Long and narrow river cyanobacterial bloom information identification method and system based on remote sensing data

Abstract

The invention discloses a long and narrow river cyanobacterial bloom information identification method and system based on remote sensing data, and the method comprises the following steps: selecting a remote sensing image in a research region range through a CEE cloud platform, and carrying out the preprocessing of the remote sensing image; a Sentinel-2 satellite image and a Landsat satellite image are respectively utilized to detect a cloud layer in the preprocessed remote sensing image within a cloud cover threshold value; based on a detection result, removing cloud pixels from the preprocessed remote sensing image to obtain a cloud-removed image; analyzing the cloud-removed image by using the improved normalized difference water body index, and determining the threshold change range of the water body in each image of the cloud-removed image so as to realize the recognition of the water body area; identifying cyanobacterial bloom information in the water body area by using a floating algae index method, and determining a threshold change range of cyanobacterial bloom in each image so as to realize identification of distribution characteristics of cyanobacterial bloom; according to the method, the accuracy and flexibility of narrow and long river cyanobacterial bloom information identification are remarkably improved.

Description

Technical Field

[0001] The invention relates to the technical field of cyanobacteria bloom information identification, and more particularly to a method and system for identifying cyanobacteria bloom information in a narrow and long river based on remote sensing data. Background Art

[0002] In recent years, with the rapid advancement of science and technology, remote sensing technology has become an indispensable and important means in the field of environmental monitoring. Through sensors mounted on platforms such as satellites, aircraft or drones, remote sensing technology can obtain surface information over a long distance and over a large range, including water quality, vegetation cover, land use and other environmental parameters. In the monitoring of cyanobacterial blooms, remote sensing technology has gradually replaced the traditional time-consuming and labor-intensive on-site sampling and laboratory analysis methods with its advantages of high efficiency, speed and non-contact, and has become the preferred monitoring method in this field. Cyanobacterial blooms are a direct consequence of eutrophication of water bodies. Their large-scale reproduction will not only lead to water quality deterioration and affect the balance of aquatic ecosystems, but may also pose a threat to human health through the food chain. Therefore, timely monitoring and early warning of cyanobacterial blooms are of great significance for protecting water resources and maintaining ecological security.

[0003] Most of the current remote sensing technologies and methods for identifying algal blooms focus on open waters such as lakes or large reservoirs, and their applicability to long and narrow rivers is obviously limited. Due to their unique geographical features and water flow characteristics, long and narrow rivers make the monitoring of cyanobacterial blooms particularly complicated. Such rivers often have narrow channels, many bends, and uneven water flow rates, which pose many challenges to remote sensing monitoring. These characteristics not only require the monitoring method to have a high degree of accuracy, but also ensure sufficient monitoring coverage, which poses a severe challenge to traditional remote sensing monitoring technology and is difficult to meet the actual needs of cyanobacterial bloom monitoring in long and narrow rivers. Therefore, it is particularly important and urgent to develop a remote sensing monitoring technology for cyanobacterial blooms that specifically targets the characteristics of long and narrow rivers. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for identifying blue algae bloom information in narrow and long rivers based on remote sensing data, which can automatically adapt to the dynamic changes of different environmental conditions in the river, realize dynamic adjustment and optimization of thresholds, thereby improving the accuracy and flexibility of identifying blue algae bloom information in narrow and long rivers.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A method for identifying blue algae bloom information in a narrow river based on remote sensing data comprises the following steps:

[0007] Using the CEE cloud platform, remote sensing images within the study area were selected and preprocessed;

[0008] The cloud threshold is set, and the cloud layer in the pre-processed remote sensing image is detected using Sentinel-2 satellite imagery and Landsat satellite imagery within the cloud threshold. Based on the detection results, cloud pixels are removed from the pre-processed remote sensing image to obtain a de-clouded image.

[0009] The cloud-removed image is analyzed using the improved normalized difference water index to determine the threshold variation range of water bodies in each scene of the cloud-removed image, so as to realize the identification of water body areas.

[0010] The phytoplankton index method is used to identify the cyanobacteria bloom information in the water area, and the threshold change range of cyanobacteria bloom in each image is determined to realize the identification of the distribution characteristics of cyanobacteria bloom.

[0011] Visualize the recognition results.

[0012] Further, the cloud layer in the preprocessed remote sensing image is detected within the cloud amount threshold using the Sentinel-2 satellite image and the Landsat satellite image respectively; wherein the cloud layer in the preprocessed remote sensing image is detected within the cloud amount threshold using the Sentinel-2 satellite image, specifically:

[0013] The QA60 band of the Sentinel-2 satellite image is used to detect clouds in the preprocessed remote sensing image within the cloud cover threshold. The expression is as follows:

[0014] Cloud QA60 =Bit101∪Bit111

[0015] Among them, Cloud QA60 Represents the cloud noise pixel in the QA60 band; Bit101 means that the 10th bit of the current pixel in the QA60 band is 1, and Bit111 means that the 11th bit of the current pixel in the QA60 band is 1.

[0016] Further, the cloud layer in the preprocessed remote sensing image is detected within the cloud amount threshold using the Sentinel-2 satellite image and the Landsat satellite image respectively; wherein the cloud layer in the preprocessed remote sensing image is detected within the cloud amount threshold using the Landsat satellite image, specifically:

[0017] The QA_PIXEL band in the Landsat image is used to detect the cloud layer in the preprocessed remote sensing image within the cloud amount threshold. The expression is as follows:

[0018] Cloud QA_PIXEL =Bit31∪Bit51

[0019] Among them, CloudQA_PIXEL Represents the cloud noise pixel in the QA60 band; Bit31 indicates that the third bit of the current pixel in the QA_PIXEL band is 1, and Bit51 indicates that the fifth bit of the current pixel in the QA_PIXEL band is 1.

[0020] Furthermore, the improved normalized difference water index is used to analyze the cloud-free image, and the expression is:

[0021] I MNDWI =(R Green -R SWIR ) / (R Green +R SWIR )

[0022] In the above formula, R Green is the reflectivity of green light band; R SWIR is the reflectivity in the shortwave infrared band.

[0023] Furthermore, the phytoplankton index method is used to identify the cyanobacteria bloom information in the water area, and the expression is:

[0024] FAI=R NIR -R NIR '

[0025] R NIR ′=R RED +(R SWIR -R RED )×(λ NIR -λ RED ) / (λ SWIR -λ RED )

[0026] In the above formula, FAI is phytoplankton, R RED , R NIR , R SWIR Corresponding to the red, near infrared, and short infrared bands, λ RED、 λ NIR、 λ SWIR Corresponding to the central wavelengths of red light, near infrared, and short infrared bands respectively; R NIR ′ is the interpolated reflectance, that is, linear interpolation is used between the red band and the short-wave infrared band to obtain the reflectance information in the near-infrared band.

[0027] Furthermore, the threshold variation range of the water body and the threshold variation range of the cyanobacteria bloom are specifically determined by the Otsu threshold method.

[0028] Furthermore, the determination is performed using the Otsu threshold method, specifically:

[0029] The slope and statistical analysis methods were used to determine the threshold variation range of cyanobacteria bloom in each image.

[0030] Furthermore, the selecting of remote sensing images within the study area specifically includes: obtaining high-quality, cloudless or lightly clouded remote sensing images within the study area.

[0031] The present invention also provides a system for identifying blue algae blooms in narrow and long rivers based on remote sensing data, comprising:

[0032] The data processing module uses the CEE cloud platform to select remote sensing images within the study area and perform preprocessing. The cloud threshold is set, and the cloud layer in the preprocessed remote sensing image is detected within the cloud threshold using Sentinel-2 satellite images and Landsat satellite images. Based on the detection results, cloud pixels are removed from the preprocessed remote sensing image to obtain a de-clouded image.

[0033] The recognition module uses the improved normalized difference water index to analyze the de-clouded image and determine the threshold change range of the water body in each image of the de-clouded image to realize the recognition of the water body area;

[0034] The phytoplankton index method is used to identify the cyanobacteria bloom information in the water area, and the threshold change range of cyanobacteria bloom in each image is determined to realize the identification of the distribution characteristics of cyanobacteria bloom.

[0035] The visualization module visualizes the recognition results.

[0036] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of a method for identifying blue algae bloom information in narrow and long rivers based on remote sensing data are implemented.

[0037] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0038] (1) The present invention relies on obtaining remote sensing image data within the study area from the Google Earth Engine (GEE) cloud platform, and uses Sentinel-2 satellite images and Landsat satellite images to detect clouds in pre-processed remote sensing images within a threshold value; then uses the improved normalized difference water-soil index and phytoplankton index method to identify the distribution characteristics of water body areas and cyanobacteria water bodies, effectively distinguishing water bodies from other types of land objects such as vegetation and bare soil, significantly amplifying the spectral characteristics of water bodies, and improving the accuracy of water body identification; it can also automatically adapt to the dynamic changes of different environmental conditions in the river, no longer completely relying on expert experience and static statistical analysis, and realize dynamic adjustment and optimization of thresholds, thereby improving the accuracy and flexibility of identification.

[0039] (2) In view of the characteristics of long and narrow rivers, by adopting efficient image processing technology, feature extraction methods and classification algorithms, as well as in-depth analysis of river morphology, hydrological characteristics and climatic conditions, we have achieved accurate monitoring of the distribution and changes of algal blooms, optimized data processing procedures, reduced the consumption of computing resources, and significantly improved the speed and accuracy of identifying algal blooms in long and narrow rivers. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0041] The following is a further description of the method and system for identifying blue algae blooms in narrow and long rivers based on remote sensing data in conjunction with the accompanying drawings;

[0042] Figure 1 This is a flow chart of a method for identifying blue algae bloom information in a narrow and long river based on remote sensing data provided by the present invention. DETAILED DESCRIPTION

[0043] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0044] In order to better understand the purpose, structure and function of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings.

[0045] like Figure 1 As shown, the present invention provides a method for identifying blue algae bloom information in a narrow and long river based on remote sensing data, comprising the following steps:

[0046] Using the CEE cloud platform, remote sensing images within the study area were selected and preprocessed;

[0047] It should be noted that the remote sensing data in the present invention are Landsat series Collection 2 (C2) level data and Sentinel-2 Level-2A (L2A) level data obtained from the Google Earth Engine (GEE) platform, and have undergone preliminary processing such as radiation calibration, atmospheric correction, geometric correction and image cropping.

[0048] A cloud cover threshold is set, and Sentinel-2 satellite images and Landsat satellite images are used to detect clouds in the preprocessed remote sensing images within the cloud cover threshold. A cloud mask is generated based on the detection results, and the cloud mask is used to remove cloud pixels from the preprocessed remote sensing images to obtain a de-clouded image, so as to detect and remove cloud-covered areas in remote sensing images.

[0049] It should be noted that: in the present invention, a cloud cover threshold is set (for example, CLOUDY_PIXEL_PE RCENTAGE is 10%) to filter out image sets with cloud cover less than the threshold.

[0050] The cloud-removed image is analyzed using the improved normalized difference water index to determine the threshold variation range of water bodies in each scene of the cloud-removed image, so as to realize the identification of water body areas.

[0051] The phytoplankton index method is used to identify the cyanobacteria bloom information in the water area, and the threshold change range of cyanobacteria bloom in each image is determined to realize the identification of the distribution characteristics of cyanobacteria bloom.

[0052] The identification results of the distribution characteristics of cyanobacteria blooms are visualized.

[0053] Sentinel-2 satellite images and Landsat satellite images are used to detect clouds in the preprocessed remote sensing images within the cloud cover threshold. Specifically, Sentinel-2 satellite images are used to detect clouds in the preprocessed remote sensing images within the cloud cover threshold.

[0054] The QA60 band of the Sentinel-2 satellite image is used to detect clouds in the preprocessed remote sensing image within the cloud cover threshold. The expression is as follows:

[0055] Cloud QA60 =Bit101∪Bit111

[0056] Among them, Cloud QA60 Represents the cloud noise pixel in the QA60 band; Bit101 means that the 10th bit of the current pixel in the QA60 band is 1, and Bit111 means that the 11th bit of the current pixel in the QA60 band is 1.

[0057] It should be noted that the QA60 band of the Sentinel-2 image is a cloud mask band containing bit mask information, in which specific bits (such as bits 10 and 11) are used to identify the presence of clouds. When bit 10 of the QA60 band is 1, it means that the current pixel is affected by thick cloud noise; when bit 11 is 1, it means that the current pixel is affected by cirrus noise.

[0058] The method of detecting the cloud layer in the preprocessed remote sensing image by using the Sentinel-2 satellite image and the Landsat satellite image within the cloud amount threshold is as follows:

[0059] The QA_PIXEL band in the Landsat image is used to detect the cloud layer in the preprocessed remote sensing image within the cloud amount threshold. The expression is as follows:

[0060] Cloud QA_PIXEL =Bit31∪Bit51

[0061] Among them, Cloud QA_PIXEL Represents the cloud noise pixel in the QA60 band; Bit31 indicates that the third bit of the current pixel in the QA_PIXEL band is 1, and Bit51 indicates that the fifth bit of the current pixel in the QA_PIXEL band is 1.

[0062] It should be noted that when the third bit of the QA_PIXEL band of the Landsat image is 1, it indicates that there is cloud shadow noise in the current pixel, and when the fifth bit is 1, it indicates that there is thick cloud noise in the current pixel.

[0063] The cloud removal image is analyzed using the improved normalized difference water index, specifically:

[0064] I MNDWI =(R Green -R SWIR ) / (R Green +R SWIR )

[0065] In the above formula, R Green is the reflectivity of green light band; R SWIR is the reflectivity in the shortwave infrared band.

[0066] It should be noted that the water body identification in the present invention is an improved normalized difference water index that is optimized on the basis of NDWI, replacing the near-infrared band with the mid-infrared band, which significantly improves the accuracy of water body identification in urban areas. In addition, MNDWI can also capture more subtle information such as suspended matter distribution and water quality changes.

[0067] The existing calculation formula for the normalized difference water index is as follows:

[0068] I NDWI =(R Green -R NIR ) / (R Green +R NIR )

[0069] Where R Green is the reflectivity of green light band; R NIR It is the reflectance of the near infrared band. When the NDWI value is greater than 0, it usually indicates that the area is a water body. The range of NDWI is [-1,1]. The closer the value is to 1, the more likely it is a water body.

[0070] The phytoplankton index method is used to identify the blue algae bloom information in the water area, and the expression is:

[0071] FAI=R NIR -R NIR '

[0072] R NIR ′=R RED +(R SWIR -R RED )×(λ NIR -λ RED ) / (λ SWIR -λ RED )

[0073] In the above formula, FAI is phytoplankton, R RED , R NIR , R SWIR Corresponding to the red, near infrared, and short infrared bands, λ RED、 λ NIR、 λ SWIR Corresponding to the central wavelengths of red light, near infrared, and short infrared bands respectively; R NIR ′ is the interpolated reflectance, that is, linear interpolation is used between the red band and the short-wave infrared band to obtain the reflectance information in the near-infrared band. The band λ involved in the calculation of Landsat 5-TM RED =660nm,λ NIR =839nm,λ SWIR =1678nm; Landsat 7-ETM+band λ involved in calculation RED =661nm,λ NIR =835nm,λ SWIR =1650nm; Band λ of Landsat 8 / 9-OLI involved in the calculation RED =655nm,λ NIR =865nm,λ SWIR =1610nm; Sentinel 2-MSI participates in the calculation of the band λ RED =665nm,λ NIR =842nm,λ SWIR =1610nm.

[0074] The threshold variation range of the water body and the threshold variation range of the cyanobacteria bloom are specifically as follows: determining by the Otsu threshold method has significant advantages in processing complex and changeable water body and algal bloom remote sensing images, thereby achieving optimal segmentation of the image.

[0075] The selecting of remote sensing images within the study area specifically includes: obtaining high-quality, cloudless or lightly clouded remote sensing images within the study area.

[0076] Finally, the calculation results were visualized using ArcMap 10.8 software. In the calculated image, the area covered by the grid identified as algal bloom is the algal bloom outbreak area. In order to gain a deeper understanding of the distribution characteristics of algal blooms, the present invention further performs overlay analysis on the grids identified as algal blooms in each image to construct an algal bloom outbreak frequency distribution map. The more grid overlays in the area in the map, the higher the frequency of algal bloom outbreaks in the area.

[0077] The present invention also provides a system for identifying blue algae blooms in narrow and long rivers based on remote sensing data, comprising:

[0078] The data processing module uses the CEE cloud platform to select remote sensing images within the study area and perform preprocessing. The cloud threshold is set, and the cloud layer in the preprocessed remote sensing image is detected within the cloud threshold using Sentinel-2 satellite images and Landsat satellite images. Based on the detection results, cloud pixels are removed from the preprocessed remote sensing image to obtain a de-clouded image.

[0079] The recognition module uses the improved normalized difference water index to analyze the de-clouded image and determine the threshold change range of the water body in each image of the de-clouded image to realize the recognition of the water body area;

[0080] The phytoplankton index method is used to identify the cyanobacteria bloom information in the water area, and the threshold change range of cyanobacteria bloom in each image is determined to realize the identification of the distribution characteristics of cyanobacteria bloom.

[0081] The visualization module visualizes the recognition results.

[0082] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a method for identifying blue algae bloom information in a narrow and long river based on remote sensing data.

[0083] Example 1

[0084] The data sources, steps and technical effects of the method for identifying blue algae blooms in narrow and long rivers based on remote sensing data in the present invention are illustrated by examples:

[0085] (1) Using the Google Earth Engine (GEE) platform, we screened and acquired high-quality, cloud-free or lightly clouded remote sensing image data within the study area, including multi-source high-resolution satellite images such as the Landsat series (Landsat 5-TM, Landsat 7-ETM+, Landsat 8-OLI, and the latest Landsat 9-OLI) and Sentinel 2-MSI.

[0086] In view of the dynamic changes of water bodies in narrow and long rivers, daily Landsat and Sentinel 2-MSI images were selected to ensure data capture with high temporal resolution. In the period when Sentinel 2-MSI data was missing (such as before 2018), monthly images of the Landsat series were cleverly used to maintain the continuity and integrity of the data series.

[0087] (2) For the Landsat Collection 2 (C2) level data and Sentinel-2 Level-2A (L2A) level data that have been preliminarily processed, the focus is on subsequent processing of cloud masking and water body identification. By setting the cloud amount threshold and using the QA60 band and QA_PIXEL band for fine cloud detection operations, the cloud-covered area in the image is removed to improve data quality. The normalized difference water index (NDWI) and improved normalized difference water index (MNDWI) are used to accurately extract the water body part, providing high-quality basic data for subsequent algal bloom identification.

[0088] (3) The floating algae index (FAI) is used as the main spectral index for identifying cyanobacterial blooms. This index is insensitive to changes in environmental and observation conditions and is not easily affected by clouds, which improves the accuracy and reliability of monitoring. By calculating the FAI algae sensitivity index, the abundance and status of algal blooms in water bodies are quantitatively evaluated, effectively distinguishing clean water bodies from algal bloom water bodies. The Otsu threshold method is innovatively applied to the setting of spectral index thresholds, realizing dynamic adjustment of the index threshold, and adaptively determining the optimal threshold based on the grayscale distribution characteristics of the image to achieve accurate extraction of water body and algal bloom information.

[0089] The technical solution of the present invention also has the following technical effects:

[0090] 1. Improved the applicability and flexibility of algal bloom information identification:

[0091] Existing remote sensing technologies for identifying algal blooms work well in open waters such as lakes or large reservoirs, but when applied to narrow and long rivers, the recognition accuracy is low due to the unique geographical morphology and water flow characteristics of the rivers. When identifying cyanobacteria blooms, the current remote sensing technology relies heavily on expert experience and statistical analysis to determine the threshold, lacking universality and flexibility, and having difficulty adapting to the rapid changes in different river environments.

[0092] The present invention adopts adaptive algorithms and models to optimize the data processing process according to the characteristics of narrow and long rivers, and uses the Otsu threshold method to dynamically adjust the phytoplankton index threshold, reducing the interference of multiple factors such as other suspended matter in the water body, complex bottom reflections and variable water flow conditions, so that it can automatically adapt to the dynamic changes of different environmental conditions in the river, thereby improving the applicability and flexibility of identification.

[0093] 2. Improved the efficiency of algal bloom information identification:

[0094] As current technology lacks specialized algorithms and model support for the characteristics of long and narrow rivers, processing remote sensing data of such rivers is not only time-consuming and requires large computing resources, but is also limited by the high cost and complex processing procedures of high-resolution data, making it difficult to guarantee the accuracy of algal bloom information identification.

[0095] The present invention utilizes the advanced cloud processing platform Google Earth Engine, relying on its huge data volume and efficient data analysis capabilities, to screen out multi-source high-resolution satellite images (such as the Landsat series and Sentinel 2-MSI), combining the advantages of data at different resolutions, and reducing direct reliance on high-cost high-resolution remote sensing data. For example, in the time period when Sentinel 2-MSI data is missing (such as before 2018), the monthly scale images of the Landsat series are cleverly used to maintain the continuity and integrity of the data time series, thereby ensuring the continuity and integrity of long-term series data monitoring. This method of combining multi-source data not only enriches the data basis of the research, but also further improves the accuracy and reliability of the final results. In addition, the present invention also specifically adopts adaptive algorithms and models to optimize the data processing process, reduce the consumption of computing resources, and improve the speed and accuracy of algal bloom identification.

[0096] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for identifying blue algae bloom information in narrow and long rivers based on remote sensing data, characterized in that: The following steps are involved: Using the CEE cloud platform, remote sensing images within the study area were selected and preprocessed; The cloud threshold is set, and the cloud layer in the pre-processed remote sensing image is detected using Sentinel-2 satellite imagery and Landsat satellite imagery within the cloud threshold. Based on the detection results, cloud pixels are removed from the pre-processed remote sensing image to obtain a de-clouded image. The cloud-removed image is analyzed using the improved normalized difference water index to determine the threshold variation range of water bodies in each scene of the cloud-removed image, so as to realize the identification of water body areas. The phytoplankton index method is used to identify the cyanobacteria bloom information in the water area, and the threshold change range of cyanobacteria bloom in each image is determined to realize the identification of the distribution characteristics of cyanobacteria bloom. Visualize the recognition results.

2. The method for identifying blue algae blooms in narrow and long rivers based on remote sensing data according to claim 1, characterized in that: The cloud layer in the pre-processed remote sensing image is detected within the cloud amount threshold using the Sentinel-2 satellite image and the Landsat satellite image respectively; wherein the cloud layer in the pre-processed remote sensing image is detected within the cloud amount threshold using the Sentinel-2 satellite image, specifically: The QA60 band of the Sentinel-2 satellite image is used to detect clouds in the preprocessed remote sensing image within the cloud cover threshold. The expression is as follows: Cloud QA60 =Bit101∪Bit111 Among them, Cloud QA60 Represents the cloud noise pixel in the QA60 band; Bit101 means that the 10th bit of the current pixel in the QA60 band is 1, and Bit111 means that the 11th bit of the current pixel in the QA60 band is 1.

3. The method for identifying blue algae blooms in narrow and long rivers based on remote sensing data according to claim 1, characterized in that: The method of detecting the cloud layer in the preprocessed remote sensing image by using the Sentinel-2 satellite image and the Landsat satellite image within the cloud amount threshold is as follows: The QA_PIXEL band in the Landsat image is used to detect the cloud layer in the preprocessed remote sensing image within the cloud amount threshold. The expression is as follows: Cloud QA_PIXEL =Bit31∪Bit51 Among them, Cloud QA_PIXEL Represents the cloud noise pixel in the QA60 band; Bit31 indicates that the third bit of the current pixel in the QA_PIXEL band is 1, and Bit51 indicates that the fifth bit of the current pixel in the QA_PIXEL band is 1.

4. The method for identifying blue algae blooms in narrow and long rivers based on remote sensing data according to claim 1, characterized in that: The improved normalized difference water index is used to analyze the cloud removal image, and the expression is: I MNDWI =(R Green -R SWIR ) / (R Green +R SWIR ) In the above formula, R Green is the reflectivity of green light band; R SWIR is the reflectivity in the shortwave infrared band.

5. The method for identifying blue algae blooms in narrow and long rivers based on remote sensing data according to claim 1, characterized in that: The phytoplankton index method is used to identify the blue algae bloom information in the water area, and the expression is: FAI=R NIR -R NIR ′ R NIR ′=R RED +(R SWIR -R RED )×(λ NIR -l RED ) / (λ SWIR -l RED ) In the above formula, FAI is phytoplankton, R RED , R NIR , R SWIR Corresponding to the red, near infrared, and short infrared bands, λ RED、 λ NIR、 λ SWIR Corresponding to the central wavelengths of red light, near infrared, and short infrared bands respectively; R NIR ′ is the interpolated reflectance, that is, linear interpolation is used between the red band and the short-wave infrared band to obtain the reflectance information in the near-infrared band.

6. The method for identifying blue algae blooms in narrow and long rivers based on remote sensing data according to claim 1, characterized in that: The threshold variation range of the water body and the threshold variation range of the cyanobacteria bloom are specifically determined by the Otsu threshold method.

7. The method for identifying blue algae blooms in narrow and long rivers based on remote sensing data according to claim 6, characterized in that: The determination is performed by the Otsu threshold method, specifically: The slope and statistical analysis methods were used to determine the threshold variation range of cyanobacteria bloom in each image.

8. The method for identifying blue algae blooms in narrow and long rivers based on remote sensing data according to claim 1, characterized in that: The selecting of remote sensing images within the study area specifically includes: obtaining high-quality, cloudless or lightly clouded remote sensing images within the study area.

9. A system for identifying blue algae blooms in narrow and long rivers based on remote sensing data, used to implement the method for identifying blue algae blooms in narrow and long rivers based on remote sensing data according to any one of claims 1 to 8, characterized in that: include: The data processing module uses the CEE cloud platform to select remote sensing images within the study area and perform preprocessing; The cloud threshold is set, and the cloud layer in the pre-processed remote sensing image is detected using Sentinel-2 satellite imagery and Landsat satellite imagery within the cloud threshold. Based on the detection results, cloud pixels are removed from the pre-processed remote sensing image to obtain a de-clouded image. The recognition module uses the improved normalized difference water index to analyze the de-clouded image and determine the threshold change range of the water body in each scene of the de-clouded image to realize the recognition of the water body area; The phytoplankton index method is used to identify the cyanobacteria bloom information in the water area, and the threshold change range of cyanobacteria bloom in each image is determined to realize the identification of the distribution characteristics of cyanobacteria bloom. The visualization module visualizes the recognition results.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for identifying blue algae bloom information in narrow and long rivers based on remote sensing data according to any one of claims 1 to 8 are implemented.

Images