A landslide early warning method based on multi-morphology target tracking intelligent identification

Abstract

This invention discloses a landslide early warning method based on intelligent recognition of multi-morphological target tracking, comprising: S1, target detection; S2, cyclic detection; S3, real-time data analysis; and S4, issuing a warning. This method utilizes advanced computer vision technology, focusing on efficient and accurate tracking detection of large facilities. Its workflow is scientifically and meticulously designed. First, specific targets are set as the basis for detection, ensuring the accuracy and reliability of the detection process. Subsequently, the device performs multiple cycles of target recognition. This process not only enhances data reliability but also improves detection accuracy. Based on the multiple cycles of target recognition, the device can further generate an overall displacement cloud map. This cloud map intuitively displays the displacement changes of large facilities such as slopes over different time periods, effectively solving the problem of single-target point recognition error tolerance.

Description

Technical Field

[0001] This invention relates to the field of monitoring technology, specifically to a landslide early warning method based on intelligent identification of multi-morphological target tracking. Background Technology

[0002] In the transportation sector, especially in the monitoring of highway slopes, bridges, and buildings, accurate target identification and monitoring are crucial for ensuring safety. With technological advancements, vision-based monitoring methods have gradually been applied, but several problems remain. First, the monitoring range is limited. Traditional detection equipment often uses fixed bases, which are insufficient to meet the comprehensive and efficient monitoring needs of long highway slopes, bridges, and other large structures. Second, monitoring accuracy is affected by environmental factors and the inherent limitations of the equipment itself. Some monitoring technologies are significantly influenced by environmental factors, leading to data errors and affecting accuracy. Furthermore, some monitoring equipment has limited accuracy and is insufficient for monitoring minute deformations. Minor deformations of highway slopes may be early signs of potential disasters, but existing equipment struggles to capture them accurately. Additionally, data processing and transmission issues remain in the current application of monitoring technologies.

[0003] Existing data processing efficiency is low. Deformation monitoring generates a large amount of data, and existing data processing algorithms and systems are inefficient in handling massive amounts of data, making it difficult to analyze and process in real time or quickly, and unable to detect deformation problems in a timely manner. In addition, data transmission is unstable, and the data transmission link in the monitoring system is easily affected by external interference, resulting in unstable or interrupted data transmission, affecting the real-time performance and integrity of the monitoring data. In remote mountainous areas or complex terrain environments, transmission problems are more prominent. Finally, the reliability of existing monitoring technology systems is insufficient. On the one hand, dynamic monitoring is unreliable, and some equipment lacks reliability in dynamic measurement, failing to guarantee accurate and stable acquisition of deformation data under dynamic conditions such as complex traffic loads and environmental changes. On the other hand, it is reflected in weak anti-interference ability and poor resistance to environmental interference, making it easy for monitoring data to be inaccurate or unstable due to external environmental factors. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a landslide early warning method based on intelligent identification of multi-morphological target tracking. This method solves the problems of inaccurate target identification, limited monitoring range, insufficient accuracy, and unstable data processing and transmission in existing monitoring technologies. It can also effectively solve the problem of fault tolerance in the identification of single target points.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a landslide early warning method based on intelligent identification of multi-morphological target tracking, specifically comprising the following steps:

[0006] S1. Target Detection: The detection equipment is deeply integrated with the deformation monitoring system to perform high-precision target detection;

[0007] S2. Cyclic Detection: The movable base performs continuous and periodic detection on targets within the visible range. During the cyclic detection process, the movable base moves according to a preset trajectory and speed, scanning and identifying each target one by one. Each scan forms an independent data point, which is transmitted in real time through signal transmission equipment for subsequent data processing and analysis.

[0008] S3. Real-time data analysis: A multi-target tracker is used to detect, identify, transmit, and process the data of the targets. Real-time transmission technology is used to quickly transmit the data sets obtained from each round of detection to the analysis system. After preliminary processing, the data sets generate dynamic cloud maps.

[0009] S4. Issue a warning: Through analysis, the system marks the sections of the base that may collapse and immediately triggers an alarm at the main controller.

[0010] Preferably, in step S1, the detection device achieves accurate detection of the target and quickly transmits the relevant data, which includes not only the target's position information but also key parameters such as its degree of deformation.

[0011] Preferably, in step S2, since each cycle forms a complete data set, comparing and analyzing the data from multiple cycles, more complex algorithms and technologies are used to ensure that targets in different frames can be accurately associated. This involves multiple steps such as target feature extraction, matching, and tracking. By combining these steps with cyclic detection, a complete and dynamic cloud map is formed, which intuitively displays the motion trajectories and interrelationships of multiple targets at different time points and spatial locations.

[0012] Preferably, in step S3, the cloud map can intuitively display the position and deformation of the target at different time points, providing an important visual reference for subsequent real-time analysis and comparison.

[0013] Preferably, in step S3, during the real-time analysis and comparison process, the data in the cloud map is thoroughly mined and processed to form an overall deformation map within the base segment. This deformation map not only shows the overall deformation trend of the target over a period of time, but also reveals possible abnormal deformation points, providing timely early warning for facility maintenance and upkeep.

[0014] Preferably, in S1, the detection device is a target device, and the target has multi-head target recognition objects that can adapt to different light waves and have a variety of different colors and shapes. Targets of different shapes have unique geometric features.

[0015] Preferably, in S2, the base of the movable platform adopts advanced mobility technology and structural design, enabling the detection equipment to easily cope with various complex terrains and detection environments, and to move quickly and accurately to the designated detection position to perform comprehensive and detailed scanning and detection of structures such as slopes.

[0016] Preferably, in step S3, multiple standard targets with known locations are set within the detection area. These targets can serve as reference points to accurately determine the position and attitude of the device during movement. During the detection process, the device will scan these standard targets sequentially and compare the collected data with the preset target information to calculate the current position and attitude deviation of the device. Then, the collected point cloud map is corrected using this data to offset the displacement error caused by movement.

[0017] Preferably, S2 introduces a multi-point automatic tracking algorithm, which uses an improved combination of SIFT and LBP features to perform non-overlapping irregular block division on the acquired image, and uses an adaptive initial value method for block processing of the SLIC segmentation algorithm; SIFT features are extracted from the obtained image blocks, and rotation-invariant uniform LBP features are extracted from the SIFT feature region to be described to establish feature description.

[0018] Beneficial effects

[0019] This invention provides a landslide early warning method based on intelligent recognition of multi-morphological target tracking. Compared with existing technologies, it has the following advantages: This landslide early warning method based on intelligent recognition of multi-morphological target tracking utilizes advanced computer vision technology, focusing on efficient and accurate tracking detection of large facilities. Its workflow is scientifically and meticulously designed. First, specific targets are set as the basis for detection, ensuring the accuracy and reliability of the detection process. Subsequently, the device performs multiple rounds of target recognition. This process not only enhances the reliability of the data but also improves the detection accuracy. Based on the multiple rounds of target recognition, the device can further generate an overall displacement cloud map. This cloud map intuitively shows the displacement changes of large facilities such as slopes over different time periods, providing important data support for subsequent deformation detection. To further analyze the deformation of the facilities, the device also generates... The device compares displacement cloud maps with a pre-set database to promptly detect abnormal deformations in facilities, providing timely warnings for maintenance and upkeep. Notably, based on existing detection instruments, the device cleverly incorporates a mobile platform. This platform design is highly flexible, including but not limited to rotating wheel sets, moving wheel sets, and hovering drone gimbals. This design allows the device to easily handle various complex terrains and detection environments, fundamentally solving the problem of traditional detection methods' difficulty in large-scale deployment. The device not only possesses high-precision detection capabilities but also excellent flexibility and adaptability. It can perform comprehensive, multi-angle tracking detection of large facilities, promptly detecting deformations and providing strong protection for the safe operation of facilities. It effectively solves the problem of single-target point identification tolerance. It is believed that this device will play an increasingly important role in future facility inspection and maintenance work. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the detection design principle of the present invention;

[0021] Figure 2 This is a schematic diagram of the identification principle of the detection device of the present invention;

[0022] Figure 3 This is a schematic diagram of the detection target structure of the present invention;

[0023] Figure 4 This is a flowchart of the automatic tracking algorithm of the present invention. Detailed Implementation

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] Please see Figure 1-3 This invention provides a technical solution: a landslide early warning method based on intelligent recognition of multi-morphological target tracking, specifically including the following steps:

[0026] S1. Target Detection: The detection equipment is deeply integrated with the deformation monitoring system to perform high-precision target detection;

[0027] S2. Cyclic Detection: The movable base performs continuous and periodic detection on targets within the visible range. During the cyclic detection process, the movable base moves according to a preset trajectory and speed, scanning and identifying each target one by one. Each scan forms an independent data point, which is transmitted in real time through signal transmission equipment for subsequent data processing and analysis.

[0028] S3. Real-time data analysis: A multi-target tracker is used to detect, identify, transmit, and process the data of the targets. Real-time transmission technology is used to quickly transmit the data sets obtained from each round of detection to the analysis system. After preliminary processing, the data sets generate dynamic cloud maps.

[0029] S4. Issue a warning: Through analysis, the system marks the sections of the base that may collapse and immediately triggers an alarm at the main controller.

[0030] Specifically, target detection is the cornerstone of visual multi-target tracking technology. Its core task is to accurately locate and identify targets from images or video frames. This technology is crucial for achieving efficient and accurate visual monitoring and analysis, and can capture and analyze minute deformations of target objects in real time.

[0031] Specifically, cyclic detection is an efficient and reliable detection method that uses a movable base to continuously and periodically detect targets within the visible range. This detection method not only improves the accuracy of the detection, but also ensures the continuity and integrity of the data.

[0032] Specifically, real-time data analysis is the core requirement for multi-target trackers in complex video stream processing. Since the video stream is continuous, the multi-target tracker must complete the detection, identification, data transmission, and processing of targets in a very short time to ensure the real-time performance and accuracy of the analysis.

[0033] In this invention, in step S1, the detection device achieves accurate detection of the target and quickly transmits the relevant data. This data includes not only the target's position information but also key parameters such as its degree of deformation.

[0034] In this invention, in S2, since each cycle forms a complete data set, the data from multiple cycles are compared and analyzed. More complex algorithms and technologies are used to ensure that targets in different frames can be accurately associated. This involves multiple steps such as target feature extraction, matching, and tracking. By combining these steps with cyclic detection, a complete and dynamic cloud map is formed, which intuitively displays the motion trajectory and interrelationship of multiple targets at different time points and spatial locations.

[0035] In this invention, in S3, the cloud map can intuitively display the position and deformation of the target at different time points, providing an important visual reference for subsequent real-time analysis and comparison.

[0036] In this invention, during the real-time analysis and comparison process in S3, the data in the cloud map is deeply mined and processed to form an overall deformation map within the base segment. This deformation map not only shows the overall deformation trend of the target over a period of time, but also reveals possible abnormal deformation points, providing timely early warning for facility maintenance and upkeep.

[0037] Specifically, real-time data analysis is a key capability of multi-target trackers in complex video stream processing. It ensures the real-time nature and accuracy of the data, providing strong support for subsequent deformation analysis and early warning.

[0038] In this invention, in S1, the detection device is a target device. The target has multi-head target recognition objects that can adapt to different light waves and have various colors and shapes. Targets of different shapes have unique geometric features.

[0039] In this invention, in S2, the base of the movable platform adopts advanced mobility technology and structural design, enabling the detection equipment to easily cope with various complex terrains and detection environments, and to move quickly and accurately to the designated detection position to perform comprehensive and detailed scanning and detection of structures such as slopes.

[0040] In this invention, in step S3, multiple standard targets with known locations are set within the detection area. These targets can serve as reference points to accurately determine the position and attitude of the device during movement. During the detection process, the device sequentially scans these standard targets and compares the collected data with the preset target information to calculate the current position and attitude deviation of the device. Then, these data are used to correct the collected point cloud map to offset the displacement error caused by movement.

[0041] Specifically, targets of different shapes (such as crosses, polygons, etc.) have unique geometric features that can provide rich information during target recognition. For example, cross-shaped targets have obvious horizontal and vertical intersecting structures, while polygonal targets have multiple sides and corners. These unique shape features can help recognition algorithms identify targets more accurately and reduce misjudgments. In complex monitoring environments, targets of different shapes can avoid recognition confusion caused by similar shapes. For example, in scenarios with multiple similar targets, using targets of different shapes can give each target a unique identifier, making it easier to distinguish them accurately. In addition, targets of different shapes present different visual features at different angles. For example, a circular target looks circular at any angle, while a polygonal target will present different combinations of sides and corners at different angles. This allows for improved overall recognition accuracy when monitoring from multiple angles by combining targets of various shapes, effectively solving the problem of single-target recognition tolerance.

[0042] Targets of different appearance, shape and color can be selected according to different usage scenarios and monitoring targets. In some monitoring scenarios, targets of a certain shape may be more suitable. For example, in an environment with a large number of circular interference objects (such as pipes, wellheads, etc.), square or polygonal targets can be more easily distinguished from the surrounding environment. In an environment with more square structures, round or irregularly shaped targets may be more recognizable.

[0043] In this invention, S2 introduces a multi-point automatic tracking algorithm, which uses an improved combination of SIFT and LBP features to perform non-overlapping irregular block division on the acquired image. The SLIC segmentation algorithm is processed by using an adaptive initial value. SIFT features are extracted from the obtained image blocks, and rotation-invariant uniform LBP features are extracted from the SIFT feature region to be described to establish a feature description.

[0044] Specifically, in S4, the alarm can quickly attract the attention of relevant personnel so that flow restriction measures can be taken in a timely manner and the area can be maintained, thereby effectively preventing landslides and ensuring safety.

[0045] Furthermore, any content not described in detail in this specification is existing technology known to those skilled in the art.

[0046] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0047] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A landslide early warning method based on intelligent recognition of multi-morphological target tracking, characterized in that: Specifically, the following steps are included: S1. Target Detection: The detection equipment is deeply integrated with the deformation monitoring system to perform high-precision target detection. The detection equipment is a target device. The target has multi-head target recognition objects that can adapt to different light waves and have various colors and shapes. The targets of different shapes have unique geometric features. The targets are cross-shaped and polygonal. The cross-shaped target has obvious horizontal and vertical intersecting structure, and the polygonal target has multiple sides and corners. The polygonal target is set at the end of the cross-shaped target. The specific steps of the target detection are to first identify the cross shape, then calculate the coordinates of the center point by measuring the coordinates of the points on the horizontal bars around the shape, and then compare them with the coordinates of the points in the middle of the cross shape to calculate the mean value of the cross shape and obtain the coordinates as b. To identify polygonal shapes, the coordinates of the center point are calculated by measuring the coordinates of points around the shape. These coordinates are then compared with the coordinates of points in the center of the shape. The average coordinates of the four polygonal shapes are calculated and denoted as a1, a2, a3, and a4. The center point coordinates of a1, a2, a3, and a4 are calculated and denoted as a0. A0 is compared with b, and the final target center point coordinates are calculated as c. S2. Cyclic Detection: The movable base performs continuous and periodic detection on targets within the visible range. During the cyclic detection process, the movable base moves according to a preset trajectory and speed, scanning and identifying each target one by one. Each scan forms an independent data point, which is transmitted in real time through signal transmission equipment for subsequent data processing and analysis. S3. Real-time data analysis: A multi-target tracker is used to detect, identify, transmit, and process the data of the targets. Real-time transmission technology is used to quickly transmit the data sets obtained from each round of detection to the analysis system. After preliminary processing, the data sets generate dynamic cloud maps. S4. Issue a warning: Through analysis, the system marks the sections of the base that may collapse and immediately triggers an alarm at the main controller.

2. The landslide early warning method based on multi-morphological target tracking and intelligent recognition according to claim 1, characterized in that: In step S1, the detection device achieves accurate detection of the target and quickly transmits the relevant data. This data includes not only the target's position information but also key parameters related to its deformation degree.

3. The landslide early warning method based on multi-morphological target tracking and intelligent recognition according to claim 1, characterized in that: In S2, since each cycle forms a complete data set, the data from multiple cycles are compared and analyzed. More complex algorithms and technologies are used to ensure that targets in different frames can be accurately associated. This involves multiple steps such as target feature extraction, matching, and tracking. By combining these steps with cycle-based detection, a complete and dynamic cloud map is formed, which intuitively displays the motion trajectories and interrelationships of multiple targets at different time points and spatial locations.

4. The landslide early warning method based on multi-morphological target tracking and intelligent recognition according to claim 1, characterized in that: In S3, the cloud map can intuitively display the position and deformation of the target at different time points, providing an important visual reference for subsequent real-time analysis and comparison.

5. A landslide early warning method based on intelligent recognition of multi-morphological target tracking according to claim 1, characterized in that: In S3, during the real-time analysis and comparison process, the data in the cloud map is deeply mined and processed to form an overall deformation map within the base segment. This deformation map not only shows the overall deformation trend of the target over a period of time, but also reveals possible abnormal deformation points, providing timely early warning for facility maintenance and upkeep.

6. The landslide early warning method based on multi-morphological target tracking and intelligent recognition according to claim 1, characterized in that: In S2, the base of the movable platform adopts advanced mobility technology and structural design, enabling the detection equipment to easily cope with various complex terrains and detection environments, and to move quickly and accurately to the designated detection position to perform comprehensive and detailed scanning and detection of the slope structure.

7. A landslide early warning method based on intelligent recognition of multi-morphological target tracking according to claim 1, characterized in that: In step S3, multiple standard targets with known locations are set within the detection area. These targets can serve as reference points to accurately determine the position and attitude of the device during movement. During the detection process, the device will scan these standard targets sequentially and compare the collected data with the preset target information to calculate the current position and attitude deviation of the device. Then, the collected point cloud map is corrected using this data to offset the displacement error caused by movement.

8. A landslide early warning method based on intelligent recognition of multi-morphological target tracking according to claim 1, characterized in that: The S2 section introduces a multi-point automatic tracking algorithm, which uses an improved combination of SIFT and LBP features to perform non-overlapping irregular block division on the acquired image. The SLIC segmentation algorithm is processed by using an adaptive initial value. SIFT features are extracted from the obtained image blocks, and rotation-invariant uniform LBP features are extracted from the SIFT feature regions to be described to establish feature descriptions.

Images