Regional ecological environment quality assessment method and device, equipment and storage medium

Abstract

The embodiment of the invention relates to the field of ecological environment evaluation, in particular to a regional ecological environment quality evaluation method and device, equipment and a storage medium, and the method comprises the steps: determining a plurality of element images according to remote sensing data of a target region; according to a category image of the target area, determining a convolution scale corresponding to each pixel position in each element image to obtain a convolution scale map of each element image, the category image being used for representing a landform category of each pixel position in the target area, and according to the corresponding convolution scale map, performing convolution processing on each element image to obtain an element convolution result. And generating an evaluation report of the target area according to the convolution result of each element. According to the embodiment of the invention, the fidelity of the spatial geometric morphology and the expression stability of the ecological elements can be improved according to different landform categories in the target region, and the convolution kernels of different pixel positions are specifically matched for different element images according to the landform categories, so that the interpretability, the reliability and the cross-region applicability of the evaluation result are integrally improved.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the field of ecological environment assessment, and relate to but are not limited to a regional ecological environment quality assessment method and device, equipment, and storage medium. BACKGROUND

[0002] Remote sensing ecological index (RSEI) is a commonly used method for comprehensive evaluation of regional ecological environment quality using remote sensing data. The typical RSEI four sub-indexes are composed of greenness (commonly used NDVI), wetness (commonly used Tasseled Cap Wetness / TCW or other wetness index), dryness (commonly used NDBSI combined by bare soil index and built-up area index), and heat (commonly used land surface temperature LST). Usually, the four types of factors are normalized and unified in positive and negative directions, and the first principal component is extracted as a comprehensive ecological index through principal component analysis (PCA). RSEI is widely used in urban ecological quality evaluation, land use / cover change impact assessment, regional ecological red line monitoring, and temporal ecological evolution analysis due to its simple index structure and wide application range. When the related technology extracts features of the above four types of factors through convolution, there are defects such as scale mismatch and boundary distortion caused by unified convolution kernel and lack of spatial processing under land class conditions, and fusion mismatch of multi-source resolution and heterogeneous data, resulting in unstable fusion of heat and spectral factors. Further, the evaluation results of RSEI are not accurate and not interpretable. SUMMARY

[0003] Therefore, the regional ecological environment quality assessment method and device, equipment, and storage medium provided by the embodiments of the present application improve the interpretability, reliability, and cross-regional applicability of the ecological evaluation results.

[0004] The regional ecological environment quality assessment method and device, equipment, and storage medium provided by the embodiments of the present application are implemented as follows: In an aspect of the embodiments of the present application, a regional ecological environment quality assessment method is provided, which includes: determining a plurality of element images according to remote sensing data of a target region; determining a convolution scale corresponding to each pixel position in each element image according to a category image of the target region, to obtain a convolution scale map of each element image, the category image being used to represent a landform category of each pixel position in the target region; performing convolution processing on each element image according to the corresponding convolution scale map to obtain element convolution results; generating an evaluation report of the target region according to the element convolution results.

[0005] In a possible implementation, the plurality of element images are determined according to remote sensing data of the target region, including: performing image preprocessing on the remote sensing data of the target region; determining element images corresponding to a plurality of preset element dimensions according to the remote sensing data after image preprocessing.

[0006] In a possible implementation, the convolution scale corresponding to each pixel position in each element image is determined according to a category image of the target region, to obtain a convolution scale map of each element image, including: determining a convolution scale set corresponding to a landform category included in the category image; determining a landform category of each pixel position in each element image according to the category image; for each element image, determining a convolution scale corresponding to each pixel position according to a corresponding scale selection rule and the landform category of the pixel position in the convolution scale set, to obtain a convolution scale map of the element image.

[0007] In a possible implementation, for each element image, a convolution scale corresponding to each pixel position is determined according to a corresponding scale selection rule and the landform category of the pixel position in the convolution scale set, to obtain a convolution scale map of the element image, including: for each element image, selecting a candidate scale of each pixel position in the convolution scale set corresponding to the pixel position; determining a convolution scale of each pixel position in the convolution scale set according to the candidate scale corresponding to the pixel position according to a corresponding scale selection rule of the element image, to obtain a convolution scale map of the element image.

[0008] In a possible implementation, the element images include a greenness image, a dryness image, a wetness image, and a heat image, and for each element image, a candidate scale of each pixel position in the convolution scale set corresponding to the pixel position is selected, including: for the greenness image, the dryness image, and the wetness image, selecting a preset reference scale as the candidate scale of each pixel position in the convolution scale set corresponding to the pixel position; for the heat image, selecting a candidate scale of each pixel position in the convolution scale set corresponding to the pixel position according to a preset spatial support condition.

[0009] In a possible implementation, for each element image, a convolution scale corresponding to each pixel position is determined according to a corresponding scale selection rule and the landform category of the pixel position in the convolution scale set, to obtain a convolution scale map of the element image, further including: performing consistency correction on the convolution scale map corresponding to each element image.

[0010] In a possible implementation, the evaluation report of the target region is generated according to the element convolution results, and the evaluation report of the target region includes: Perform principal component analysis based on the element convolution results to obtain a comprehensive ecological quality score; Generate the evaluation report of the target region according to the element images, the element convolution results, and the comprehensive ecological quality scores.

[0011] Another aspect of the embodiment of the application also provides a device for evaluating the ecological environment quality of a region, and the device includes: An image acquisition module configured to determine a plurality of element images according to remote sensing data of a target region; A scale determination module configured to determine a convolution scale corresponding to each pixel position in each element image according to a category image of the target region, to obtain a convolution scale map of each element image, and the category image is used to represent the landform category of each pixel position in the target region; A convolution processing module configured to perform convolution processing on each element image according to the corresponding convolution scale map to obtain element convolution results; A quality evaluation module configured to generate an evaluation report of the target region according to the element convolution results.

[0012] In a possible implementation, the image acquisition module is further configured to: Perform image preprocessing on the remote sensing data of the target region; Determine element images corresponding to a plurality of preset element dimensions according to the remote sensing data after image preprocessing.

[0013] In a possible implementation, the scale determination module is further configured to: Determine a convolution scale set corresponding to the landform category included in the category image; Determine the landform category of each pixel position in each element image according to the category image; For each element image, determine the convolution scale corresponding to each pixel position according to the corresponding scale selection rule and the landform category of the pixel position in the convolution scale set, to obtain the convolution scale map of the element image.

[0014] In a possible implementation, the scale determination module is further configured to: For each element image, select a candidate scale of each pixel position in the convolution scale set corresponding to the pixel position; According to the scale selection rule corresponding to the element image, determine the convolution scale in the convolution scale set according to the candidate scale corresponding to each pixel position, to obtain the convolution scale map of the element image.

[0015] In a possible implementation, the element images include a greenness image, a dryness image, a wetness image, and a heat image, and the scale determination module is further configured to: For the greenness image, the dryness image, and the wetness image, a preset reference scale is selected as a candidate scale of each pixel position from a set of convolution scales corresponding to the pixel position; For the heat image, a candidate scale of each pixel position is selected from a set of convolution scales corresponding to the pixel position according to a preset spatial support condition.

[0016] In a possible implementation, the scale determination module is further configured to: perform consistency correction on the convolution scale map corresponding to each element image.

[0017] In a possible implementation, the quality assessment module is further configured to: perform principal component analysis based on the element convolution results to obtain a comprehensive ecological quality score; generate an evaluation report of the target region according to the element images, the element convolution results, and the comprehensive ecological quality scores.

[0018] The electronic device provided in the embodiments of the present application includes a memory and a processor, the memory stores a computer program that can run on the processor, and the processor implements the method provided in the embodiments of the present application when executing the program.

[0019] The computer readable storage medium provided in the embodiments of the present application stores a computer program, and the computer program is executed by a processor to implement the method provided in the embodiments of the present application.

[0020] In the embodiments of the present application, a plurality of element images are determined according to remote sensing data of a target region. Convolution scales corresponding to each pixel position in each element image are determined according to a category image of the target region, to obtain a convolution scale map of each element image, the category image is used to represent a landform category of each pixel position in the target region, each element image is processed by convolution according to the corresponding convolution scale map, to obtain element convolution results. An evaluation report of the target region is generated according to the element convolution results. The embodiments of the present application can improve the fidelity of spatial geometric shapes and the expression stability of ecological elements according to different landform categories in the target region, and match different convolution kernels for different pixel positions for different element images according to the landform categories, to improve the explainability, reliability, and cross-region applicability of the evaluation results as a whole. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. Obviously, the drawings in the following description only constitute some embodiments of the present application. For those skilled in the field, other drawings can also be obtained from these drawings without any creative effort.

[0022] Figure 1 A flow chart of a regional ecological environment quality assessment method according to an embodiment of the present application is shown; Figure 2 A schematic diagram of a regional ecological environment quality assessment process according to an embodiment of the present application is shown; Figure 3 A schematic diagram of a regional ecological environment quality assessment device according to an embodiment of the present application is shown; Figure 4 A schematic diagram of an electronic device according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0023] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the following will further describe the specific technical solutions of the present application with reference to the drawings in the embodiments of the present application. The following embodiments are used to illustrate the present application, but not to limit the scope of the present application.

[0024] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application, and are not intended to limit the present application.

[0025] In the following description, "some embodiments" are related to a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict.

[0026] It should be noted that the terms "first, second, third" used in the embodiments of the present application are used to distinguish similar or different objects, and do not represent a specific order of the objects. It can be understood that "first, second, third" can be interchanged in a specific order or sequence as allowed, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.

[0027] The regional ecological environment quality evaluation method of the embodiments of the present application can be executed by any electronic device, which can include but is not limited to a mobile phone, a wearable device (such as a smart watch, a smart bracelet, smart glasses, etc.), a tablet computer, a notebook computer, a vehicle-mounted terminal, a PC (Personal Computer), and the like. The functions implemented by the method can be realized by calling program codes by a processor in the electronic device. Of course, the program codes can be saved in a computer storage medium. Therefore, the electronic device at least includes a processor and a storage medium.

[0028] The regional ecological environment quality evaluation method of the embodiments of the present application can be used in any application scenario that needs to evaluate the ecological environment quality of a region. For example, the embodiments of the present application can be applied to evaluate the ecological quality of a city in the application scenario of city development and planning management. Alternatively, the embodiments of the present application can also be applied to dynamically monitor a nature reserve / national park in the application scenario of ecological system protection and restoration.

[0029] At present, the regional ecological environment quality evaluation is usually realized by calculating a remote sensing ecological index (RSEI) of a region. The remote sensing ecological index is used to comprehensively evaluate the ecological environment quality of a region by using remote sensing data. The typical RSEI four-type sub-indexes include greenness (commonly used NDVI), wetness (commonly used Tasseled Cap Wetness / TCW or other wetness indexes), dryness (commonly used NDBSI combined by a bare soil index and a built-up area index), and heat (commonly used land surface temperature LST). The four types of factors are usually normalized and unified in positive and negative directions, and the first principal component is extracted as a comprehensive ecological index by principal component analysis (PCA). The RSEI is widely used in scenes such as city ecological quality evaluation, land use / cover change impact assessment, regional ecological red line monitoring, and time series ecological evolution analysis because of its simple index structure and wide application range.

[0030] The data sources of the remote sensing ecological index usually include multispectral (such as Landsat, Sentinel-2) and thermal infrared (for LST inversion), and need to be preprocessed by atmospheric correction, cloud shadow mask, resampling and registration, time series synthesis, and the like. In order to suppress the strip, spot and random noise, the smoothing method (such as mean filtering, Gaussian filtering, fixed-size convolution / window operation) in the spatial dimension is usually used in the traditional process, and the window or convolution kernel size is usually unified globally.

[0031] In the related art, a large number of studies attempt to replace or extend four factors (such as introducing Albedo, soil moisture index, vegetation coverage, etc.), adjust weights (PCA variants, entropy weights, random forest weights), or perform RSEI calculation in sub-regions / sub-land classes (statistical fusion according to functional areas or land cover types). However, in spatial processing, most of the work still uses a globally unified smoothing or fixed convolution kernel size. In the field of remote sensing semantic segmentation and target detection, methods such as multi-scale convolution, dynamic convolution, deformable convolution, and selective convolution kernel are used to improve the adaptability of feature extraction. However, these methods are mostly used for feature learning in recognition and segmentation tasks, and are less directly embedded into the sub-index spatial processing and fusion of RSEI, and there is no standard process of "adaptive selection of convolution kernel size according to land cover categories".

[0032] At the same time, LST often has resolution differences with multispectral data, and the traditional method uses uniform resampling and uniform window smoothing, which lacks targeted processing of scale differences and noise characteristics of different land classes.

[0033] Therefore, the related art has the following technical problems in evaluating the ecological environment quality by calculating the remote sensing ecological index of a region: Uniform convolution kernel and lack of land class conditions in spatial processing lead to scale mismatch and boundary distortion. Existing methods often use a single, globally uniform convolution kernel size for smoothing all land classes, ignoring the significant differences in spatial texture, structure scale, and noise spectrum of forests, grasslands, farmland, water bodies, and built-up areas. The result is that in high-texture and hard-boundary areas such as urban edges, street networks, shorelines, and waterfront transitions, a uniform kernel will over-smooth, blurring details, widening boundaries, and even causing "bleeding"; in high-frequency spot areas such as deserts and bare land, a uniform kernel is prone to under-smoothing, leaving salt and pepper noise and random fluctuations, which disrupts spatial continuity and the interpretability of ecological semantics; further, spatial convolution is not adaptive to land class conditions, and class boundaries and transition zones are prone to "mixed pixel" effects and boundary blurring, causing evaluation distortion, hot spot misjudgment, and local instability of RSEI in areas with strong heterogeneity.

[0034] The fusion mismatch of multi-source resolution and heterogeneous data is not effectively handled, resulting in unstable fusion of heat and spectral factors. RSEI usually fuses LST (thermal infrared) and multi-spectral sub-indexes (NDVI / humidity / dryness), which have systematic differences in spatial resolution, noise spectrum and sampling support (e.g., LST resolution is coarser, thermal noise and field of view effect are stronger). The unified spatial convolution cannot select the optimal smoothing scale of heat and spectral factors according to land types, nor can it align the "effective spatial support" of the two types of data, thereby introducing false differences and misaligned hot spots in the fusion stage, destroying the stability of the covariance structure, causing the PCA load direction to fluctuate, the principal component to be unstable, the comparability of time series and cross-regional comparison to decrease, and finally affecting the robustness and application credibility of RSEI.

[0035] Therefore, the technical problem solved by the embodiments of the present application is how to improve the interpretability, reliability and cross-regional applicability of the ecological evaluation results.

[0036] The regional ecological environment quality evaluation scheme of the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0037] Figure 1 A flowchart of a regional ecological environment quality evaluation method according to an embodiment of the present application is shown. As shown in Figure 1 The regional ecological environment quality evaluation method of the embodiments of the present application can include the following steps S10-S40.

[0038] For ease of description, the regional ecological environment quality evaluation method of the embodiments of the present application is described with an electronic device as the execution subject. It should be understood that the execution subject of the embodiments of the present application can also be a processor or a chip in the electronic device, and the embodiments of the present application are not limited in any way.

[0039] Step S10, determining a plurality of element images according to remote sensing data of a target region.

[0040] In one possible implementation, the electronic device can perform ecological environment quality evaluation on the target region by determining a remote sensing ecological index based on the remote sensing data of the target region. The remote sensing data can include a plurality of images of the target region collected by different wave bands at the same time, which are used to determine corresponding element images in a plurality of different element dimensions. Each element image corresponds to an element dimension and is used to represent the feature distribution of the target region in the element dimension.

[0041] Exemplarily, the remote sensing data in the embodiments of the present application can include visible light images, near-infrared light images, and thermal infrared light images of the target region, which can be obtained by Landsat 8 / 9 OLI / TIRS, Sentinel-2 with a matching LST product, or obtained by a land surface temperature inversion method with a matching TIRS (thermal infrared sensor). The multiple factor images can include multiple images used to calculate a remote sensing ecological index, that is, the factor dimensions of the embodiments of the present application can include greenness, dryness, wetness, and heat, and corresponding greenness images, dryness images, wetness images, and heat images are determined respectively.

[0042] In some embodiments, the remote sensing data may, in the acquisition process, have interference noise due to cloud cover and the like, and therefore the electronic device can first perform image preprocessing on the remote sensing data of the target region before determining the multiple factor images based on the remote sensing data, and then determine the factor images corresponding to the multiple preset factor dimensions according to the remote sensing data after image preprocessing.

[0043] Optionally, the preprocessing process can include radiation calibration, terrain correction, and atmospheric correction on the remote sensing data, and simultaneously locate the cloud cover area as an invalid area by cloud mask. For example, radiation calibration is performed by LEDAPS or LaSRC, terrain correction is performed by DEM, atmospheric correction is performed by Sen2Cor (Sentinel-2 dedicated atmospheric correction processor), and invalid area calibration is performed by Fmask. At the same time, the preprocessing process can also perform geometric fine correction and resampling to a unified resolution on each image data. Further, for the invalid area calibrated after cloud mask processing, it can be directly marked as invalid or interpolated and completed.

[0044] In some embodiments, after preprocessing the remote sensing data, the electronic device can directly determine the multiple factor images based on the factor images corresponding to each factor dimension according to a rule. For example, the rule for determining the greenness image (G) can be: NDVI = (NIR - Red) / (NIR + Red), where NIR is the near-infrared band and Red is the red band. The rule for determining the wetness image (W) can be: Tasseled Cap wetness component (Landsat coefficient) or using NDWI / modified wetness index; prefer Tasseled Cap W component to take into account physical meaning and stability. The rule for determining the dryness image (D) can be: NDBSI (combining bare soil index SI and built-up index IBI, or directly using NDBI and SI weighted average); ensure negative correlation with ecological quality. The rule for determining the heat image (L) can be: land surface temperature LST (estimated by TIRS single-window or split-window algorithm, combined with land surface emissivity).

[0045] In some embodiments, the electronic device can first determine the candidate element images corresponding to the multiple element dimensions by the above method, and then perform pre-standardization and polarity check on each candidate element image to obtain element images that pass the rapid quality inspection. The pre-standardization process can be normalization processing of each candidate element image to ensure that the pixel value of each pixel position is between 0 and 1. The polarity check process also includes polarity correction, which is used to check that the greenness image and the humidity image are positively correlated with ecological quality, and the aridity image and the heat image are negatively correlated with ecological quality. For element images that do not pass the polarity check, polarity correction can be performed by taking the inverse.

[0046] Step S20: determining a convolution scale corresponding to each pixel position in each of the element images according to the category image of the target area, to obtain a convolution scale map of each of the element images.

[0047] In a possible implementation, the electronic device obtains a category image of the target area, where each pixel in the category image is used to indicate the landform category of the corresponding position in the target area. The resolution of the category image can be the same as that of each element image, and is used to indicate the landform category corresponding to each pixel position in each element image, so as to determine the convolution scale corresponding to each pixel position in each element image according to the corresponding landform category, and obtain a convolution scale map of the element image.

[0048] In some embodiments, the electronic device can generate the category image of the target area by using a land feature classification network (such as DeepLabv3+, U-Net, SegFormer) or a traditional classification (random forest, SVM). The landform categories can include vegetation, water body, construction land, bare land / desert, wetland, farmland, forest land, road, and the like. Optionally, the electronic device can also calculate a boundary distance map according to the category image, and set a boundary bandwidth b (recommended 2-3 pixels); generate a mask for each category for intra-category k smoothing and water-land separation processing. The category division process can ensure that the overall accuracy of the category image is greater than or equal to 85%, and the boundary quality is good. Morphological refinement is provided for slender elements (roads, water channels).

[0049] Optionally, each landform category in the embodiments of the present application can be pre-configured with a corresponding set of convolution scales, which includes multiple candidate convolution scales. When determining the convolution scale map corresponding to the element image, the electronic device can match the corresponding set of convolution scales according to the landform category of each pixel position, and select a convolution scale in the corresponding set of convolution scales. That is, the electronic device can first determine the set of convolution scales corresponding to the landform category included in the category image, and then determine the landform category of each pixel position in each element image according to the category image. Then, for each element image, the corresponding convolution scale of each pixel position is determined according to the corresponding scale selection rule and the landform category of the pixel position in the set of convolution scales, and the convolution scale map of the element image is obtained.

[0050] For example, the landform categories in the embodiments of the present application can include construction land / road, water body, vegetation (forest / grass / farmland), bare land / barren desert, bare rock / rocky mountain, sand dune / wind-blown sand, snow / ice sheet, paddy field, intertidal zone / beach, and urban green space / park. Different landform categories can be pre-configured with corresponding sets of convolution scales, construction land / road: Kc={3, 5, 7}, water body: Kc={5, 7, 9}, vegetation (forest / grass / farmland): Kc={7, 9, 11}, bare land / barren desert: Kc={5, 7, 9}, bare rock / rocky mountain: Kc={3, 5, 7}, sand dune / wind-blown sand: Kc={7, 9, 11}, snow / ice sheet: Kc={7, 9, 11}, paddy field: Kc={5, 7, 9}, intertidal zone / beach: Kc={3, 5, 7}, urban green space / park: Kc={3, 5, 7}.

[0051] Optionally, the set of convolution scales of each landform category in the embodiments of the present application can also include at least one reference scale, which is one of the multiple convolution scales, and is used as the default convolution scale. For example, the reference scale of construction land / road can be determined as k_c=3 or 5 (in the case of preserving details and preventing color bleeding at the pixel position). The reference scale of water body can be determined as k_c=5 or 7 (in the case of internal uniformity and clear outline at the pixel position). The reference scale of vegetation (forest / grass / farmland) can be determined as k_c=7 or 9 (in the case of large patches and enhanced coherence at the pixel position). The reference scale of bare land / barren desert can be determined as k_c=5 or 7. The reference scale of bare rock / rocky mountain can be determined as k_c=5, which is used to ensure the coexistence of hard boundaries and prevent wide dragging. The reference scale of sand dune / wind-blown sand can be determined as k_c=9. The reference scale of snow / ice sheet can be determined as k_c=9. The reference scale of paddy field can be determined as k_c=7. The reference scale of intertidal zone / beach can be determined as k_c=5. The reference scale of urban green space / park can be determined as k_c=5.

[0052] In some embodiments, the electronic device can, after obtaining the set of convolution scales corresponding to each pixel position, first determine a convolution scale as a candidate scale from the set, and then correct the candidate scale based on a preset scale selection rule to accurately determine the corresponding convolution scale. That is, for each element image, the electronic device selects a candidate scale for each pixel position from the set of convolution scales corresponding to the pixel position. Then, according to the scale selection rule corresponding to the element image, the electronic device determines the convolution scale for each pixel position from the set of convolution scales based on the candidate scale for the pixel position, to obtain the convolution scale map of the element image.

[0053] Optionally, for different types of element images, the electronic device can directly select a reference scale as the candidate scale from the set of convolution scales corresponding to each pixel position, or select the candidate scale according to other rules. For example, in the case where the element images in the embodiments of the present application include greenness images, dryness images, humidity images, and heat images, for greenness images, dryness images, and humidity images, the electronic device selects a preset reference scale as the candidate scale for each pixel position from the set of convolution scales corresponding to the pixel position. For heat images, the electronic device selects the candidate scale for each pixel position from the set of convolution scales corresponding to the pixel position according to a preset spatial support condition.

[0054] Further, the spatial support condition in the embodiments of the present application can include first calculating q = r_LST / r_MS, where r_MS is the target grid resolution and r_LST is the native resolution of the heat image, and then determining that the candidate scale covers a neighborhood of approximately q x q (example: r_LST = 100 m, r_MS = 30 m, q ≈ 3.3 → k takes 9 or 11). Then, one of the above two scales is further selected through a simple adaptive and boundary protection method. The processing process of the simple adaptive method can be to calculate the local variance or entropy (recommended 5 x 5 window, set threshold according to intra-class statistical quantile), take the 25% quantile q25 and the 75% quantile q75 within the class. If the local variance ≤ q25 → use the larger kernel; ≥ q75 → use the smaller kernel; and if it is between them, use the reference scale. The processing process of the boundary protection method can include using the class boundary distance d and the boundary bandwidth b (2-3 pixels), d ≤ b → forcedly use the smallest kernel k_min in Kc; d > b → select k_c or fine-tune according to the above rule. Further, 3 x 3 median smoothing is performed on the pixel level to avoid jumps and ensure that the four-factor kernel scale of the same pixel in the corresponding set of convolution scales does not differ by more than one grade; and k_LST is forced to be no less than the other factors.

[0055] In other embodiments, after determining the candidate scale for each pixel position in the element image, the electronic device can correct the candidate scale through the scale selection rule corresponding to each element image to obtain the corresponding convolution scale.​

[0056] Exemplarily, in the case that the factor images of the embodiments of the present application include a greenness image, a dryness image, a wetness image, and a heat image, the scale selection rule of the greenness image can include: in the case that the pixel position belongs to the high-NDVI and low-texture vegetation inner area, selecting the up one step convolution scale in the convolution scale set based on the candidate scale; in the case that the pixel position belongs to the urban green land, broken vegetation, or seasonal leaf fall period, selecting the down one step convolution scale in the convolution scale set based on the candidate scale; in the case that the pixel position belongs to the near water bank or built-up area junction, selecting the down one step convolution scale in the convolution scale set based on the candidate scale.

[0057] The scale selection rule of the wetness image can include: in the case that the pixel position belongs to the wetland / paddy field inner area, selecting the up one step convolution scale in the convolution scale set based on the candidate scale; in the case that the pixel position belongs to the near water body boundary (d≤b) or mixed pixel, selecting the down one step convolution scale in the convolution scale set based on the candidate scale; in the case that the pixel position belongs to the arid area high noise (sandstorm, thin cloud), taking the candidate scale as the corresponding convolution scale.

[0058] The scale selection rule of the dryness image can include: in the case that the pixel position belongs to the farmland bare land strip / cultivation texture area, selecting the up one step convolution scale in the convolution scale set based on the candidate scale; in the case that the pixel position belongs to the built-up area high D and gradient high area, selecting the down one step convolution scale in the convolution scale set based on the candidate scale; in the case that the pixel position belongs to the desert hinterland micro-texture area, selecting the up one step convolution scale in the convolution scale set based on the candidate scale; in the case that the pixel position belongs to the rocky mountain high-contrast fissure, selecting the down one step convolution scale in the convolution scale set based on the candidate scale.

[0059] The scale selection rule of the heat image can include: in the case that the pixel position belongs to the uniform open surface (lake center, desert hinterland) area, selecting the up one step convolution scale in the convolution scale set based on the candidate scale; in the case that the pixel position belongs to the industrial heat island / point heat source, shoreline thermal gradient area, selecting the down one step convolution scale in the convolution scale set based on the candidate scale.

[0060] In another embodiment, after determining the convolution scale map corresponding to each element image, the electronic device can further perform consistency correction based on the convolution scale maps of different element images to further improve the accuracy of the final convolution result. The consistency correction process can include performing consistency checking on each convolution scale map to ensure that the convolution scale corresponding to the heat map image in the same pixel position is the largest convolution scale among the multiple element images, and the difference in the convolution scale corresponding to different element images in the same pixel position in the convolution scale set does not exceed one level. Further, the electronic device can further perform 3x3 median smoothing on each convolution scale map to prevent convolution scale mutation while avoiding scale jump and cross-class pull by performing only within the same mask (without crossing classes).

[0061] Step S30, convolve each element image according to the corresponding convolution scale map to obtain an element convolution result.

[0062] In one possible implementation, after determining the convolution scale map corresponding to each element image, the electronic device can perform Gaussian convolution on each element image in parallel based on the convolution scale map of each element image to obtain an element convolution result corresponding to each element image.

[0063] Step S40, generating an evaluation report of the target region according to each element convolution result.

[0064] In one possible implementation, after completing the convolution of each element image to obtain multiple convolution scale maps, the electronic device can generate an evaluation report of the target region according to each element convolution result. The evaluation report further includes a comprehensive ecological quality score calculated based on the above-mentioned element convolution results. That is, the electronic device can perform principal component analysis based on each element convolution result to obtain a comprehensive ecological quality score. Then, the evaluation report of the target region is generated according to each element image, each element convolution result, and each comprehensive ecological quality score.

[0065] In some embodiments, the process of principal component analysis can include uniformly performing Z-score or robust Z (MAD) standardization on each element convolution result, and performing polarity alignment. The processed element convolution results are subjected to principal component analysis, and the first principal component PC1 is taken as the comprehensive ecological quality score. The electronic device can further check the sign according to the component load, that is, if PC1 is opposite to the ecological quality direction, then take the negative, and then linearly normalize to [0, 1] to obtain the remote sensing ecological index. At the same time, in the process of principal component analysis, if PCA is sensitive to anomalies, robust PCA or factor analysis can be used instead; PCA can also be subjected to correlation constraint, and the contribution of the element convolution results corresponding to the greenness image and the wetness image is positive, and the contribution of the element convolution results corresponding to the dryness image and the heat image is negative.

[0066] Optionally, after the above calculation is completed, for each pixel position marked as invalid, morphological closing operation can be performed on the small hole, and the invalid value is filled by means of neighborhood interpolation. At the same time, the electronic device can also perform scale consistency check (multi-temporal contrast), and calculate the correlation with external ecological indicators (such as ground object coverage, ground surface humidity sample point). After the above processing is completed, based on each element image, each element convolution result and each comprehensive ecological quality score, an evaluation report of the target area is generated, and each element image, each element convolution result and each comprehensive ecological quality score, and the evaluation report of the target area are output. At the same time, the sensor information, parameters for obtaining the remote sensing data, and the generation date of the evaluation report can also be output.

[0067] Figure 2 A schematic diagram of a regional ecological environment quality evaluation process according to an embodiment of the application is shown. As shown in Figure 2 When performing ecological environment quality evaluation of the target area, the embodiment of the application can first obtain remote sensing data of the target area. Based on the remote sensing data, greenness image, dryness image, humidity image and heat image are determined, as well as the category image (ground object category map) representing the landform category of each position in the target area. Further, the correspondence between each landform type in the category image and the convolution scale set is determined. At the same time, after the electronic device performs normalization processing on the greenness image, dryness image, humidity image and heat image, the convolution scale map of each element image is determined according to the correspondence between each landform type in the category image and the convolution scale set. Then, based on the convolution scale map of each element image, PCA principal component analysis, post-processing and quality control are performed to obtain remote sensing ecological index (RSEI) and further generate the ecological environment quality evaluation report of the target area.

[0068] Based on the above technical features, the embodiment of the present application can adapt to the spatial scale based on the category, and balance the detail retention and boundary authenticity. That is, by using the land cover category and its typical texture scale, different kernel sizes are actively selected or weighted, so that each sub-index is smoothed and enhanced at the "most suitable" spatial scale. For high-texture, boundary-complex areas such as built-up areas, shorelines and roads, small kernels are preferred, which can not only suppress local random fluctuations, but also maximize the retention of detailed structures and clear boundaries, significantly reducing the edge smearing and "bleeding" effects caused by cross-class mixing; for blocky, structure-continuous areas such as forests and farmlands, medium and large kernels are used to improve patch consistency and regional coherence, avoiding patchiness and fragmentation caused by too small windows. Overall, this class-matched scale strategy improves the fidelity of spatial geometric morphology and the stability of ecological element expression, providing clearer and more reliable input features for subsequent RSEI fusion. At the same time, the embodiment of the present application is robust to complex scenes and cooperates with multi-source resolution, improving fusion stability and spatio-temporal consistency. That is, for high-frequency spots such as bare land and desert, and areas with obvious random fluctuations, larger kernels or multi-scale weighting strategies are used to effectively suppress salt and pepper noise, reduce isolated small patches, and improve the spatial continuity and statistical robustness of sub-indices; at the same time, the "effective spatial support" alignment strategy and the class-adaptive window are introduced to alleviate the resolution difference and sampling mismatch between land surface temperature (LST) and multi-spectral indicators, and to reduce the pseudo-difference and mispositioning hot spots caused by coarse and fine resolution mixing. After this processing, the covariance structure of the heat factor and the greenness, wetness and dryness in PCA is more stable, the principal component direction is more explicit, the fusion result is less sensitive to outliers and noise, the time series comparison is more coherent, and the overall interpretation, reliability and cross-regional applicability of the evaluation result are improved.

[0069] It should be understood that, although each step in the above flowcharts is displayed in sequence according to the direction of the arrow, these steps are not necessarily executed in the order indicated by the arrow. Unless otherwise specified herein, there is no strict order limitation for the execution of these steps, and these steps can be executed in other orders. Moreover, at least part of the steps in the above flowcharts can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily sequential, but can be alternately or alternately executed with at least part of other steps or sub-steps or stages of other steps.

[0070] Based on the foregoing embodiments, the embodiments of the present application provide a regional ecological environment quality assessment device, which comprises various modules and units included in the modules, and can be realized by a processor. Of course, it can also be realized by a specific logic circuit. In the implementation process, the processor can be a central processing unit (CPU), a microprocessor (MPU), a digital signal processor (DSP), or a field programmable gate array (FPGA).

[0071] Figure 3 A schematic diagram of a regional ecological environment quality assessment device according to an embodiment of the present application is shown. As shown in Figure 3 The regional ecological environment quality assessment device according to the embodiment of the present application comprises: An image acquisition module 30 configured to determine a plurality of element images according to remote sensing data of a target region; A scale determination module 31 configured to determine a convolution scale corresponding to each pixel position in each element image according to a category image of the target region, to obtain a convolution scale map of each element image, the category image being configured to represent a landform category of each pixel position in the target region; A convolution processing module 32 configured to perform convolution processing on each element image according to the corresponding convolution scale map, to obtain an element convolution result; A quality assessment module 33 configured to generate an assessment report of the target region according to the element convolution result.

[0072] In a possible implementation, the image acquisition module 30 is further configured to: perform image preprocessing on the remote sensing data of the target region; determine element images corresponding to a plurality of preset element dimensions according to the remote sensing data after image preprocessing.

[0073] In a possible implementation, the scale determination module 31 is further configured to: determine a convolution scale set corresponding to the landform category included in the category image; determine a landform category of each pixel position in each element image according to the category image; for each element image, determine a convolution scale corresponding to each pixel position according to a corresponding scale selection rule and the landform category of the pixel position in the convolution scale set, to obtain a convolution scale map of the element image.

[0074] In a possible implementation, the scale determination module 31 is further configured to: for each element image, select a candidate scale of each pixel position in the convolution scale set corresponding to the pixel position; Based on the scale selection rules corresponding to the feature image, the convolution scale is determined from the convolution scale set according to the candidate scale corresponding to each pixel position, thus obtaining the convolution scale map of the feature image.

[0075] In one possible implementation, the feature images include greenness images, dryness images, humidity images, and heat images, and the scale determination module 31 is further used for: For greenness images, dryness images, and humidity images, a preset baseline scale is selected as the candidate scale for each pixel location from the set of convolution scales. For thermal images, candidate scales are selected from the set of convolution scales corresponding to each pixel location based on preset spatial support conditions.

[0076] In one possible implementation, the scale determination module 31 is further used for: Consistency correction is performed on the convolution scale map corresponding to each feature image.

[0077] In one possible implementation, the quality assessment module 33 is further used for: Principal component analysis was performed based on the convolution results of each element to obtain the comprehensive ecological quality score; An assessment report for the target area is generated based on the images of each element, the convolution results of each element, and the comprehensive ecological quality score of each element.

[0078] The descriptions of the above device embodiments are similar to those of the above method embodiments, and have similar beneficial effects. For technical details not disclosed in the device embodiments of this application, please refer to the descriptions of the method embodiments of this application for understanding.

[0079] It should be noted that, in the embodiments of this application... Figure 3 The module division of the regional ecological environment quality assessment device shown is illustrative and represents only a logical functional division; in actual implementation, other division methods may be used. Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, exist as separate physical units, or be integrated into one unit with two or more units. The integrated units can be implemented in hardware, as software functional units, or a combination of both.

[0080] It should be noted that, in the embodiments of the present application, if the above-mentioned method is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the embodiments of the present application can be embodied in the form of a software product, and the computer software product is stored in a storage medium, and includes a plurality of instructions for causing an electronic device to execute all or part of the methods described in the embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read only memory (ROM), a magnetic disk or an optical disk, and various media that can store program codes. Thus, the embodiments of the present application are not limited to any specific hardware and software combination.

[0081] Figure 4 A schematic diagram of an electronic device according to an embodiment of the present application is shown. As shown in Figure 4 , the present application provides an electronic device, which can be a server, and the internal structure diagram thereof can be as shown in Figure 4 . The electronic device includes a processor 420, a memory and a transceiver 440 connected through a system bus 410. The processor 420 of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium 431 and an internal memory 432. The non-volatile storage medium 431 stores an operating system, a computer program and a database. The internal memory 432 provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium 431. The database of the electronic device is used to store data. The transceiver 440 of the electronic device is used to communicate with external terminals through network connection. The computer program is executed by the processor 420 to implement the above-mentioned method.

[0082] The embodiments of the present application provide a computer readable storage medium having a computer program stored thereon, which is executed by the processor 420 to implement the steps in the method provided in the above-mentioned embodiments.

[0083] The embodiments of the present application provide a computer program product containing instructions, which, when executed on a computer, cause the computer to perform the steps in the method provided by the above-mentioned method embodiments.

[0084] Those skilled in the art can understand that Figure 4 the structure shown in the above-mentioned embodiments is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the electronic device to which the scheme of the present application is applied. The specific electronic device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0085] In a possible implementation, the photographing prompting device provided in the present application can be implemented in the form of a computer program, which can run on an electronic device as shown in Figure 4 The memory of the electronic device can store various program modules constituting the above device. The computer program constituted by the various program modules causes the processor 420 to perform the steps in the method of various embodiments of the present application described in the specification.

[0086] It should be noted that the description of the above storage medium and device embodiments is similar to the description of the above method embodiments, and has similar beneficial effects as the method embodiments. For technical details of the present application that are not disclosed in the storage medium, storage medium and device embodiments, please refer to the description of the method embodiments of the present application.

[0087] It should be understood that the "one embodiment" or "an embodiment" or "some embodiments" mentioned throughout the specification means that the specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present application. Therefore, "in a possible implementation" or "in an embodiment" or "in some embodiments" appearing throughout the specification does not necessarily mean the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. It should be understood that in various embodiments of the present application, the size of the sequence number of the above processes does not mean the execution order, and the execution order of the processes should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application. The above sequence number of the embodiments of the present application is only for description, not representing the advantages and disadvantages of the embodiments. The above description of each embodiment tends to emphasize the differences between each embodiment, and the same or similar parts can be referred to each other. For the sake of brevity, this paper will not repeat here.

[0088] The term "and / or" in this paper is only a description of the association relationship of the associated objects, which means that there can be three kinds of relationships, for example, object A and / or object B, which can represent three cases of the existence of object A, the existence of object A and object B, and the existence of object B.

[0089] It should be noted that in this paper, the term "include", "contain" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or includes elements inherent to such process, method, article or device. Without more limitations, the element defined by the sentence "including a…" does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0090] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other manners. The embodiments described above are merely exemplary, for example, the division of the modules is only a logical function division, and there can be another division manner for the actual implementation, for example, a plurality of modules or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the various components shown or discussed can be indirect coupling or communication connection through some interfaces, devices or modules, and can be electrical, mechanical or other forms.

[0091] The modules described above as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules; they can be located in one place, or distributed on multiple network units; and some or all of the modules can be selected according to actual needs to achieve the purpose of the embodiments.

[0092] In addition, each functional module in each embodiment of the present application can be integrated into one processing unit, or each module can be a separate unit, or two or more modules can be integrated into one unit; the integrated module can be realized in the form of hardware or hardware plus software functional unit.

[0093] Those of ordinary skill in the art can understand that all or part of the steps of the above-mentioned method embodiments can be completed by a program instructing related hardware, and the foregoing program can be stored in a computer readable storage medium, and the program executes the steps of the above-mentioned method embodiments when executed; and the foregoing storage medium includes mobile storage devices, read-only memories (ROM), magnetic discs or optical discs and various storage media that can store program codes.

[0094] Alternatively, the integrated units of the present application, if implemented in the form of software functional modules and sold or used as independent products, can also be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the embodiments of the present application can be embodied in the form of a software product, and the computer software product is stored in a storage medium, and includes a plurality of instructions for causing an electronic device to execute all or part of the methods described in the embodiments of the present application. The foregoing storage medium includes mobile storage devices, ROM, magnetic discs or optical discs and various storage media that can store program codes.

[0095] The methods disclosed in the several method embodiments provided in the present application can be combined arbitrarily without conflict, to obtain new method embodiments.

[0096] The features disclosed in several product embodiments provided by the present application can be arbitrarily combined, without conflict, to obtain new product embodiments.

[0097] The features disclosed in several method or device embodiments provided by the present application can be arbitrarily combined, without conflict, to obtain new method embodiments or device embodiments.

[0098] The above description is merely illustrative of the application, and the scope of the application is not limited thereto. Any variations and modifications of the application, which would occur to those skilled in the art, are to be considered within the scope of the application. Therefore, the scope of the application is to be determined by the claims.

Claims

1. A regional ecological environment quality assessment method, characterized in that, The method comprises: determining a plurality of element images according to remote sensing data of a target region; determining a convolution scale corresponding to each pixel position in each of the element images according to a category image of the target region, wherein the category image is used to represent a geomorphologic category of each pixel position in the target region, and a convolution scale map of each of the element images is obtained; performing convolution processing on each of the element images according to the corresponding convolution scale map to obtain an element convolution result; generating an evaluation report of the target region according to each of the element convolution results.

2. The method of claim 1, wherein, The method comprises: performing image preprocessing on the remote sensing data of the target region; determining element images corresponding to a plurality of preset element dimensions according to the remote sensing data after image preprocessing.

3. The method of claim 1, wherein, The method comprises: determining a convolution scale set corresponding to the geomorphologic categories included in the category image; determining a geomorphologic category of each pixel position in each of the element images according to the category image; for each of the element images, determining a convolution scale corresponding to each of the pixel positions in the element image according to a corresponding scale selection rule and the geomorphologic category of each of the pixel positions in the convolution scale set, and obtaining a convolution scale map of the element image.

4. The method of claim 3, wherein, The method comprises: for each of the element images, selecting a candidate scale of each of the pixel positions in the convolution scale set corresponding to each of the pixel positions; determining a convolution scale in the convolution scale set according to the candidate scale corresponding to each of the pixel positions according to a corresponding scale selection rule of the element image, and obtaining a convolution scale map of the element image.

5. The method of claim 4, wherein, The element images comprise a greenness image, a dryness image, a wetness image and a heat image, and the method comprises: for the greenness image, the dryness image and the wetness image, selecting a preset reference scale as the candidate scale of each of the pixel positions in the convolution scale set corresponding to each of the pixel positions; for the heat image, selecting the candidate scale of each of the pixel positions in the convolution scale set corresponding to each of the pixel positions according to a preset spatial support condition.

6. The method of claim 3, wherein, The method further comprises: performing consistency correction on the convolution scale map corresponding to each of the element images.

7. The method of claim 1, wherein, The method comprises: perform principal component analysis based on the convolution results of the elements to obtain comprehensive ecological quality scores; generate an evaluation report of the target region according to the element images, the convolution results of the elements, and the comprehensive ecological quality scores.

8. A regional ecological environment quality assessment device, characterized in that, The device comprises: An image acquisition module configured to determine a plurality of element images according to remote sensing data of a target region; A scale determination module configured to determine a convolution scale corresponding to each pixel position in each element image according to a category image of the target region, and obtain a convolution scale map of each element image, wherein the category image is used to represent the landform category of each pixel position in the target region; A convolution processing module configured to perform convolution processing on each element image according to the corresponding convolution scale map to obtain element convolution results; A quality evaluation module configured to generate an evaluation report of the target region according to the convolution results of the elements.

9. An electronic device comprising a memory and a processor, the memory storing a computer program operable to run on the processor, characterized in that, The processor implements the steps of the method of any one of claims 1 to 7 when executing the program.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program, when executed by the processor, implements the method of any one of claims 1 to 7.

Images