Sea grass distribution map drawing method and device

Abstract

The invention relates to the technical field of satellite remote sensing images, and discloses a sea grass distribution map drawing method and device, and the method comprises the steps: obtaining a time sequence remote sensing image of a to-be-recognized region, selecting a sea-land separation boundary based on the water body index of each satellite remote sensing image, and constructing a time sequence ocean region remote sensing image; acquiring a plurality of homogeneous image spots in each ocean region remote sensing image by using a multi-scale segmentation method, and classifying the homogeneous image spots; acquiring an area formed by all homogeneous image spots belonging to the type of green plants in ocean water as a green plant area remote sensing image; based on the red light wave band reflection value of the remote sensing image of each scene green plant area and the first vegetation red edge wave band reflection value, calculating a submerged sea grass spectral index to mark suspected sea grass pixel points; calculating an amplitude deviation index of each suspected sea grass pixel point, and selecting a target sea grass pixel point; and extracting target sea grass pixel points of all green plant area remote sensing images in the time sequence green plant area remote sensing images, and drawing a sea grass distribution map.

Description

Technical Field

[0001] The invention relates to the technical field of satellite remote sensing images, and in particular to a method and device for drawing a seaweed distribution map. Background Art

[0002] Seagrass is a higher aquatic plant that grows in shallow ocean waters. It has high ecological value and is a key component of the marine blue carbon ecosystem. However, seagrass has been declining rapidly around the world in recent years, so it is of great significance to protect seagrass and its ecological functions.

[0003] Seagrass surveys are the prerequisite for implementing seagrass protection. The current conventional methods for seagrass surveys are field surveys and remote sensing technology surveys. Based on field observations, this method has defects such as long operation cycles, high implementation costs, and difficulty in detecting unknown sea areas. Satellite remote sensing technology not only has the ability to make large-scale statistical inferences quickly and conveniently, but is also cheaper. Using remote sensing technology to survey seagrass can quickly, conveniently, efficiently, and cheaply achieve large-scale seagrass distribution mapping, which has significant advantages over traditional field observation survey methods.

[0004] However, in the current process of using remote sensing technology to survey seagrass, since seagrass and seaweed floating on the surface of ocean water both contain chlorophyll, both appear as plant spectral characteristics in remote sensing images. Therefore, it is difficult to accurately distinguish seagrass from seaweed, resulting in low accuracy in seagrass surveys and low precision in drawing seagrass distribution maps. Summary of the invention

[0005] To this end, the technical problem to be solved by the present invention is to overcome the problem that the prior art cannot effectively distinguish seagrass from floating algae when drawing seagrass distribution maps using remote sensing technology, resulting in inaccurate identification of seagrass distribution areas and low accuracy in drawing seagrass distribution maps.

[0006] In order to solve the above technical problems, the present invention provides a method for drawing a seaweed distribution map, comprising:

[0007] Obtain satellite remote sensing images of the area to be identified at each sampling time within a preset time period to form a time series remote sensing image of the area to be identified;

[0008] Based on the water body index of each satellite remote sensing image in the time series remote sensing image, the land-sea separation boundary is selected, and the ocean area in each satellite remote sensing image is extracted to form a time series ocean area remote sensing image;

[0009] For each ocean area remote sensing image in the time series ocean area remote sensing image, a multi-scale segmentation method is used to perform image segmentation to obtain multiple homogeneous image spots in each ocean area remote sensing image;

[0010] Based on the characteristic parameters of the homogeneous image spots, the categories of each homogeneous image spot in each ocean area remote sensing image are obtained, wherein the categories include marine water green plant types, marine water body types and benthic types;

[0011] Acquire the area composed of all homogeneous image spots belonging to the type of green plants in the ocean water in each ocean area remote sensing image as the green plant area remote sensing image corresponding to each ocean area remote sensing image, and form a time series green plant area remote sensing image;

[0012] For each scene of the green plant area remote sensing images in the time series of green plant area remote sensing images, the submerged seaweed spectral index of each scene of the green plant area remote sensing image is calculated based on the red light band reflectance value of each scene of the green plant area remote sensing image and the first vegetation red edge band reflectance value, and compared with the preset spectral index range, the suspected seaweed pixel points are marked; the amplitude deviation index of each suspected seaweed pixel point is calculated, and all suspected seaweed pixel points whose amplitude deviation index is not greater than the preset amplitude threshold are obtained as target seaweed pixel points;

[0013] The target seagrass pixels of all green plant area remote sensing images in the time series green plant area remote sensing images are extracted to draw a seagrass distribution map.

[0014] Preferably, after acquiring the time series remote sensing images of the area to be identified, the method further includes preprocessing each satellite remote sensing image in the time series remote sensing images, including:

[0015] Perform radiation calibration, atmospheric correction, geometric correction, splicing and cropping on each satellite remote sensing image to obtain the corresponding optimized remote sensing image;

[0016] Select one scene from all satellite remote sensing images after atmospheric correction as a reference image, and extract the scale-invariant feature transformation features of the reference image;

[0017] Based on the scale-invariant feature transformation features of each optimized remote sensing image and the scale-invariant feature transformation features of the reference image, each optimized remote sensing image is registered with the reference image as a benchmark to complete the preprocessing of each satellite remote sensing image.

[0018] Preferably, the water body index of each scene of satellite remote sensing images in the time series remote sensing images is used to select the land-sea separation boundary, extract the ocean area in each scene of satellite remote sensing images, and form a time series ocean area remote sensing image, including:

[0019] Based on the green band and near-infrared band reflectance values ​​of each satellite remote sensing image in the time series remote sensing image, the water body index of each satellite remote sensing image in the time series remote sensing image is calculated;

[0020] Based on the water body index of each satellite remote sensing image and the preset water body index threshold, the water boundary corresponding to each satellite remote sensing image is obtained; the water boundary with the largest ocean area is obtained as the sea-land separation boundary;

[0021] Using the land-sea separation boundary, each satellite remote sensing image in the time series remote sensing image of the identified area is cropped, and the ocean area of ​​each satellite remote sensing image is extracted as the corresponding ocean area remote sensing image to form a time series ocean area remote sensing image.

[0022] Preferably, the water index of each satellite remote sensing image in the time series remote sensing image is calculated based on the green band and near infrared band reflectance values ​​of each satellite remote sensing image in the time series remote sensing image, which is expressed as:

[0023]

[0024] Among them, NDWI represents the water body index, G represents the green band, and NIR represents the near-infrared band reflectance value.

[0025] Preferably, the segmentation scale, shape factor, compactness, spectral weight and spatial weight of the multi-scale segmentation method are preset to divide each ocean area remote sensing image to obtain multiple homogeneous image spots in each ocean area remote sensing image.

[0026] Preferably, the step of obtaining the category of each homogeneous image spot in each ocean area remote sensing image based on the characteristic parameters of the homogeneous image spots includes:

[0027] A variety of typical homogeneous image spots and their characteristic parameters are obtained to construct a training set, and a gradient boosting decision tree is trained to obtain a target gradient boosting decision tree;

[0028] Extract characteristic parameters of each homogeneous image spot, including spectral characteristics, texture characteristics, contour characteristics and shape characteristics;

[0029] Based on the characteristic parameters of each homogeneous image spot, the target gradient boosting decision tree is used for classification, and the homogeneous image spots are divided into one of the marine water green plant type, marine water body type or benthic type.

[0030] Preferably, the submerged seaweed spectral index of each scene of green plant area remote sensing image is calculated based on the red light band reflectance value of each scene of green plant area remote sensing image and the first vegetation red edge band reflectance value, expressed as:

[0031]

[0032] Among them, SSII represents the submerged seagrass spectral index, VRE represents the reflectance value of the first vegetation red edge band, R represents the reflectance value of the red light band, and a represents the preset adjustment coefficient.

[0033] Preferably, the amplitude deviation index is expressed as:

[0034]

[0035] Where ADI represents the amplitude deviation index, σ A is the standard deviation of the pixel amplitude, m A is the mean amplitude of the pixels.

[0036] Preferably, the satellite remote sensing image is a Sentinel-2 satellite remote sensing image.

[0037] This embodiment provides a seaweed distribution map drawing device, including:

[0038] An image acquisition module is used to obtain satellite remote sensing images of the area to be identified at each sampling time within a preset time period to form a time-series remote sensing image of the area to be identified;

[0039] The ocean area acquisition module is used to select the land-sea separation boundary based on the water body index of each satellite remote sensing image in the time series remote sensing image, extract the ocean area in each satellite remote sensing image, and form a time series ocean area remote sensing image;

[0040] A homogeneous image spot extraction and classification module is used to perform image segmentation on each ocean area remote sensing image in the time series ocean area remote sensing image using a multi-scale segmentation method to obtain a plurality of homogeneous image spots in each ocean area remote sensing image; based on characteristic parameters of the homogeneous image spots, obtain the category of each homogeneous image spot in each ocean area remote sensing image, the categories including marine water green plant type, marine water body type and benthic type;

[0041] The green plant area acquisition module is used to obtain the area composed of all homogeneous image spots belonging to the type of green plants in the ocean water in each ocean area remote sensing image, as the green plant area remote sensing image corresponding to each ocean area remote sensing image, to form a time series green plant area remote sensing image;

[0042] The target seaweed pixel extraction module is used to calculate the submerged seaweed spectral index of each scene of the green plant area remote sensing image in the time-series green plant area remote sensing image based on the red light band reflectance value of each scene of the green plant area remote sensing image and the first vegetation red edge band reflectance value, compare it with the preset spectral index range, and mark the suspected seaweed pixel points; calculate the amplitude deviation index of each suspected seaweed pixel point, and obtain all suspected seaweed pixel points whose amplitude deviation index is not greater than the preset amplitude threshold as the target seaweed pixel points;

[0043] The distribution map drawing module is used to extract the target seagrass pixels of all green plant area remote sensing images in the time series green plant area remote sensing images and draw the seagrass distribution map.

[0044] The above technical solution of the present invention has the following beneficial effects compared with the prior art:

[0045] The method for drawing a seagrass distribution map of the present invention comprises the following steps: after obtaining a satellite remote sensing image of an area to be identified, using a water body index to distinguish between land and sea areas, and obtaining possible seagrass growth areas; after obtaining a plurality of homogeneous image spots using a multi-scale segmentation method, classifying the homogeneous image spots, and obtaining types of green plants in the ocean water; calculating the submerged seagrass spectral index and amplitude deviation index of a remote sensing image of a green plant area composed of homogeneous image spots of green plant types in the ocean water, and obtaining target seagrass pixel points therefrom to draw a seagrass distribution map; based on the different ecological attributes of seagrass and floating algae, that is, the growth position of seagrass is fixed over time, while the growth position of floating algae is variable, the present invention uses the plant spectral index, the submerged seagrass spectral index and the amplitude deviation index in the remote sensing image to distinguish the image pixels of seagrass and floating algae, accurately obtains the seagrass area in the area to be identified, improves the accuracy of seagrass remote sensing survey, and improves the accuracy of the drawn seagrass distribution map. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to make the content of the present invention more clearly understood, the present invention is further described in detail below according to specific embodiments of the present invention in conjunction with the accompanying drawings, wherein

[0047] Figure 1 is a flow chart of the steps of the method for drawing a seagrass distribution map provided by the present invention;

[0048] Figure 2 It is a detailed step identification flow chart of seagrass map drawing provided by the present invention;

[0049] Figure 3 The present invention provides a flow chart for drawing a seagrass distribution map based on the Sentinel-2 satellite remote sensing image. DETAILED DESCRIPTION

[0050] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.

[0051] Although seagrass and algae both show the spectral characteristics of plants in remote sensing images, in reality, seagrass is a perennial plant with a developed root system fixed to the bottom of the ocean. Therefore, except for the natural extinction and new growth of seagrass areas, the location of most seagrass growth does not change; while algae do not have a fixed root system, and the growth burst is often closely related to the deterioration of ocean water quality, showing random changes, and the location of its burst spread is also affected by waves and sea breeze, so its location is always changing. Therefore, based on the different ecological properties of seagrass and seaweed, that is, the significant difference in the fixed / changing growth location of seagrass and seaweed over time, the present invention uses the amplitude deviation index ADI of the submerged seagrass spectral index SSII in the remote sensing image to distinguish the image pixels of seagrass and seaweed, so as to improve the accuracy of seagrass remote sensing survey.

[0052] Reference Figure 1 As shown, a flow chart of the steps of a method for drawing a seaweed distribution map provided by the present invention specifically includes:

[0053] S101: Acquire satellite remote sensing images of the area to be identified at each sampling time within a preset time period to form a time series remote sensing image of the area to be identified;

[0054] S102: based on the water body index of each satellite remote sensing image in the time series remote sensing image, selecting the land-sea separation boundary, extracting the ocean area in each satellite remote sensing image, and forming a time series ocean area remote sensing image;

[0055] The water body index NDWI is expressed as: G is the green band, and NIR is the near-infrared band reflectance value;

[0056] S103: for each ocean area remote sensing image in the time series ocean area remote sensing image, perform image segmentation using a multi-scale segmentation method to obtain a plurality of homogeneous image spots in each ocean area remote sensing image;

[0057] S104: based on the characteristic parameters of the homogeneous image spots, obtaining the category of each homogeneous image spot in each ocean area remote sensing image, wherein the category includes the type of green plants in the ocean water, the type of ocean water body and the benthic type;

[0058] S105: Acquire an area composed of all homogeneous image spots belonging to the type of green plants in the ocean water in each ocean area remote sensing image as a green plant area remote sensing image corresponding to each ocean area remote sensing image, to form a time series green plant area remote sensing image;

[0059] S106: for each scene of the green plant area remote sensing images in the time series of green plant area remote sensing images, based on the red light band reflectance value of each scene of the green plant area remote sensing image and the first vegetation red edge band reflectance value, calculate the submerged seaweed spectral index of each scene of the green plant area remote sensing image, compare it with the preset spectral index range, and mark the suspected seaweed pixel points; calculate the amplitude deviation index of each suspected seaweed pixel point, and obtain all suspected seaweed pixel points whose amplitude deviation index is not greater than the preset amplitude threshold as the target seaweed pixel points;

[0060] The submerged seagrass spectral index SSII is expressed as: VRE represents the reflectance value of the first vegetation red edge band, R represents the reflectance value of the red light band, and a represents the preset adjustment coefficient;

[0061] The amplitude deviation index ADI is expressed as: σ A is the standard deviation of the pixel amplitude, m A is the mean amplitude of the pixels;

[0062] S107: extracting target seagrass pixels from all green plant area remote sensing images in the time series green plant area remote sensing images, and drawing a seagrass distribution map.

[0063] In the embodiment of the present invention, after obtaining the time series remote sensing images of the area to be identified in step S101, the process further includes preprocessing each satellite remote sensing image in the time series remote sensing images, including:

[0064] Perform radiation calibration, atmospheric correction, geometric correction, splicing and cropping on each satellite remote sensing image to obtain the corresponding optimized remote sensing image;

[0065] Select one scene from all satellite remote sensing images after atmospheric correction as a reference image, and extract the scale-invariant feature transformation features of the reference image;

[0066] Based on the scale-invariant feature transformation features of each optimized remote sensing image and the scale-invariant feature transformation features of the reference image, each optimized remote sensing image is registered with the reference image as a benchmark to complete the preprocessing of each satellite remote sensing image.

[0067] Specifically, in step S102, the construction of the time series ocean area remote sensing image includes:

[0068] S102-1: Calculate the water index of each satellite remote sensing image in the time series remote sensing image based on the green band and near infrared band reflectance values ​​of each satellite remote sensing image in the time series remote sensing image;

[0069] S102-2: based on the water body index of each satellite remote sensing image and a preset water body index threshold, obtaining the water boundary corresponding to each satellite remote sensing image; obtaining the water boundary with the largest ocean area as the sea-land separation boundary;

[0070] S102-3: Using the land-sea separation boundary, crop each satellite remote sensing image in the time series remote sensing image of the area to be identified, extract the ocean area of ​​each satellite remote sensing image as the corresponding ocean area remote sensing image, and form a time series ocean area remote sensing image.

[0071] In an embodiment of the present invention, when a multi-scale segmentation method is used to perform image segmentation to obtain homogeneous image spots, the segmentation scale, shape factor, compactness, spectral weight and spatial weight of the multi-scale segmentation method are preset, and each ocean area remote sensing image is divided to obtain multiple homogeneous image spots in each ocean area remote sensing image.

[0072] Specifically, in step S104, the homogeneous image spots are classified, including:

[0073] S104-1: Acquire multiple typical homogeneous image spots and their characteristic parameters to construct a training set, train a gradient boosting decision tree, and obtain a target gradient boosting decision tree;

[0074] S104-2: extracting characteristic parameters of each homogeneous image spot, including spectral characteristics, texture characteristics, contour characteristics and shape characteristics;

[0075] S104-3: Based on the characteristic parameters of each homogeneous image spot, the target gradient boosting decision tree is used for classification, and the homogeneous image spots are divided into one of the marine water green plant type, the marine water body type or the benthic type.

[0076] The method for drawing a seagrass distribution map of the present invention comprises the following steps: after obtaining a satellite remote sensing image of an area to be identified, using a water body index to distinguish between land and sea areas, and obtaining possible seagrass growth areas; after obtaining a plurality of homogeneous image spots using a multi-scale segmentation method, classifying the homogeneous image spots, and obtaining types of green plants in the ocean water; calculating the submerged seagrass spectral index and amplitude deviation index of a remote sensing image of a green plant area composed of homogeneous image spots of green plant types in the ocean water, and obtaining target seagrass pixel points therefrom to draw a seagrass distribution map; based on the different ecological attributes of seagrass and floating algae, that is, the growth position of seagrass is fixed over time, while the growth position of floating algae is variable, the present invention uses the plant spectral index, the submerged seagrass spectral index and the amplitude deviation index in the remote sensing image to distinguish the image pixels of seagrass and floating algae, accurately obtains the seagrass area in the area to be identified, improves the accuracy of seagrass remote sensing survey, and improves the accuracy of the drawn seagrass distribution map.

[0077] Reference Figure 2As shown, it is a detailed step discrimination flow chart of seagrass map drawing provided by an embodiment of the present invention; the present invention obtains remote sensing images based on time series in the area to be identified, and performs preliminary preprocessing on the time series remote sensing images; water boundary extraction and cropping are performed on the time series images to determine that the seawater coverage area is the possible growth area of ​​seagrass, that is, the ocean area; for the time series remote sensing images of the ocean area, an object-oriented method is used for classification, and the time series remote sensing images are divided into three categories of green plants in the ocean water, marine water bodies and other benthic plants through the steps of segmenting homogeneous image spots, extracting feature parameters, and running classification algorithms; for the green plants in the ocean water, the submerged seagrass extraction index is calculated to obtain a time series seagrass spectral image; for the time series seagrass spectral image, the amplitude deviation index of each pixel is calculated; the amplitude deviation index can reflect whether the seagrass spectrum remains unchanged in each ocean position; according to the calculation result of the amplitude deviation index, the position where the seagrass spectrum remains unchanged in the image is marked as seagrass, and the position where the seagrass spectrum changes is marked as impetuous and deleted, thereby realizing the removal of impetuousness in the process of seagrass identification.

[0078] Based on the above embodiment, in the embodiment of the present invention, the satellite remote sensing images collected by the Sentinel-2 satellite are used to draw the seagrass distribution map. Figure 3 As shown in the figure, it is a flow chart of drawing seagrass distribution map based on Sentinel-2 satellite remote sensing image. The specific steps include:

[0079] S201: Acquire the time series of Sentinel-2 satellite remote sensing images of the study area within 2 years with a time interval of 10 days;

[0080] S202: Use SANP software to perform radiometric calibration on the Sentinel-2 satellite remote sensing images, use ALLCOT software to perform atmospheric correction using the dark spectrum fitting method, and use ARCGIS software to perform geometric correction, stitching, and cropping of all images;

[0081] S203: Select an image in the middle period from the atmospherically corrected time series images as a reference image, and use ARCGIS software to extract the SIFT features of the image; and use ARCGIS software to register other images with the reference image as a benchmark;

[0082] S204: Calculate the water index NDWI of each image in the aligned time series image using ENVI software. The calculation formula is: G is the green band and NIR is the reflectance value of the near infrared band.

[0083] S205: setting a threshold of the NDWI calculation result to determine a threshold range for water boundary extraction; determining a maximum range water boundary at the highest tidal level from the extracted multi-period water boundaries as a sea-land separation boundary for seagrass growth;

[0084] S206: Using the maximum range water boundary to crop the time series image, the seawater coverage is determined as the possible seagrass growth area (i.e., the ocean area), which is retained, and the rest is removed to eliminate the spectral interference of land vegetation and objects;

[0085] S207: Segment each image of the possible seagrass growth area to extract homogeneous image spots, using a multi-scale segmentation method; divide the segmented image homogeneous image spots into three types: green plants in the ocean, ocean water bodies and other benthic plants;

[0086] During classification, characteristic parameters are extracted for each image spot, including spectral characteristics, texture characteristics, contour characteristics and shape characteristics, and typical homogeneous image spots are selected as samples for classification using the gradient boosting decision tree (GBDT) method, resulting in three image types: green plants in the ocean, ocean water bodies and other benthic images.

[0087] S208: extracting the distribution range of green plants in the ocean, performing image cropping, and obtaining a time-series image of the green vegetation area;

[0088] S209: Calculate the submerged seagrass spectral index SS II for the time series images of the green vegetation area and obtain the time series seagrass spectral images;

[0089] The calculation formula of submerged seagrass spectral index SS II is: VRE represents the reflectance value of the first vegetation red edge band, R represents the reflectance value of the red light band, and a represents the preset adjustment coefficient, which takes the minimum value that is not 0;

[0090] The calculated SSI I value of the image in the seagrass covered area should be greater than 1, and the larger the SSI I value, the higher the seagrass coverage density in the area;

[0091] S210: setting the SSI I threshold for the time series seaweed spectral image, marking the pixels within the threshold range as suspected seaweed; calculating the amplitude deviation index ADI of each pixel in the time series seaweed spectral image, the amplitude deviation index can reflect whether the seaweed spectrum at each position remains unchanged;

[0092] The calculation formula of the amplitude deviation index ADI is expressed as: σ A is the amplitude standard deviation, m A is the mean amplitude.

[0093] Set the threshold of the amplitude deviation index ADI and mark the pixels with a value greater than the threshold as floating algae;

[0094] S211: All pixel points marked as algae in the image are removed, and the remaining pixel points are determined to be seaweed, so that the algae can be removed during the seaweed identification process, so that a seaweed distribution map can be drawn according to the area formed by the retained pixel points.

[0095] The present invention is based on the property that the geographical spatial positions of the seagrass and floating algae growth areas in the real surface environment are fixed or changing. The invention uses remote sensing images to effectively distinguish seagrass from algae, and reduces the error influence caused by floating algae factors in seagrass remote sensing monitoring, thereby improving the accuracy of seagrass remote sensing surveys and further improving the accuracy of drawing seagrass distribution maps.

[0096] Based on the above embodiments, an embodiment of the present invention provides a seaweed distribution map drawing device, which may include:

[0097] The image acquisition module 100 is used to acquire satellite remote sensing images of the area to be identified at each sampling time within a preset time period to form a time series remote sensing image of the area to be identified;

[0098] The ocean area acquisition module 200 is used to select the land-sea separation boundary based on the water body index of each satellite remote sensing image in the time series remote sensing image, extract the ocean area in each satellite remote sensing image, and form a time series ocean area remote sensing image;

[0099] The homogeneous image spot extraction and classification module 300 is used to perform image segmentation on each ocean area remote sensing image in the time series ocean area remote sensing image using a multi-scale segmentation method to obtain a plurality of homogeneous image spots in each ocean area remote sensing image; based on the characteristic parameters of the homogeneous image spots, obtain the category of each homogeneous image spot in each ocean area remote sensing image, wherein the category includes the type of green plants in the ocean water, the type of ocean water body and the benthic type;

[0100] The green plant area acquisition module 400 is used to acquire an area composed of all homogeneous image spots belonging to the type of green plants in the ocean water in each ocean area remote sensing image as the green plant area remote sensing image corresponding to each ocean area remote sensing image, so as to form a time series green plant area remote sensing image;

[0101] The target seaweed pixel extraction module 500 is used to calculate the submerged seaweed spectral index of each scene of the green plant area remote sensing image in the time-series green plant area remote sensing image based on the red light band reflectance value of each scene of the green plant area remote sensing image and the first vegetation red edge band reflectance value, compare it with the preset spectral index interval, and mark the suspected seaweed pixel points; calculate the amplitude deviation index of each suspected seaweed pixel point, and obtain all suspected seaweed pixel points whose amplitude deviation index is not greater than the preset amplitude threshold as the target seaweed pixel points;

[0102] The distribution map drawing module 600 is used to extract target seagrass pixels from all green plant area remote sensing images in the time series green plant area remote sensing images, and draw a seagrass distribution map.

[0103] The seaweed distribution map drawing device of this embodiment is used to implement the aforementioned seaweed distribution map drawing method. Therefore, the specific implementation of the seaweed distribution map drawing device can be seen in the embodiment part of the seaweed distribution map drawing method in the above text, for example,

[0104] The image acquisition module 100 and the ocean area acquisition module 200 are respectively used to implement steps S101 and S102 in the above-mentioned seagrass distribution map drawing method; the homogeneous image spot extraction and classification module 300 is used to implement steps S103 and S104 in the above-mentioned seagrass distribution map drawing method; the green plant area acquisition module 400, the target seagrass pixel point extraction module 500, and the distribution map drawing module 600 are respectively used to implement steps S105, S106 and S107 in the above-mentioned seagrass distribution map drawing method. Therefore, its specific implementation method can refer to the description of the corresponding various partial embodiments and will not be repeated here.

[0105] The method for drawing a seagrass distribution map of the present invention comprises the following steps: after obtaining a satellite remote sensing image of an area to be identified, using a water body index to distinguish between land and sea areas, and obtaining possible seagrass growth areas; after obtaining a plurality of homogeneous image spots using a multi-scale segmentation method, classifying the homogeneous image spots, and obtaining types of green plants in the ocean water; calculating the submerged seagrass spectral index and amplitude deviation index of a remote sensing image of a green plant area composed of homogeneous image spots of green plant types in the ocean water, and obtaining target seagrass pixel points therefrom to draw a seagrass distribution map; based on the different ecological attributes of seagrass and floating algae, that is, the growth position of seagrass is fixed over time, while the growth position of floating algae is variable, the present invention uses the plant spectral index, the submerged seagrass spectral index and the amplitude deviation index in the remote sensing image to distinguish the image pixels of seagrass and floating algae, accurately obtains the seagrass area in the area to be identified, improves the accuracy of seagrass remote sensing survey, and improves the accuracy of the drawn seagrass distribution map.

[0106] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0107] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0108] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0109] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0110] Obviously, the above embodiments are merely examples for the purpose of clear explanation and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the present invention.

Claims

1. A method for drawing a seagrass distribution map, characterized in that: include: Obtain satellite remote sensing images of the area to be identified at each sampling time within a preset time period to form a time series remote sensing image of the area to be identified; Based on the water body index of each satellite remote sensing image in the time series remote sensing image, the land-sea separation boundary is selected, and the ocean area in each satellite remote sensing image is extracted to form a time series ocean area remote sensing image; For each ocean area remote sensing image in the time series ocean area remote sensing image, a multi-scale segmentation method is used to perform image segmentation to obtain multiple homogeneous image spots in each ocean area remote sensing image; Based on the characteristic parameters of the homogeneous image spots, the categories of each homogeneous image spot in each ocean area remote sensing image are obtained, wherein the categories include marine water green plant types, marine water body types and benthic types; Acquire the area composed of all homogeneous image spots belonging to the type of green plants in the ocean water in each ocean area remote sensing image as the green plant area remote sensing image corresponding to each ocean area remote sensing image, and form a time series green plant area remote sensing image; For each scene of the green plant area remote sensing images in the time series of green plant area remote sensing images, the submerged seaweed spectral index of each scene of the green plant area remote sensing image is calculated based on the red light band reflectance value of each scene of the green plant area remote sensing image and the first vegetation red edge band reflectance value, and compared with the preset spectral index range, the suspected seaweed pixel points are marked; the amplitude deviation index of each suspected seaweed pixel point is calculated, and all suspected seaweed pixel points whose amplitude deviation index is not greater than the preset amplitude threshold are obtained as target seaweed pixel points; The target seagrass pixels of all green plant area remote sensing images in the time series green plant area remote sensing images are extracted to draw a seagrass distribution map.

2. The method for drawing a seaweed distribution map according to claim 1, characterized in that: After obtaining the time series remote sensing images of the area to be identified, each satellite remote sensing image in the time series remote sensing images is preprocessed, including: Perform radiation calibration, atmospheric correction, geometric correction, splicing and cropping on each satellite remote sensing image to obtain the corresponding optimized remote sensing image; Select one scene from all satellite remote sensing images after atmospheric correction as a reference image, and extract the scale-invariant feature transformation features of the reference image; Based on the scale-invariant feature transformation features of each optimized remote sensing image and the scale-invariant feature transformation features of the reference image, each optimized remote sensing image is registered with the reference image as a benchmark to complete the preprocessing of each satellite remote sensing image.

3. The method for drawing a seaweed distribution map according to claim 1, characterized in that: The method of selecting the land-sea separation boundary based on the water body index of each satellite remote sensing image in the time series remote sensing image, extracting the ocean area in each satellite remote sensing image, and forming a time series ocean area remote sensing image includes: Based on the green band and near-infrared band reflectance values ​​of each satellite remote sensing image in the time series remote sensing image, the water body index of each satellite remote sensing image in the time series remote sensing image is calculated; Based on the water body index of each satellite remote sensing image and the preset water body index threshold, the water boundary corresponding to each satellite remote sensing image is obtained; the water boundary with the largest ocean area is obtained as the sea-land separation boundary; Using the land-sea separation boundary, each satellite remote sensing image in the time series remote sensing image of the identified area is cropped, and the ocean area of ​​each satellite remote sensing image is extracted as the corresponding ocean area remote sensing image to form a time series ocean area remote sensing image.

4. The method for drawing a seaweed distribution map according to claim 3, characterized in that: The water index of each satellite remote sensing image in the time series remote sensing image is calculated based on the green band and near infrared band reflectance values ​​of each satellite remote sensing image in the time series remote sensing image, which is expressed as: Among them, NDWI represents the water body index, G represents the green band, and NIR represents the near-infrared band reflectance value.

5. The method for drawing a seaweed distribution map according to claim 1, characterized in that: The segmentation scale, shape factor, compactness, spectral weight and spatial weight of the multi-scale segmentation method are preset to divide each ocean area remote sensing image and obtain multiple homogeneous image spots in each ocean area remote sensing image.

6. The method for drawing a seaweed distribution map according to claim 1, characterized in that: The method of obtaining the category of each homogeneous image spot in each ocean area remote sensing image based on the characteristic parameters of the homogeneous image spot includes: A variety of typical homogeneous image spots and their characteristic parameters are obtained to construct a training set, and a gradient boosting decision tree is trained to obtain a target gradient boosting decision tree; Extract characteristic parameters of each homogeneous image spot, including spectral characteristics, texture characteristics, contour characteristics and shape characteristics; Based on the characteristic parameters of each homogeneous image spot, the target gradient boosting decision tree is used for classification, and the homogeneous image spots are divided into one of the marine water green plant type, marine water body type or benthic type.

7. The method for drawing a seaweed distribution map according to claim 1, characterized in that: Based on the red light band reflectance value of each scene of green plant area remote sensing image and the first vegetation red edge band reflectance value, the submerged seaweed spectral index of each scene of green plant area remote sensing image is calculated, which is expressed as: Among them, SSII represents the submerged seagrass spectral index, VRE represents the reflectance value of the first vegetation red edge band, R represents the reflectance value of the red light band, and a represents the preset adjustment coefficient.

8. The method for drawing a seaweed distribution map according to claim 1, characterized in that: The amplitude deviation index is expressed as: Where ADI represents the amplitude deviation index, σ A is the standard deviation of the pixel amplitude, m A is the mean amplitude of the pixels.

9. The method for drawing a seaweed distribution map according to claim 1, characterized in that: The satellite remote sensing image is a Sentinel-2 satellite remote sensing image.

10. A seaweed distribution map drawing device, characterized in that: include: An image acquisition module is used to obtain satellite remote sensing images of the area to be identified at each sampling time within a preset time period to form a time-series remote sensing image of the area to be identified; The ocean area acquisition module is used to select the land-sea separation boundary based on the water body index of each satellite remote sensing image in the time series remote sensing image, extract the ocean area in each satellite remote sensing image, and form a time series ocean area remote sensing image; A homogeneous image spot extraction and classification module is used to perform image segmentation on each ocean area remote sensing image in the time series ocean area remote sensing image using a multi-scale segmentation method to obtain multiple homogeneous image spots in each ocean area remote sensing image; Based on the characteristic parameters of the homogeneous image spots, the categories of each homogeneous image spot in each ocean area remote sensing image are obtained, wherein the categories include marine water green plant types, marine water body types and benthic types; The green plant area acquisition module is used to obtain the area composed of all homogeneous image spots belonging to the type of green plants in the ocean water in each ocean area remote sensing image, as the green plant area remote sensing image corresponding to each ocean area remote sensing image, to form a time series green plant area remote sensing image; The target seaweed pixel extraction module is used to calculate the submerged seaweed spectral index of each scene of the green plant area remote sensing image in the time-series green plant area remote sensing image based on the red light band reflectance value of each scene of the green plant area remote sensing image and the first vegetation red edge band reflectance value, compare it with the preset spectral index range, and mark the suspected seaweed pixel points; calculate the amplitude deviation index of each suspected seaweed pixel point, and obtain all suspected seaweed pixel points whose amplitude deviation index is not greater than the preset amplitude threshold as the target seaweed pixel points; The distribution map drawing module is used to extract the target seagrass pixels of all green plant area remote sensing images in the time series green plant area remote sensing images and draw the seagrass distribution map.

Images