score based on the acquired surface floating object status parameters and water storage capacity of each sub-area of ​​the current lake.

[0086] It should be noted that, for the specific implementation process of the above steps S1 to S4, please refer to the processing process of each module and unit in the above image analysis system, which will not be repeated here.

[0087] To summarize, the integrated satellite remote sensing image analysis system and method based on deep learning in the above-mentioned embodiment obtains the visible light remote sensing image of the current lake through the satellite and preprocesses it to obtain the preprocessed visible light remote sensing image, obtains the water storage capacity based on the preprocessed visible light remote sensing image, and then obtains the three surface floating object state parameters of the local area ratio of floating objects, the local number of floating objects and the local dispersion of floating objects, which can accurately characterize the surface floating object state of each sub-area of ​​the corresponding lake, and conduct subsequent overall state assessment in combination with the water storage capacity, so that the assessment result is more accurate.

[0088] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A comprehensive satellite remote sensing image analysis system based on deep learning, characterized in that: include: Image acquisition and preprocessing module, water storage detection module, surface floating object detection module and overall status assessment module; The image acquisition and preprocessing module is used to acquire the visible light remote sensing image of the current lake through a satellite and preprocess it to obtain the preprocessed visible light remote sensing image; The water storage capacity detection module is used to detect the water storage capacity of the current lake based on the pre-processed visible light remote sensing image to obtain the water storage capacity A; The surface floating object detection module is used to detect the state parameters of surface floating objects in each sub-area of ​​the current lake after preprocessing the visible light remote sensing image, and obtain the local area ratio S of the floating objects. i 、Local number of floating objects C i Local dispersion of floating objects F i , where the subscript i represents the i-th sub-region; The overall state evaluation module is used to calculate the overall state score of the lake based on the acquired surface floating object state parameters and water storage capacity of each sub-area of ​​the current lake.

2. The integrated satellite remote sensing image analysis system based on deep learning according to claim 1, characterized in that: The image acquisition and preprocessing module includes a satellite shooting unit and a preprocessing unit. The satellite shooting unit is used to shoot and obtain a visible light remote sensing image of the current lake on a satellite; the preprocessing unit is used to perform noise reduction and image enhancement processing on the visible light remote sensing image to obtain a preprocessed visible light remote sensing image.

3. The integrated satellite remote sensing image analysis system based on deep learning according to claim 2 is characterized in that: The water storage capacity detection module includes a target detection unit, a contour detection unit and a water storage capacity acquisition unit; the target detection unit is used to detect the lake target in the preprocessed visible light remote sensing image using the trained lake target detection model to obtain a lake target detection frame; the contour detection unit is used to use the contour detection function in OpenCV to perform outer contour detection on the lake area in the lake target detection frame, obtain the coordinates of each point on the outer contour line of the lake area in the image, and calculate the area of ​​the lake area in the image according to the coordinates of each point in the image, which is recorded as Sp; the water storage capacity acquisition unit is used to search and compare in a preset area-water level height database according to the area Sp of the lake area in the image, obtain the water level height Hp corresponding to the area Sp of the lake area in the image, and then search and compare in a preset water level height-water storage capacity database according to the water level height Hp to obtain the water storage capacity corresponding to the water level height Hp, that is, obtain the water storage capacity A of the current lake.

4. The integrated satellite remote sensing image analysis system based on deep learning according to claim 3 is characterized in that: The surface floating object detection module includes a sub-area division unit, an area ratio calculation unit, a quantity calculation unit and a dispersion calculation unit; the sub-area division unit is used to divide the current lake area into multiple sub-areas; the area ratio calculation unit is used to calculate the local area ratio S of floating objects in each sub-area according to the divided sub-areas. i The number calculation unit is used to count and record the local number of floating objects in each sub-area according to the divided sub-areas C i The dispersion calculation unit is used to calculate the local dispersion F of floating objects in each sub-area according to the divided sub-areas. i .

5. The integrated satellite remote sensing image analysis system based on deep learning according to claim 4 is characterized in that: The specific processing process of the surface floating object detection module is as follows: Step S31: Based on the coordinates of each point on the outer contour line of the lake area obtained by the contour detection unit, OpenCV is used to calculate the geometric center point Z of the lake area, and a straight line L is drawn through the geometric center point Z in the direction parallel to the long side of the lake target detection frame. z , intersecting the two short sides of the lake target detection frame at points D1 and D2 respectively, calculating the midpoint of line segment ZD1, recorded as point D3, and calculating the midpoint of line segment ZD2, recorded as point D4. Points D1, D3, Z, D4, and D2 are arranged on line segment D1D2 in sequence, and straight lines H are drawn through points D3, Z, and D4 in the direction parallel to the short side of the lake target detection frame. D3 , H Z , H D4 , and record the straight line H D3 , H Z , H D4 The coordinates of the intersection with the outer contour of the lake area, and then divide the current lake area into four sub-areas; Step S32: Calculate the local area ratio S of the floating objects in each sub-area according to the divided sub-areas. i , the calculation formula is as follows: S i =Pz i / P i ; Among them, Pz i represents the sum of the areas of all floating objects in the i-th sub-region in the image, P i represents the area of ​​the i-th sub-region in the image; Step S33: According to the divided sub-areas, count and record the local number of floating objects C in each sub-area i ; Step S34: Calculate the local dispersion of floating objects in each sub-area according to the divided sub-areas. i , the calculation formula is as follows: F i =(Q1+Q2+……Q j-1 +Q j ); Among them, Q j Represents the distance between the geometric center point of the jth floating object in the ith sub-region and the geometric center point of the sub-region in the image.

6. The integrated satellite remote sensing image analysis system based on deep learning according to claim 5, characterized in that: In steps S32 to S34, the floating objects in the sub-area are detected using the trained floating object detection model.

7. The integrated satellite remote sensing image analysis system based on deep learning according to claim 5, characterized in that: In the overall status assessment module, the calculation formula of the overall status score T of the lake is as follows: T=S′ wavg +C′ wavg +F′ wavg +A′; Among them, S′ wavg is the local area ratio of floating objects in each sub-area S i The weighted average S wavg The result obtained after dimensionless processing; C′ wavg is the local number of floating objects in each sub-area C i The weighted average C wavg The result obtained after dimensionless processing; F′ wavg is the local dispersion of floating objects in each sub-area F i The weighted average F wavg A′ is the result obtained after dimensionless processing; A′ is the result obtained after dimensionless processing of water storage capacity A.

8. The integrated satellite remote sensing image analysis system based on deep learning according to claim 7, characterized in that: In the overall lake condition score T: Weighted average S wavg The calculation formula is as follows: <h2 style=";text-align:left;direction:ltr">S<h2 style=";text-align:left;direction:ltr"> wavg <h2 style=";text-align:left;direction:ltr"> (w1*S1+w2*S2+w3*S3+w4*S4) Weighted average C wavg The calculation formula is as follows: C wavg =w1*C1+w2*C2+w3*C3+w4*C4; Weighted average F wavg The calculation formula is as follows: <h2 style=";text-align:left;direction:ltr">F<h2 style=";text-align:left;direction:ltr"> wavg <h2 style=";text-align:left;direction:ltr"> (w1*F1+w2*F2+w3*F3+w4*F4) Among them, w1, w2, w3, and w4 are the weights of the corresponding sub-regions, which are positively correlated with the distance between the geometric center point of the corresponding sub-region and the geometric center point of the lake area. That is, the farther the geometric center point of the sub-region is from the geometric center point of the lake area, the greater the corresponding weight.

9. A comprehensive satellite remote sensing image analysis method based on deep learning, characterized in that: Analyzing satellite remote sensing images using the image analysis system according to any one of claims 1 to 8, and then evaluating the overall state of the lake, comprises the following steps: Step S1: acquiring a visible light remote sensing image of the current lake through a satellite and preprocessing the image to obtain a preprocessed visible light remote sensing image; Step S2: Based on the pre-processed visible light remote sensing image, the water storage capacity of the current lake is detected to obtain the water storage capacity; Step S3: dividing the lake area into sub-areas, and then detecting the state parameters of surface floating objects in each sub-area of ​​the current lake; Step S4: Calculate the overall lake status score based on the acquired surface floating object status parameters and water storage capacity of each sub-area of ​​the current lake.

Images