Saline-alkali land information monitoring and treatment method and system

Abstract

This application relates to the technical field of saline-alkali land monitoring and treatment. Specifically, it discloses a method and system for monitoring and treating saline-alkali land information. It uses a camera to collect the overall terrain data of the target saline-alkali land before treatment and the overall terrain data after treatment, and introduces an image processing algorithm based on deep learning to extract features from the overall terrain data before and after treatment, so as to capture the topographic and geomorphic features of the target saline-alkali land before and after treatment respectively. Furthermore, taking the topographic and geomorphic features of the saline-alkali land before treatment as the baseline state, through causal alignment analysis of the topographic and geomorphic features of the saline-alkali land after treatment and the baseline state, the potential differences between the two are excavated, so as to realize the intelligent evaluation of the treatment effect. In this way, by considering the topographic and geomorphic information before and after the treatment of the saline-alkali land, the treatment effect can be analyzed more accurately, and the influence of short-term fluctuations of soil physical property parameters on the evaluation of the treatment effect can be avoided.

Description

Technical Field

[0001] This application relates to the technical field of saline-alkali land monitoring and treatment, and more specifically, to a method and system for monitoring and treating saline-alkali land information. Background Art

[0002] Saline-alkali land refers to low-yield soil with excessive soluble salts in the soil, which hinders plant growth. Soil salinization not only affects the effective utilization of land resources but also poses a threat to the ecological environment. To improve the efficiency and effectiveness of saline-alkali land treatment, the invention patent with the publication number CN118966611A proposes a method and system for monitoring and treating saline-alkali land information. According to the area and severity of salinity of the saline-alkali land, it is refined and divided into priority sub-regions and non-priority sub-regions. Through sub-regional treatment, resources are concentrated to deal with serious areas, while monitoring and treatment of non-priority areas are maintained. By analyzing data such as changes in soil moisture content, conductivity, soil structure, and vegetation cover before and after saline-alkali land treatment, the treatment coefficient of different regions is calculated to achieve a quantitative assessment of the treatment effect. This method helps to reasonably allocate resources and improve treatment efficiency through sub-regional treatment and quantitative assessment of treatment effects.

[0003] However, when evaluating the treatment effect in the prior art, it mainly focuses on physical measurements and statistical analyses of limited data dimensions at the soil level. However, soil physical property parameters are easily affected by various external factors such as seasonal precipitation, temperature fluctuations, and human farming, resulting in short-term fluctuations, which may lead to insufficient accuracy and reliability of the evaluation results.

[0004] Therefore, an optimized method and system for monitoring and treating saline-alkali land information are expected. Summary of the Invention

[0005] To solve the above technical problems, this application is proposed. Embodiments of this application provide a method and system for monitoring and treating saline-alkali land information, which use a camera to collect the full-view terrain data of the target saline-alkali land before treatment and the full-view terrain data after treatment, and introduce an image processing algorithm based on deep learning to extract features from the full-view terrain data before and after treatment, so as to capture the topographic features of the target saline-alkali land before and after treatment respectively. Furthermore, taking the topographic features of the saline-alkali land before treatment as the baseline state, through causal alignment analysis of the topographic features of the saline-alkali land after treatment and the baseline state, the potential differences between the two are excavated, thereby realizing an intelligent evaluation of the treatment effect. In this way, by considering the topographic information of the saline-alkali land before and after treatment, the treatment effect can be analyzed more accurately, avoiding the influence of short-term fluctuations of soil physical property parameters on the evaluation of the treatment effect.

[0006] Correspondingly, according to one aspect of this application, a system for monitoring and treating saline-alkali land information is provided, which includes:

[0007] The pre-treatment overall terrain data acquisition module is used to acquire the pre-treatment overall terrain data of the target saline-alkali land collected by the camera;

[0008] The post-treatment overall terrain data acquisition module is used to acquire the post-treatment overall terrain data of the target saline-alkali land collected by the camera;

[0009] The topographic and geomorphic feature extraction module is used to extract topographic and geomorphic features from the pre-treatment overall terrain data and the post-treatment overall terrain data to obtain the target saline-alkali land topographic and geomorphic baseline state feature coding map and the target saline-alkali land topographic and geomorphic post-treatment state feature coding map;

[0010] The causal alignment analysis module is used to perform causal alignment analysis based on topographic and geomorphic features on the target saline-alkali land topographic and geomorphic baseline state feature coding map and the target saline-alkali land topographic and geomorphic post-treatment state feature coding map to obtain the target saline-alkali land pre-treatment - post-treatment strong causal joint perception coding feature map;

[0011] The treatment effect evaluation module is used to determine the evaluation result of the treatment effect based on the target saline-alkali land pre-treatment - post-treatment strong causal joint perception coding feature map.

[0012] According to another aspect of the present application, there is provided a method for monitoring and treating saline-alkali land information, which includes:

[0013] Acquire the pre-treatment overall terrain data of the target saline-alkali land collected by the camera;

[0014] Acquire the post-treatment overall terrain data of the target saline-alkali land collected by the camera;

[0015] Extract topographic and geomorphic features from the pre-treatment overall terrain data and the post-treatment overall terrain data to obtain the target saline-alkali land topographic and geomorphic baseline state feature coding map and the target saline-alkali land topographic and geomorphic post-treatment state feature coding map;

[0016] Perform causal alignment analysis based on topographic and geomorphic features on the target saline-alkali land topographic and geomorphic baseline state feature coding map and the target saline-alkali land topographic and geomorphic post-treatment state feature coding map to obtain the target saline-alkali land pre-treatment - post-treatment strong causal joint perception coding feature map;

[0017] Determine the evaluation result of the treatment effect based on the target saline-alkali land pre-treatment - post-treatment strong causal joint perception coding feature map.

[0018] Compared with the prior art, the saline-alkali land information monitoring and governance method and system provided by this application utilize a camera to collect the overall terrain data of the target saline-alkali land before treatment and the overall terrain data after treatment, introduce an image processing algorithm based on deep learning to extract features from the overall terrain data before and after treatment, so as to capture the topographic features of the target saline-alkali land before and after treatment respectively. Furthermore, taking the topographic features of the saline-alkali land before treatment as the baseline state, through causal alignment analysis of the topographic features of the saline-alkali land after treatment and the baseline state, the potential differences between the two are excavated, thereby realizing the intelligent evaluation of the treatment effect. In this way, by considering the topographic information of the saline-alkali land before and after treatment, the treatment effect can be analyzed more accurately, and the influence of short-term fluctuations in soil physical property parameters on the evaluation of the treatment effect can be avoided. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] By describing the embodiments of this application in more detail with reference to the accompanying drawings, the above and other objects, features, and advantages of this application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of this application, and constitute a part of the specification. Together with the embodiments of this application, they are used to explain this application, but do not constitute a limitation to this application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0020] Figure 1 It is a block diagram of a saline-alkali land information monitoring and governance system according to an embodiment of this application.

[0021] Figure 2 It is a schematic diagram of data flow of a saline-alkali land information monitoring and governance system according to an embodiment of this application.

[0022] Figure 3 It is a block diagram of a causal alignment analysis module in a saline-alkali land information monitoring and governance system according to an embodiment of this application.

[0023] Figure 4 It is a block diagram of a causal association strength measurement unit in a saline-alkali land information monitoring and governance system according to an embodiment of this application.

[0024] Figure 5 It is a flowchart of a saline-alkali land information monitoring and governance method according to an embodiment of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0025] As shown in this application and the claims, unless the context clearly indicates otherwise, words such as "a", "an", "one", and / or "the" are not specifically singular and may also include plural. Generally speaking, the terms "including" and "comprising" only indicate the inclusion of the clearly identified steps and elements, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0026] Although the present application makes various references to certain modules in the system according to embodiments of the present application, any number of different modules may be used and run on the user terminal and / or the server. The modules are merely illustrative, and different aspects of the system and method may use different modules.

[0027] Flowcharts are used in the present application to illustrate the operations performed by the system according to embodiments of the present application. It should be understood that the operations above or below do not necessarily have to be executed precisely in order. On the contrary, various steps may be processed in reverse order or simultaneously as needed. Also, other operations may be added to these processes, or one or more steps may be removed from these processes.

[0028] Hereinafter, exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all embodiments of the present application. It should be understood that the present application is not limited by the exemplary embodiments described herein.

[0029] As mentioned in the above background art, the patent publication number CN118966611A proposes a method and system for monitoring and managing saline-alkali land information. According to the area and severity of the saline-alkali land, it is refined and divided into priority sub-regions and non-priority sub-regions. Through sub-regional management, resources are concentrated to deal with serious areas, while monitoring and management of non-priority areas are maintained. By analyzing data such as changes in soil moisture content, conductivity, soil structure, and vegetation cover before and after the treatment of saline-alkali land, the treatment coefficients of different regions are calculated to achieve a quantitative evaluation of the treatment effect. This method helps to reasonably allocate resources and improve the treatment efficiency through sub-regional management and quantitative evaluation of the treatment effect.

[0030] In the prior art, the focus was mainly on the physical measurement and statistical analysis of limited data dimensions at the soil level. However, the physical property parameters of the soil are easily affected by various external factors such as seasonal precipitation, temperature fluctuations, and human tillage, resulting in short-term fluctuations, which may in turn lead to insufficient accuracy and reliability of the evaluation results. To address this technical problem, the present application proposes an optimized saline-alkali land information monitoring and treatment system. It uses cameras to collect the full-view terrain data of the target saline-alkali land before treatment and the full-view terrain data after treatment, and introduces an image processing algorithm based on deep learning to extract features from the full-view terrain data before and after treatment, so as to capture the topographic and geomorphic features of the target saline-alkali land before and after treatment respectively. Then, taking the topographic and geomorphic features of the saline-alkali land before treatment as the baseline state, through causal alignment analysis of the topographic and geomorphic features of the saline-alkali land after treatment and the baseline state, the potential differences between the two are excavated, thereby realizing the intelligent evaluation of the treatment effect. In this way, by considering the topographic and geomorphic information of the saline-alkali land before and after treatment, the treatment effect can be analyzed more accurately, avoiding the influence of short-term fluctuations of soil physical property parameters on the evaluation of the treatment effect.

[0031] Figure 1 It is a block diagram of a saline-alkali land information monitoring and treatment system according to an embodiment of the present application. Figure 2 It is a schematic diagram of data flow of a saline-alkali land information monitoring and treatment system according to an embodiment of the present application. As Figure 1 and Figure 2 shown, the saline-alkali land information monitoring and treatment system 100 includes: a pre-treatment full-view terrain data acquisition module 110 for acquiring the pre-treatment full-view terrain data of the target saline-alkali land collected by the camera; a post-treatment full-view terrain data acquisition module 120 for acquiring the post-treatment full-view terrain data of the target saline-alkali land collected by the camera; a topographic and geomorphic feature extraction module 130 for extracting topographic and geomorphic features from the pre-treatment full-view terrain data and the post-treatment full-view terrain data to obtain a target saline-alkali land topographic and geomorphic baseline state feature encoding map and a target saline-alkali land topographic and geomorphic post-treatment state feature encoding map; a causal alignment analysis module 140 for performing causal alignment analysis based on topographic and geomorphic features on the target saline-alkali land topographic and geomorphic baseline state feature encoding map and the target saline-alkali land topographic and geomorphic post-treatment state feature encoding map to obtain a target saline-alkali land pre-treatment - post-treatment strong causal joint perception encoding feature map; a treatment effect evaluation module 150 for determining an evaluation result of the treatment effect based on the target saline-alkali land pre-treatment - post-treatment strong causal joint perception encoding feature map.

[0032] In the above saline-alkali land information monitoring and treatment system, the pre-treatment overall terrain data acquisition module 110 is used to acquire the pre-treatment overall terrain data of the target saline-alkali land collected by the camera. It should be understood that the terrain and landform, as an important intuitive manifestation of the treatment effect of saline-alkali land, its changes often imply the treatment effects in many aspects such as soil structure and vegetation restoration. For example, during the treatment process, changes in terrain and landform such as the change in land flatness, the adjustment of water accumulation areas, and the optimization of slopes may directly affect the drainage, ventilation, and lighting conditions of saline-alkali land, and thus have a profound impact on the overall treatment effect. Therefore, in order to comprehensively and accurately evaluate the treatment effect, this application acquires the pre-treatment overall terrain data of the target saline-alkali land collected by the camera as the baseline data.

[0033] In the above saline-alkali land information monitoring and treatment system, the post-treatment overall terrain data acquisition module 120 is used to acquire the post-treatment overall terrain data of the target saline-alkali land collected by the camera. That is, this application acquires the pre-treatment overall terrain data of the target saline-alkali land collected by the camera as the baseline data, and the post-treatment overall terrain data of the target saline-alkali land as the comparison data, and evaluates the treatment effect of the saline-alkali land by comparing and analyzing the two. Specifically, the camera is a high-definition camera carried on a drone. It should be understood that the drone has the characteristics of being flexible and maneuverable, and can fly at low altitude. It can quickly reach the target saline-alkali land area, is not restricted by terrain and traffic conditions, and can conduct covering shooting of large areas of saline-alkali land in a short time. The high-definition camera focuses light on the image sensor through an optical lens, uses the image sensor to convert the optical signal into an electrical signal or a digital signal, and then performs encoding, compression and other processing through the image processing circuit, and finally forms high-resolution image data, so as to clearly present various details of the saline-alkali land, such as the texture and color difference of the soil surface, providing rich information for subsequent analysis.

[0034] Specifically, first, it is necessary to use a suitable camera to acquire the pre-treatment overall terrain data of the target saline-alkali land and the post-treatment overall terrain data of the target saline-alkali land. Specifically, the camera should have high resolution and a wide-angle lens to be able to capture as much detailed information as possible. At the same time, in order to meet the shooting requirements under different lighting conditions, the camera also needs to have good low-light performance. Considering that saline-alkali land is often vast in area and it is difficult for a single camera to cover the entire area, the method of carrying a camera on a drone is usually adopted for operation. The drone can not only flexibly adjust the flight height and speed to achieve refined shooting of areas of different scales, but also adjust the shooting route according to actual needs to ensure that every corner can be photographed.

[0035] During the preparation phase, the drone's flight path and shooting parameters must be carefully planned. This includes determining the takeoff and landing points, calculating the optimal flight altitude, and setting the number of photos per second. To minimize the impact of external environmental factors on data accuracy, shooting should be conducted on clear days with low winds. Furthermore, understanding the geographic information of the target saline-alkali land, such as altitude variations and topography, will help optimize the drone's flight path and avoid blind spots caused by complex terrain. Every shooting mission requires thorough preparation in advance to ensure a smooth process.

[0036] To ensure data integrity and consistency, the camera should be calibrated before each capture. During calibration, not only should the camera's focal length and exposure time be adjusted, but the white balance should also be set according to the current lighting conditions to ensure the captured photos retain true color. To address the potential reflections caused by the white or light-colored soil surfaces characteristic of saline-alkali land, special attention should be paid to adjusting camera settings to minimize overexposure. Furthermore, given the sparse vegetation cover in saline-alkali land, the overlap of the shots should be appropriately increased during capture to facilitate the blending of transitions between images when subsequently stitched together into a panorama.

[0037] In addition to preparing the hardware and shooting parameters, strict quality control measures are also required to ensure the reliability of the collected data. For example, before the official shooting, a test shot can be conducted in a small area to check whether the image quality meets the requirements. If problems such as image blur or color distortion are found, the camera settings should be adjusted promptly until the problem is resolved. Throughout the shooting process, the operator needs to monitor the status of the drone and the working conditions of the camera in real time, and take immediate corrective measures if any anomalies occur. For example, if the drone deviates from the planned route or encounters unexpected weather conditions, a quick response can be taken, ensuring the safety of the equipment while also ensuring the integrity of the data.

[0038] To ensure efficiency and data consistency during the specific photography task, the target saline-alkali land can be divided into several relatively independent small areas, and photographed sequentially in a pre-planned sequence. After completing the photography work in each small area, it is necessary to carefully check whether the photos obtained cover all the key features of the area and have sufficient overlap with photos of other adjacent areas. This method not only improves the efficiency of the photography work, but also facilitates the subsequent accurate splicing of photos to form a complete pre-treatment terrain dataset.

[0039] Finally, after completing all the shooting tasks, it is an essential step to preliminarily organize and back up the collected data. This includes deleting duplicate or unclear photos, numbering and archiving the remaining photos according to the shooting location and time sequence to ensure that each photo can be traced back to its specific shooting location and moment. At the same time, promptly upload these original data to cloud storage or copy them to an external hard drive for preservation to prevent data loss.

[0040] For the overall terrain data of the target saline-alkali land after treatment, the acquisition operation process is similar to the acquisition steps of the overall terrain data of the target saline-alkali land before treatment. After the treatment is completed, it is also necessary to select appropriate weather conditions for shooting to maintain the consistency and comparability of the data. Considering that the treatment measures may cause significant changes in the topography, such as vegetation restoration and soil structure improvement, more attention needs to be paid to capturing details during shooting, especially those key areas that may show the treatment effect.

[0041] In the shooting work after treatment, in addition to following the aforementioned basic principles regarding equipment calibration, flight path planning, and quality control, it is also necessary to pay extra attention to the specific implementation of the treatment measures. For example, certain areas may change in color or texture due to the planting of specific plants, or the introduction of an irrigation system may cause an increase in local humidity, affecting the reflectivity of the soil surface. In such cases, it is particularly important to adjust the camera settings to adapt to the new environmental conditions. In this way, it can be ensured that the terrain data after treatment has the same quality and accuracy as the data before treatment, providing a solid foundation for subsequent analysis.

[0042] In the above saline-alkali land information monitoring and treatment system, the topographic feature extraction module 130 is used to extract topographic features from the pre-treatment overall topographic data and the post-treatment overall topographic data to obtain the target saline-alkali land topographic baseline state feature encoding map and the target saline-alkali land topographic post-treatment state feature encoding map. In a specific example of the present application, the topographic feature extraction module 130 is used to: use a topographic feature extractor based on the FPT model to perform image feature extraction on the pre-treatment overall topographic data and the post-treatment overall topographic data to obtain the target saline-alkali land topographic baseline state feature encoding map and the target saline-alkali land topographic post-treatment state feature encoding map. Specifically, since the topographic features of saline-alkali land are complex and diverse, including aspects such as soil texture, color, texture, slope, gully distribution, etc., and have different spatial distributions and scale characteristics in images. Therefore, in order to comprehensively cover various topographic features of saline-alkali land, the present application constructs a topographic feature extractor based on the FPT (Feature Pyramid Transformer) model, and performs image feature extraction on the pre-treatment overall topographic data and the post-treatment overall topographic data respectively. The FPT model can capture the detailed information of topographic features in images at different scales by constructing a multi-scale feature pyramid and introducing the Transformer architecture. At the same time, by considering the global context dependence of cross-scale features, it effectively correlates and fuses multi-scale features, so as to achieve a comprehensive description of the topographic features of saline-alkali land, obtain the target saline-alkali land topographic baseline state feature encoding map and the target saline-alkali land topographic post-treatment state feature encoding map, and further provide basic data for subsequent treatment effect analysis.

[0043] In the above-mentioned saline-alkali land information monitoring and governance system, the causal alignment analysis module 140 is used to perform causal alignment analysis based on topographic and geomorphic features on the encoded map of the baseline state features of the target saline-alkali land topographic and geomorphic features and the encoded map of the state features of the target saline-alkali land topographic and geomorphic features after governance to obtain the strong causal joint perception encoded feature map of the target saline-alkali land before-after governance. That is, the present application further reveals the potential connections and change laws between the topographic and geomorphic features before and after governance through feature comparison and interaction analysis of the encoded map of the baseline state features of the target saline-alkali land topographic and geomorphic features and the encoded map of the state features of the target saline-alkali land topographic and geomorphic features after governance, so as to realize the intelligent evaluation of the governance effect. In particular, considering that during the governance process of saline-alkali land, some changes in topographic and geomorphic features may be directly caused by governance measures, while other changes may be caused by natural factors or other unknown factors. Therefore, in order to accurately evaluate the actual effect of governance measures, the present application proposes a causal alignment analysis method based on topographic and geomorphic features, by mining local feature pairs with strong causal associations between the encoded map of the baseline state features of the target saline-alkali land topographic and geomorphic features and the encoded map of the state features of the target saline-alkali land topographic and geomorphic features after governance, to construct a potential causal alignment relationship of the changes in the topographic and geomorphic features of saline-alkali land, thereby helping to reveal the direct impact of governance measures on topographic and geomorphic features, suppressing the interference of other factors, and improving the accuracy of governance effect evaluation. Among them, Figure 3 is a block diagram of the causal alignment analysis module in the saline-alkali land information monitoring and governance system according to an embodiment of the present application. As Figure 3 shown, the causal alignment analysis module 140 includes: a causal association strength measurement unit 141, which is used to measure the causal association strength based on local fine-grained decoupled features on the encoded map of the baseline state features of the target saline-alkali land topographic and geomorphic features and the encoded map of the state features of the target saline-alkali land topographic and geomorphic features after governance to obtain a set of closed-loop strength factors of the causal chain units of the local geomorphic features of the target saline-alkali land before-after governance; a strong causal association pairing unit 142, which is used to perform strong causal association pairing based on local fine-grained decoupled features on the encoded map of the baseline state features of the target saline-alkali land topographic and geomorphic features and the encoded map of the state features of the target saline-alkali land topographic and geomorphic features after governance based on the set of closed-loop strength factors of the causal chain units of the local geomorphic features of the target saline-alkali land before-after governance to obtain a set of {local feature vectors of the baseline geomorphic state of the target saline-alkali land, local feature vectors of the geomorphic state of the target saline-alkali land after governance} feature pairs that constitute the causal chain units; a strong causal joint perception unit 143, which is used to perform multi-dimensional joint perception aggregation processing on the set of {local feature vectors of the baseline geomorphic state of the target saline-alkali land, local feature vectors of the geomorphic state of the target saline-alkali land after governance} feature pairs that constitute the causal chain units to obtain the strong causal joint perception encoded feature map of the target saline-alkali land before-after governance.

[0044] Figure 4Block diagram of the causal association strength measurement unit in the saline-alkali land information monitoring and governance system according to an embodiment of the present application. As Figure 4 shown, the causal association strength measurement unit 141 includes: a feature decoupling subunit 1411, configured to perform feature decoupling on the target saline-alkali land topographic baseline state feature encoding map and the target saline-alkali land topographic post-governance state feature encoding map along the channel dimension to obtain a set of target saline-alkali land baseline geomorphic state local feature vectors and a set of target saline-alkali land post-governance geomorphic state local feature vectors; an alternative feature extraction subunit 1412, configured to respectively extract an alternative target saline-alkali land baseline geomorphic state local feature vector and an alternative target saline-alkali land post-governance geomorphic state local feature vector from the set of target saline-alkali land baseline geomorphic state local feature vectors and the set of target saline-alkali land post-governance geomorphic state local feature vectors; an association strength calculation subunit 1413, configured to perform causal association strength measurement on the alternative target saline-alkali land baseline geomorphic state local feature vector and the alternative target saline-alkali land post-governance geomorphic state local feature vector to obtain the closed-loop strength factor of the local geomorphic feature causal chain unit before-after governance of the target saline-alkali land.

[0045] More specifically, the feature decoupling subunit 1411 is represented by the formula:

[0046]

[0047]

[0048] where represents the feature decoupling operation, and respectively represent the target saline-alkali land topographic baseline state feature encoding map and the target saline-alkali land topographic post-governance state feature encoding map, , , and respectively represent the first, second, th, and th target saline-alkali land baseline geomorphic state local feature vectors in the set of target saline-alkali land baseline geomorphic state local feature vectors, , , and respectively represent the first, second, th, and th target saline-alkali land post-governance geomorphic state local feature vectors in the set of target saline-alkali land post-governance geomorphic state local feature vectors.

[0049] That is, in this application, by performing a decoupling operation on the encoded map of the baseline state characteristics of the target saline-alkali land topographic features and the encoded map of the state characteristics of the target saline-alkali land topographic features after treatment, it is decomposed into a set of local feature vectors of the baseline landform state of the target saline-alkali land and a set of local feature vectors of the landform state of the target saline-alkali land after treatment. In this way, the highly coupled feature relationships are separated, the feature expressions are refined, the semantic representations of each local feature vector are made clearer and more independent, providing finer and more accurate feature units for the subsequent calculation of the causal association strength, and helping to improve the understanding ability of the changes in the saline-alkali land topographic features.

[0050] More specifically, the association strength calculation sub-unit 1413 is used to: construct a semantic alignment modulation flow field between the local feature vector of the baseline landform state of the alternative target saline-alkali land and the local feature vector of the landform state of the alternative target saline-alkali land after treatment, which is expressed by the formula:

[0051]

[0052] where represents the transpose of a vector, represents matrix multiplication, represents a 3×3 convolution operation, represents a 1×1 convolution operation, represents the semantic alignment modulation flow field.

[0053] That is, a semantic alignment modulation flow field is constructed for the local feature vector of the baseline landform state of the alternative target saline-alkali land and the local feature vector of the landform state of the alternative target saline-alkali land after treatment. In an actual scenario, due to factors such as acquisition time and environment, there may be semantic differences and offsets between the local feature vector of the baseline landform state of the alternative target saline-alkali land and the local feature vector of the landform state of the alternative target saline-alkali land after treatment. By constructing the semantic alignment modulation flow field, the local feature vector of the baseline landform state of the alternative target saline-alkali land and the local feature vector of the landform state of the alternative target saline-alkali land after treatment can be calibrated and adjusted at the semantic level, making them more comparable in the semantic space. This operation refines the expression of features at the semantic level, enabling the subsequent feature alignment mapping to be based on a more accurate semantic basis, thus providing a more reliable semantic environment for accurately measuring the causal association strength and improving the accuracy of causal relationship analysis at the semantic level.

[0054] More specifically, the correlation strength calculation subunit 1413 is further configured to: perform feature alignment mapping on the local feature vectors of the baseline landform state of the alternative target saline-alkali land and the local feature vectors of the landform state after treatment of the alternative target saline-alkali land based on the semantic alignment modulation flow field to obtain the aligned local feature vectors of the baseline landform state of the alternative target saline-alkali land and the aligned local feature vectors of the landform state after treatment of the alternative target saline-alkali land, which is expressed by the formula:

[0055]

[0056]

[0057] Wherein, and respectively represent the aligned local feature vectors of the baseline landform state of the alternative target saline-alkali land and the aligned local feature vectors of the landform state after treatment of the alternative target saline-alkali land.

[0058] That is, based on the semantic calibration provided by the semantic alignment modulation flow field, feature alignment mapping is performed on the local feature vectors of the baseline landform state of the alternative target saline-alkali land and the local feature vectors of the landform state after treatment of the alternative target saline-alkali land. In this way, each dimension of each feature vector can establish a more reasonable corresponding relationship with the corresponding dimension of the other feature vector, thereby refining the expression of the feature in terms of dimension and scale. The obtained aligned local feature vectors of the baseline landform state of the alternative target saline-alkali land and the aligned local feature vectors of the landform state after treatment of the alternative target saline-alkali land will have higher consistency and comparability at the feature level, thus further improving the accuracy of causal relationship analysis at the semantic level.

[0059] Specifically, in a preferred example of the present application, the correlation strength calculation subunit 1413 is further configured to: input the aligned local feature vectors of the baseline landform state of the alternative target saline-alkali land and the aligned local feature vectors of the landform state after treatment of the alternative target saline-alkali land into the causal chain link unit screening network to obtain the closed-loop strength factor of the causal chain link unit of the local landform features before and after treatment of the target saline-alkali land, which is expressed by the formula:

[0060]

[0061] Wherein, represents the arccosine function, represents the square of the norm of the calculated vector, represents and the closed-loop strength factor of the causal chain link unit of the local landform features before and after treatment of the target saline-alkali land between.

[0062] That is, the closed-loop intensity factor of the causal chain link unit of the local geomorphic feature before and after the treatment of the target saline-alkali land includes a knowledge storage part based on the feature difference norm between the local feature vector of the baseline geomorphic state of the aligned candidate target saline-alkali land and the local feature vector of the geomorphic state after the treatment of the aligned candidate target saline-alkali land. and the norm representations of both 、 The environmental interaction part under it is used to cope with the forgetting of causal evolution knowledge and the loss of autocorrelation plasticity. Further, based on the positive contribution (as the numerator) of the distributed evolution knowledge to the closed-loop intensity of the causal chain link and the negative environmental interaction 、 The reverse contribution (as the denominator) is used to intuitively reflect the directional similarity of the causal closed-loop intensity through geometric correlation causal intensity quantification representation ( function), so as to obtain the closed-loop intensity factor of the causal chain link unit of the local geomorphic feature before and after the treatment of the target saline-alkali land for screening highly relevant and causally related feature pairs.

[0063] Specifically, in a specific example of the present application, the strong causal association pairing unit 142 is used to: based on the comparison between the closed-loop intensity factor of the causal chain link unit of the local geomorphic feature before and after the treatment of the target saline-alkali land and a preset threshold, determine whether the local feature vector of the baseline geomorphic state of the candidate target saline-alkali land and the local feature vector of the geomorphic state after the treatment of the candidate target saline-alkali land form a causal chain link unit, which is expressed by the formula:

[0064]

[0065] Wherein, represents the preset threshold, represents the feature pair {local feature vector of the baseline geomorphic state of the target saline-alkali land, local feature vector of the geomorphic state after the treatment of the target saline-alkali land} that forms a causal chain link unit.

[0066] [[ID=**28**]]That is, the closed-loop intensity factor of the causal chain link unit of the local geomorphic feature before and after the treatment of the target saline-alkali land is compared with the preset threshold one by one. Through this comparison operation, those feature pairs with a causal association intensity reaching or exceeding the preset threshold can be accurately screened out. These feature pairs represent a strong causal connection between the topographic and geomorphic features before and after the treatment. After excluding the feature pairs with weak causal association intensity or no obvious causal relationship, the subsequent analysis and processing can focus on the truly meaningful strong causal association features, which helps to reduce the interference of irrelevant information and improve the accuracy and pertinence of the analysis of the influencing factors of the saline-alkali land treatment effect. [[ID=**30**]]

[0067] Specifically, in a specific example of the present application, the strong causal joint perception unit 143 is configured to: input each {target saline-alkali land baseline geomorphic state local feature vector, target saline-alkali land post-treatment geomorphic state local feature vector} feature pair of the set of feature pairs of the causal chain link units into the causal chain link unit semantic dynamic response network to obtain a set of strong causal joint perception coding feature vectors of the pre-treatment - post-treatment geomorphology of the target saline-alkali land; perform feature aggregation on the set of strong causal joint perception coding feature vectors of the pre-treatment - post-treatment geomorphology of the target saline-alkali land to obtain the strong causal joint perception coding feature map of the pre-treatment - post-treatment of the target saline-alkali land, which is expressed by the formula:

[0068]

[0069] wherein, and represent different weight parameters, and represent different weight matrices, represents point subtraction, represents point addition, represents the and of the causal chain link unit, and the strong causal joint perception coding feature vector of the pre-treatment - post-treatment geomorphology of the target saline-alkali land obtained after joint perception coding.

[0070] That is, on the basis of the previous refinement of feature expression and screening of strong causal association feature pairs, multi-dimensional joint perception aggregation processing is performed on each feature in the set of {target saline-alkali land baseline geomorphic state local feature vector, target saline-alkali land post-treatment geomorphic state local feature vector} feature pairs of the causal chain link units, and the feature pairs are effectively integrated, so that the finally generated strong causal joint perception coding feature map of the pre-treatment - post-treatment of the target saline-alkali land can more comprehensively and accurately reflect the topographic and geomorphic changes before and after saline-alkali land treatment and their causal relationships, provide rich and reliable feature information for treatment effect evaluation, and improve the accuracy and credibility of the evaluation results.

[0071] In the above saline-alkali land information monitoring and treatment system, the treatment effect evaluation module 150 is used to determine the evaluation result of the treatment effect based on the pre-treatment - post-treatment strong causal joint perception coding feature map of the target saline-alkali land. In a specific example of the present application, the treatment effect evaluation module 150 is used to: input the pre-treatment - post-treatment strong causal joint perception coding feature map of the target saline-alkali land into the intelligent treatment effect evaluation module based on a classifier to obtain the evaluation result, and the evaluation result is used to indicate whether the treatment effect meets the preset requirements. Here, the pre-treatment - post-treatment strong causal joint perception coding feature map of the target saline-alkali land contains the strong causal association information of the topographic and geomorphic features before and after treatment. After receiving it as input, the intelligent treatment effect evaluation module based on a classifier automatically determines whether the treatment effect reaches the expected goal by learning the associated change pattern of the topographic and geomorphic features before and after treatment in the pre-treatment - post-treatment strong causal joint perception coding feature map of the target saline-alkali land, and gives the corresponding evaluation result. If the evaluation result shows that the treatment effect meets the preset requirements, it means that the treatment measures are effective and can be further promoted and applied; if the evaluation result shows that the treatment effect does not meet the preset requirements, it is necessary to re-examine the treatment plan and adjust the treatment strategy in order to achieve better results in subsequent treatment work.

[0072] Specifically, the classifier first unfolds the pre-treatment - post-treatment strong causal joint perception coding feature map of the target saline-alkali land into a pre-treatment - post-treatment strong causal joint perception coding feature vector, and then uses the fully connected layer of the intelligent treatment effect evaluation module to perform fully connected coding on the pre-treatment - post-treatment strong causal joint perception coding feature vector to obtain a pre-treatment - post-treatment strong causal joint perception fully connected coding feature vector. Then, the pre-treatment - post-treatment strong causal joint perception fully connected coding feature vector is input into the Softmax classification function of the intelligent treatment effect evaluation module to obtain the probability values of the pre-treatment - post-treatment strong causal joint perception coding feature vector belonging to each classification label, where the classification labels include that the treatment effect meets the preset requirements and that the treatment effect does not meet the preset requirements. Finally, the classification label corresponding to the largest of the probability values is determined as the evaluation result.

[0073] When the governance effect meets the preset requirements, it means that the governance measures taken have largely achieved the expected goals, such as significantly reducing soil salinity, increasing vegetation coverage, or improving soil structure. At this time, it is important to identify which specific strategies and methods have played a key role in achieving success. For example, if it is found that the planting of a certain type of plant has significantly promoted vegetation restoration, then it will be particularly important to further study the ecological adaptability and growth requirements of this plant. In addition, the synergistic effects between different governance measures need to be examined to understand how they work together to achieve the best results. Through such analysis, not only can an effective governance plan be summarized, but it can also provide a reference for governance projects in other similar environments. At the same time, consideration should be given to expanding the application scope of this governance model, promoting it to areas with similar conditions, and making appropriate adjustments according to specific circumstances. To ensure long-term effects, a continuous monitoring mechanism should also be established to regularly evaluate the governance results and promptly discover and solve new problems that may arise.

[0074] On the other hand, if the governance effect fails to meet the preset requirements, it is necessary to analyze the reasons in more detail. First, it is possible to check whether there are any biases or deficiencies in the data collection and processing process. For example, whether the quality of the original data is high enough, whether the feature selection is appropriate, and whether the sample set used for model training is representative. Any oversight in any link may cause the final evaluation result to deviate from the actual situation. Second, starting from the governance measures themselves, examine whether there are certain aspects that have not been considered or implemented properly. For example, whether the layout of the irrigation system is reasonable, whether the selected plant species are suitable for the local climate conditions, and whether the land management measures are appropriate. In response to these problems, new technologies and methods may need to be introduced to optimize the existing governance plan. For example, using more advanced sensor technology to monitor soil humidity and salinity changes in real time, or exploring emerging governance means such as bioremediation.

[0075] In addition, when faced with poor governance effects, interdisciplinary cooperation is particularly crucial. Inviting experts from different fields to participate in the discussion together and putting forward opinions from their respective professional perspectives helps to form a comprehensive and balanced solution. For example, agricultural scientists can give suggestions on the crop rotation system; soil scientists can recommend suitable soil amendments according to the characteristics of different types of soil; and ecologists are responsible for evaluating the effectiveness of biodiversity protection measures. By integrating the wisdom of all parties, a more scientific and reasonable governance strategy can be formulated.

[0076] In summary, the saline-alkali land information monitoring and governance system based on the embodiments of the present application is elucidated. It uses cameras to collect the overall topographic data of the target saline-alkali land before treatment and the overall topographic data after treatment, and introduces an image processing algorithm based on deep learning to extract features from the overall topographic data before and after treatment, so as to capture the topographic and geomorphic features of the target saline-alkali land before and after treatment respectively. Furthermore, taking the topographic and geomorphic features of the saline-alkali land before treatment as the baseline state, through causal alignment analysis of the topographic and geomorphic features of the saline-alkali land after treatment and the baseline state, the potential differences between the two are excavated, thereby realizing the intelligent evaluation of the treatment effect. In this way, by considering the topographic and geomorphic information of the saline-alkali land before and after treatment, the treatment effect can be analyzed more accurately, and the influence of short-term fluctuations in soil physical property parameters on the evaluation of the treatment effect can be avoided.

[0077] Furthermore, a saline-alkali land information monitoring and governance method is also provided.

[0078] Figure 5 The flowchart of the saline-alkali land information monitoring and governance method according to the embodiments of the present application is as follows. As Figure 5 shown, the saline-alkali land information monitoring and governance method includes the steps: S1, obtaining the overall topographic data of the target saline-alkali land before treatment collected by the camera; S2, obtaining the overall topographic data of the target saline-alkali land after treatment collected by the camera; S3, extracting topographic and geomorphic features from the overall topographic data before treatment and the overall topographic data after treatment to obtain the target saline-alkali land topographic and geomorphic baseline state feature coding map and the target saline-alkali land topographic and geomorphic state feature coding map after treatment; S4, performing causal alignment analysis based on topographic and geomorphic features on the target saline-alkali land topographic and geomorphic baseline state feature coding map and the target saline-alkali land topographic and geomorphic state feature coding map after treatment to obtain the target saline-alkali land before-treatment - after-treatment strong causal joint perception coding feature map; S5, determining the evaluation result of the treatment effect based on the target saline-alkali land before-treatment - after-treatment strong causal joint perception coding feature map.

[0079] Here, those skilled in the art can understand that the specific operations of each step in the above-mentioned saline-alkali land information monitoring and governance method have been introduced in detail in the description of the saline-alkali land information monitoring and governance system above with reference to Figures 1 to 4 and therefore, the repeated description thereof will be omitted.

[0080] The basic principles of the present invention have been described above in combination with specific embodiments. However, it should be noted that the advantages, advantages, effects, etc. mentioned in the present invention are only examples and not limitations, and it cannot be considered that these advantages, advantages, effects, etc. are essential for each embodiment of the present invention. In addition, the specific details of the above embodiments are only for the purpose of illustration and easy understanding, rather than limitations, and the above details do not limit the present invention to necessarily adopt the above specific details to implement.

[0081] In the above embodiments, the descriptions of the various embodiments each have their own emphasis. For parts not detailed or recorded in a certain embodiment, reference may be made to the relevant descriptions of other embodiments. In the several embodiments provided by the present invention, it should be understood that the disclosed system and method can be implemented in other ways. For example, the system embodiment described above is merely illustrative. For example, the unit division is only a logical function division, and there may be other division methods in actual implementation. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0082] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present invention. Any associated drawing marks in the claims should not be regarded as limiting the claims involved.

[0083] In addition, it is obvious that the word "including" does not exclude other units or steps, and the singular does not exclude the plural. The multiple units stated in the system claims can also be implemented by one unit through software or hardware.

[0084] Finally, it should be noted that the above description has been given for purposes of illustration and description. In addition, the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A saline-alkali land information monitoring and governance system, characterized in that, Including: An ungoverned overall terrain data acquisition module, configured to acquire the ungoverned overall terrain data of the target saline-alkali land collected by a camera; A governed overall terrain data acquisition module, configured to acquire the governed overall terrain data of the target saline-alkali land collected by the camera; A terrain and landform feature extraction module, configured to extract terrain and landform features from the ungoverned overall terrain data and the governed overall terrain data to obtain a target saline-alkali land terrain and landform baseline state feature coding map and a target saline-alkali land terrain and landform governed state feature coding map; A causal alignment analysis module, configured to perform causal alignment analysis based on terrain and landform features on the target saline-alkali land terrain and landform baseline state feature coding map and the target saline-alkali land terrain and landform governed state feature coding map to obtain a target saline-alkali land before-governance-after-governance strong causal joint perception coding feature map; A governance effect evaluation module, configured to determine an evaluation result of the governance effect based on the target saline-alkali land before-governance-after-governance strong causal joint perception coding feature map; The causal alignment analysis module includes: A causal association intensity measurement unit, configured to measure the causal association intensity based on local fine-grained decoupled features on the target saline-alkali land terrain and landform baseline state feature coding map and the target saline-alkali land terrain and landform governed state feature coding map to obtain a set of closed-loop intensity factors of the target saline-alkali land before-governance-after-governance local landform feature causal chain units; A strong causal association pairing unit, configured to perform strong causal association pairing based on local fine-grained decoupled features on the target saline-alkali land terrain and landform baseline state feature coding map and the target saline-alkali land terrain and landform governed state feature coding map based on the set of closed-loop intensity factors of the target saline-alkali land before-governance-after-governance local landform feature causal chain units to obtain a set of {target saline-alkali land baseline landform state local feature vectors, target saline-alkali land governed landform state local feature vectors} feature pairs that constitute causal chain units; A strong causal joint perception unit, configured to perform multi-dimensional joint perception aggregation processing on the set of {target saline-alkali land baseline landform state local feature vectors, target saline-alkali land governed landform state local feature vectors} feature pairs that constitute causal chain units to obtain the target saline-alkali land before-governance-after-governance strong causal joint perception coding feature map; The causal association intensity measurement unit includes: A feature decoupling subunit, configured to perform feature decoupling on the target saline-alkali land terrain and landform baseline state feature coding map and the target saline-alkali land terrain and landform governed state feature coding map along the channel dimension of the target saline-alkali land terrain and landform baseline state feature coding map and the target saline-alkali land terrain and landform governed state feature coding map to obtain a set of target saline-alkali land baseline landform state local feature vectors and a set of target saline-alkali land governed landform state local feature vectors; An alternative feature extraction sub-unit is used to extract an alternative target saline-alkali land baseline geomorphic state local feature vector and an alternative target saline-alkali land post-treatment geomorphic state local feature vector from the set of target saline-alkali land baseline geomorphic state local feature vectors and the set of target saline-alkali land post-treatment geomorphic state local feature vectors respectively; An association strength calculation sub-unit is used to measure the causal association strength between the alternative target saline-alkali land baseline geomorphic state local feature vector and the alternative target saline-alkali land post-treatment geomorphic state local feature vector to obtain the closed-loop strength factor of the target saline-alkali land pre-treatment - post-treatment local geomorphic feature causal chain unit.

2. The saline-alkali land information monitoring and treatment system according to claim 1, wherein, The camera is a high-definition camera mounted on a drone.

3. The saline-alkali land information monitoring and governance system according to claim 2, wherein, The topographic and geomorphic feature extraction module is used for: Using a topographic and geomorphic feature extractor based on the FPT model to perform image feature extraction on the pre-treatment overall topographic data and the post-treatment overall topographic data to obtain the target saline-alkali land topographic and geomorphic baseline state feature encoding map and the target saline-alkali land topographic and geomorphic post-treatment state feature encoding map.

4. The saline-alkali land information monitoring and treatment system according to claim 3, characterized in that The association strength calculation sub-unit is used for: Construct a semantic alignment modulation flow field between the alternative target saline-alkali land baseline geomorphic state local feature vector and the alternative target saline-alkali land post-treatment geomorphic state local feature vector; Based on the semantic alignment modulation flow field, perform feature alignment mapping on the alternative target saline-alkali land baseline geomorphic state local feature vector and the alternative target saline-alkali land post-treatment geomorphic state local feature vector to obtain an aligned alternative target saline-alkali land baseline geomorphic state local feature vector and an aligned alternative target saline-alkali land post-treatment geomorphic state local feature vector; Input the aligned alternative target saline-alkali land baseline geomorphic state local feature vector and the aligned alternative target saline-alkali land post-treatment geomorphic state local feature vector into a causal chain unit screening network to obtain the closed-loop strength factor of the target saline-alkali land pre-treatment - post-treatment local geomorphic feature causal chain unit.

5. The saline-alkali land information monitoring and governance system according to claim 4, wherein The strong causal association pairing unit is used for: Based on the comparison between the closed-loop strength factor of the target saline-alkali land pre-treatment - post-treatment local geomorphic feature causal chain unit and a preset threshold, determine whether the alternative target saline-alkali land baseline geomorphic state local feature vector and the alternative target saline-alkali land post-treatment geomorphic state local feature vector form a causal chain unit.

6. The saline-alkali land information monitoring and treatment system according to claim 5, characterized in that, The strong causal joint perception unit is used for: Input each {target saline-alkali land baseline geomorphic state local feature vector, target saline-alkali land post-treatment geomorphic state local feature vector} feature pair in the set of {target saline-alkali land baseline geomorphic state local feature vector, target saline-alkali land post-treatment geomorphic state local feature vector} feature pairs that form a causal chain unit into a causal chain unit semantic dynamic response network to obtain a set of target saline-alkali land pre-treatment - post-treatment geomorphic strong causal joint perception encoding feature vectors; Perform feature aggregation on the set of target saline-alkali land pre-treatment - post-treatment geomorphic strong causal joint perception encoding feature vectors to obtain the target saline-alkali land pre-treatment - post-treatment strong causal joint perception encoding feature map.

7. The saline-alkali land information monitoring and governance system according to claim 6, wherein The governance effect evaluation module is used for: Inputting the pre-governance and post-governance strong causal joint perception coding feature map of the target saline-alkali land into the intelligent governance effect evaluation module based on a classifier to obtain the evaluation result, where the evaluation result is used to indicate whether the governance effect meets the preset requirements.

8. A method for monitoring and managing saline-alkali land information, which is executed by the saline-alkali land information monitoring and management system according to any one of claims 1 to 7, characterized in that, It includes: Obtaining the pre-governance overall terrain data of the target saline-alkali land collected by a camera; Obtaining the post-governance overall terrain data of the target saline-alkali land collected by the camera; Extracting topographic and geomorphic features from the pre-governance overall terrain data and the post-governance overall terrain data to obtain the target saline-alkali land topographic and geomorphic baseline state feature coding map and the target saline-alkali land topographic and geomorphic post-governance state feature coding map; Performing causal alignment analysis based on topographic and geomorphic features on the target saline-alkali land topographic and geomorphic baseline state feature coding map and the target saline-alkali land topographic and geomorphic post-governance state feature coding map to obtain the pre-governance and post-governance strong causal joint perception coding feature map of the target saline-alkali land; Determining the evaluation result of the governance effect based on the pre-governance and post-governance strong causal joint perception coding feature map of the target saline-alkali land.

Images