CN-122023498-A - Method, device, equipment and storage medium for determining water level threshold of bird breeding habitat

CN122023498ACN 122023498 ACN122023498 ACN 122023498ACN-122023498-A

Abstract

The invention provides a method, a device, equipment and a storage medium for determining a water level threshold value of a bird breeding habitat, which belong to the technical fields of ecological remote sensing, lake wetland monitoring and water bird habitat management, and the method comprises the steps of acquiring multi-source remote sensing images in a breeding period time window and carrying out cloud/shadow elimination and synthesis treatment; the method comprises the steps of carrying out water body identification on images, counting the occurrence proportion of water bodies in a time window, constructing a stable water body mask in a propagation period, inverting the stable water body in a lake region range to obtain a non-water body region, carrying out connected domain analysis, extracting candidate patches, removing the shore-connection patches connected with a lake shoreline or touching the boundary of the lake region, reserving the independent patches completely surrounded by the stable water body as a center island, improving the robustness of a result through morphological post-treatment and minimum drawing unit control, fusing bird propagation monitoring information to construct a comprehensive suitability objective function, and determining a suitable water level of a propagation habitat or a threshold area of the lake surface.

Inventors

CUI GENG
ZHANG LIU
LIU YAN
TONG SHOUZHENG
AN YU
ZHANG MINGYE

Assignees

中国科学院东北地理与农业生态研究所

Dates

Publication Date: 20260512
Application Date: 20260120

Claims (8)

1. A method for determining a threshold level of a bird breeding habitat, comprising: The method comprises the steps of selecting a target closed lake as a research area, collecting multisource remote sensing data, acquiring contemporaneous water level observation data, lake boundary vectors and bird reproduction monitoring data, and constructing a data set covering remote sensing, hydrology and ecological factors; for multisource remote sensing image data synthesized in each year, a multisource data fusion water body identification method is adopted for identification, and a stable water body mask in the propagation period is constructed; based on the water body regions extracted year by year, identifying independent water body plaques, and using a stable water body mask to carry out space constraint, wherein only non-water body plaques within the range of the stable water body mask are reserved and used as a candidate plaque aggregate of the annual lake center island; removing the plaque connected with the shoreline or touching the boundary, reserving an independent lake center island completely surrounded by the water body, and identifying a lake center island distribution area; morphological processing is carried out on the identified lake center island to smooth the boundary, and a vector sequence of the lake center island is obtained after vectorization; selecting a core index, carrying out normalization processing on the core index, and mapping the core index to a [0,1] interval; Determining the weight of the core index in comprehensive evaluation; fusing the weight of the processed core index and the weight of the core index in comprehensive evaluation to construct a comprehensive suitability index; Determining an island quantity threshold value for supporting effective propagation based on long-term observation data, screening out up-to-standard years, calculating the comprehensive fitness index corresponding to the up-to-standard years, and establishing the lower limit of the comprehensive fitness index as a comprehensive fitness threshold value; and calculating the water level threshold of the bird reproduction habitat by backtracking hydrologic data of the year reaching the standard based on the comprehensive suitability threshold.
2. The method of claim 1, wherein the constructing the propagation phase stable water mask for each year of the synthesized multi-source remote sensing image data by using a water identification method of multi-source data fusion comprises: Carrying out pixel-by-pixel water body identification by combining NDWI with SAR data; the frequency of each pixel determined to be a body of water during the study period is counted as follows: Wherein, the The occurrence frequency of the water body of the pixel is represented by the probability that the pixel is identified as the water body in the research period, The number of years for which the pixel is determined to be the body of water, i.e., the body of water identifies the number of years, To study total years; The frequency threshold alpha is selected by an accuracy verification method, namely different alpha values are respectively taken to generate a stable water mask, And comparing with a visual interpretation result of the high-resolution historical image, selecting an alpha value corresponding to the highest overall classification precision as a threshold value, and generating the stable water mask in the propagation period, wherein alpha is 0.6-0.9.
3. The method of claim 1, wherein the core metrics include island number, island area, lake water level, bird breeding nest number.
4. The method of claim 1, wherein determining the weight of the core indicator in the composite evaluation comprises: Calculating the information entropy of each core index and measuring the discrete degree of data; Calculating weights according to the following formula; Wherein, the For the weight value of the i-th index, For the information entropy of the ith index, the discrete degree of the index data is measured, Total number of indicators for participation in the evaluation.
5. The method of claim 1, wherein the formula for constructing the integrated fitness index is: Wherein, the In order to synthesize a habitat fitness index, For the weight value of each index, For the normalized values of the various indices, The number of indexes for participation in the evaluation.
6. A bird breeding habitat water level threshold determination device, comprising: the acquisition module is used for selecting a target closed lake as a research area and collecting multi-source remote sensing data, acquiring contemporaneous water level observation data, lake boundary vectors and bird reproduction monitoring data, and constructing a data set covering remote sensing, hydrology and ecological factors; The construction module is used for constructing a stable water mask in the propagation period by adopting a water identification method of multi-source data fusion for multi-source remote sensing image data synthesized in each year; the candidate module is used for identifying independent water patches based on the water areas extracted year by year, and then using the stable water mask to carry out space constraint, and only retaining non-water patches within the range of the stable water mask as a candidate patch set of the annual lake center island; The identification module is used for removing patches connected with the shoreline or touching the boundary, reserving independent lake center islands completely surrounded by water body and identifying a lake center island distribution area; the vectorization module is used for carrying out morphological processing on the identified lake center island to smooth the boundary, and obtaining a lake center island vector sequence after vectorization; The processing module is used for selecting a core index, carrying out normalization processing on the core index and mapping the core index to a [0,1] interval; the evaluation module is used for determining the weight of the core index in comprehensive evaluation; The construction module is used for fusing the weight of the processed core index and the weight of the core index in the comprehensive evaluation to construct a comprehensive suitability index; The first calculation module is used for determining the number threshold value of islands which support effective reproduction based on long-term observation data, screening out the up-to-standard years, calculating the comprehensive fitness index corresponding to the up-to-standard years, and establishing the lower limit of the comprehensive fitness index as the comprehensive fitness threshold value; and the second calculation module is used for calculating the water level threshold of the bird reproduction habitat by backtracking hydrologic data of the year reaching the standard based on the comprehensive suitability threshold.
7. An electronic device comprising a memory for storing a computer program and a processor for implementing the method of any one of claims 1-5 when the computer program is executed.
8. A storage medium having a computer program stored thereon, which, when executed by a processor, implements the method according to any of claims 1-5.

Description

Method, device, equipment and storage medium for determining water level threshold of bird breeding habitat Technical Field The application relates to the technical field of ecological remote sensing, lake wetland monitoring and water bird habitat management, in particular to a method, a device, equipment and a storage medium for determining a water level threshold of a bird reproduction habitat. Background The water level fluctuation of the closed lake is obvious under the actions of seasonal replenishment, evaporation, stormy waves transportation and the like, and the fluctuation of the lake surface can cause the formation, expansion, connection Liu Hua or inundation of a center island of the lake, thereby influencing the isolation, the safety and the available area of a water bird breeding place. The existing lake center island extraction is mostly dependent on manual interpretation or single-period image threshold segmentation, is easily affected by cloud cover, sensor difference, water turbidity, algal bloom and the like, and is insufficient in annual consistency and repeatability, meanwhile, the traditional method is often only used for providing island boundaries, and the 'change of the pattern of the lake center island' is difficult to be further quantized into a reproductive habitat threshold target executable by water level regulation. Therefore, an automatic and standardized lake center island identification method for closed lakes is needed, and an interpretable and executable bird breeding habitat water level threshold determination method is formed on the basis of the method, so that basis is provided for ecological water supply scheduling, protection area patrol and habitat management. Disclosure of Invention In view of the above, embodiments of the present invention are directed to providing a method, an apparatus, a device, and a storage medium for determining a threshold level of a bird breeding habitat, which are used for providing basis for ecological water supply scheduling, protection area patrol, and habitat management. In a first aspect, the invention provides a method for determining a threshold level of a bird breeding habitat comprising: The method comprises the steps of selecting a target closed lake as a research area, collecting multisource remote sensing data, acquiring contemporaneous water level observation data, lake boundary vectors and bird reproduction monitoring data, and constructing a data set covering remote sensing, hydrology and ecological factors; for multisource remote sensing image data synthesized in each year, a multisource data fusion water body identification method is adopted for identification, and a stable water body mask in the propagation period is constructed; based on the water body regions extracted year by year, identifying independent water body plaques, and using a stable water body mask to carry out space constraint, wherein only non-water body plaques within the range of the stable water body mask are reserved and used as a candidate plaque aggregate of the annual lake center island; removing the plaque connected with the shoreline or touching the boundary, reserving an independent lake center island completely surrounded by the water body, and identifying a lake center island distribution area; morphological processing is carried out on the identified lake center island to smooth the boundary, and a vector sequence of the lake center island is obtained after vectorization; selecting a core index, carrying out normalization processing on the core index, and mapping the core index to a [0,1] interval; Determining the weight of the core index in comprehensive evaluation; fusing the weight of the processed core index and the weight of the core index in comprehensive evaluation to construct a comprehensive suitability index; Determining an island quantity threshold value for supporting effective propagation based on long-term observation data, screening out up-to-standard years, calculating the comprehensive fitness index corresponding to the up-to-standard years, and establishing the lower limit of the comprehensive fitness index as a comprehensive fitness threshold value; and calculating the water level threshold of the bird reproduction habitat by backtracking hydrologic data of the year reaching the standard based on the comprehensive suitability threshold. Further, the identifying the multisource remote sensing image data synthesized in each year by adopting a multisource data fusion water body identification method, and constructing a stable water body mask in the propagation period comprises the following steps: Carrying out pixel-by-pixel water body identification by combining NDWI with SAR data; the frequency of each pixel determined to be a body of water during the study period is counted as follows: Wherein, the The occurrence frequency of the water body of the pixel is represented by the probability that the pixel is identified as the w