A floating garbage multi-modal intelligent monitoring method

Abstract

This invention discloses a multimodal intelligent monitoring method for floating debris, relating to the fields of water environment monitoring and intelligent environmental sensing technology. The method includes the following steps: acquiring water surface change information and water flow change information at a monitoring section, and organizing them according to a unified time sequence to construct a continuous change sequence reflecting the relationship between changes in the position of water surface targets and changes in water flow. This invention constructs a velocity distribution representation of the monitoring section by associating water surface change information and water flow change information, making the relationship between changes in the movement of water surface targets and the propulsion of water flow a unified reference. This distinguishes interference changes from the perspective of motion patterns, improving the accuracy of floating debris identification. Simultaneously, by extracting target change segments consistent with water flow changes and continuously tracking their positions and counting the number of occurrences, continuous monitoring and automatic statistics of floating debris are achieved.

Description

Technical Field

[0001] This invention relates to the field of water environment monitoring and intelligent environmental sensing technology, specifically to a multimodal intelligent monitoring method for floating debris. Background Technology

[0002] Multimodal intelligent monitoring of floating debris refers to a technical approach that automatically identifies and monitors floating debris on the water surface in river or aquatic environments by integrating multiple monitoring methods. This process begins with the continuous acquisition of video data from the water surface using surveillance cameras. Suspected floating objects in the video footage are extracted, tracked, and their positional changes are calculated. Simultaneously, radar velocity measurement equipment acquires water flow velocity information at the monitoring cross-section of the river, and this information is combined with the hydrodynamic parameters of the river cross-section to construct the velocity distribution of the monitoring section. Subsequently, the target's motion velocity extracted from the video is correlated with the theoretical velocity of the water flow at the corresponding location. By determining the degree of matching between the target's lateral and longitudinal motion velocities and the river flow velocity components, the system achieves automatic identification, continuous tracking, and counting of floating debris targets. An alarm is triggered when the number of debris reaches a preset threshold, thus realizing continuous, automated, and intelligent monitoring of floating debris in rivers.

[0003] The existing technology has the following shortcomings: When monitoring floating debris in real-world aquatic environments, existing technologies generally suffer from insufficient accuracy due to the complexity of the surface environment and numerous interfering factors. Manual inspections require significant manpower and time, resulting in low efficiency and difficulty in continuous monitoring over long distances. Furthermore, monitoring results are easily affected by weather changes, lighting conditions, and differences in the experience of inspectors, leading to missed detections or statistical errors. While video monitoring allows for unattended operation, surface ripples, wave disturbances, aquatic plants, duckweed, and light reflections are visually similar to floating debris, making image recognition prone to misjudgment. Additionally, existing radar monitoring technologies typically only acquire localized, single-point flow velocity information, failing to reflect the overall water flow distribution across the monitored section and hindering effective correlation with video monitoring results. This can affect the judgment of the true movement of surface targets, ultimately leading to low reliability in floating debris identification and counting.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a multimodal intelligent monitoring method for floating debris to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a multimodal intelligent monitoring method for floating debris, comprising the following steps: Information on water surface changes and water flow changes at monitoring sections is acquired and organized in a uniform time sequence to construct a continuous change sequence reflecting the relationship between changes in the target water surface location and changes in water flow. The water flow change information corresponding to each time position is read around the continuous change sequence, and the water flow change information is sorted horizontally and vertically in combination with the hydrodynamic characteristics of the monitoring section, so as to construct a velocity distribution expression covering the monitoring section in the continuous change sequence. Based on the velocity distribution, the change process of the water surface target position in the continuous change sequence is read, and the change of the target's lateral motion and longitudinal motion are formed according to the change process of the water surface target position. Based on the changes in the lateral and longitudinal motion of the target, the corresponding positions of the velocity distribution are returned. The relationship between the changes in the lateral and longitudinal motion of the target and the water flow propulsion is compared and organized, and the target change segments that are consistent with the changes in the water flow are extracted. Based on the target change segment, the subsequent changes in the continuous change sequence are read, and the identified floating targets are continuously tracked and the number of times are sorted out, so as to complete the continuous monitoring and statistics of floating garbage.

[0007] Preferably, the steps for obtaining water level change information and water flow change information at the monitoring section and constructing a continuous change sequence are as follows: Acquire information on water surface changes and water flow changes within the monitoring section area, and perform unified time-identification processing on the information on water surface changes and water flow changes; Based on a unified time identifier, the information on water surface changes and water flow changes is organized in chronological order to form a time series record in which the information on water surface changes and water flow changes correspond one-to-one. Based on the time series records, water surface change information is read and spatial location is recorded to establish the position coordinate record of the water surface target in the monitoring section, and the water surface target position change trajectory is formed by combining the time series records; By integrating water surface change information, water surface target position change information, and water flow change information from time series records around the trajectory of water surface target position change, a continuous change sequence reflecting the relationship between water surface target position change and water flow change is formed.

[0008] Preferably, the process of constructing a velocity distribution representation covering the monitoring section based on a continuously changing sequence is as follows: Read the water flow change information corresponding to each time position in the continuous change sequence, and establish the correspondence between the water flow change information and the spatial position of the monitoring section around the time position. At the same time, use the radar velocity measurement position as the reference position to record the actual single-point flow velocity vector, and form a water flow change recording structure with horizontal and vertical components. The information on water flow changes in a continuous sequence is organized laterally based on the water flow change record structure, and a theoretical expression of the lateral flow velocity within the lateral range of the monitoring section is formed based on the lateral component at the radar velocity measurement location and the distance relationship between the spatial locations of the monitoring section. Based on the transverse theoretical velocity expression, the water flow change information in the continuous change sequence is longitudinally organized, and combined with the river roughness coefficient, water depth information and river water surface gradient to form a longitudinal theoretical velocity expression within the longitudinal range of the monitoring section. The theoretical flow velocity expressions for the transverse and longitudinal directions are combined and recorded in a continuous sequence to form a flow velocity distribution expression covering the spatial range of the monitoring section.

[0009] Preferably, the transverse theoretical velocity expression and the longitudinal theoretical velocity expression are recorded in a continuous change sequence according to their corresponding time and spatial positions, and a joint recording structure of transverse theoretical velocity information and longitudinal theoretical velocity information is formed around the spatial range of the monitoring section, thereby forming a complete velocity distribution expression in the continuous change sequence.

[0010] Preferably, the steps for expressing the lateral and longitudinal motion changes of the target based on the velocity distribution are as follows: Read the water surface change records in the continuous change sequence, and record the coordinates of the water surface target position around the spatial coordinates of the monitoring section. At the same time, combine the flow velocity distribution to establish the correspondence between the water surface target position change records and the spatial position of the monitoring section. The system records the changes in the position of a target on the water surface by reading the spatial coordinates of the target at adjacent time points, and then generates a record of the target's lateral movement based on the changes in its lateral spatial coordinates. Based on the record of changes in the position of the water surface target, the spatial coordinates of the water surface target at adjacent time positions are read, and a record of the longitudinal motion change of the target is formed based on the changes in the longitudinal spatial coordinates; The records of lateral and longitudinal motion changes around the target are combined in a continuous sequence of changes and combined with the velocity distribution to form a structure for expressing the motion changes of the water surface target.

[0011] Preferably, the target's lateral movement change record and the target's longitudinal movement change record are linked and organized through a continuous change sequence, and a continuous movement trajectory record is formed around the target's position coordinates on the water surface. At the same time, the correspondence between the target's movement change on the water surface and the spatial position of the monitoring section is established by combining the flow velocity distribution.

[0012] Preferably, the steps for extracting the target change segment that is consistent with the water flow change are as follows: Read the target's lateral and longitudinal motion change records in the continuous change sequence, and obtain the corresponding lateral and longitudinal water flow change records in the velocity distribution expression around the target's spatial position on the water surface, forming a spatial correspondence record between the target's motion change and the water flow change. The lateral movement changes of the target and the lateral water flow changes are compared and organized based on the spatial correspondence records. At the same time, the longitudinal movement changes of the target and the longitudinal water flow changes are compared and organized to form a motion change comparison record in a continuous change sequence. Based on the motion change comparison record, read the time sequence record in the continuous change sequence, and form the target motion change segment record around the horizontal motion change record and the vertical motion change record; The records of the target movement change section are integrated into a continuous change sequence to form a target change section that is consistent with the water flow change.

[0013] The preferred steps for continuous monitoring and statistics of floating debris are as follows: Read the water surface change records corresponding to the target change segment in the continuous change sequence, and continue to read the water surface change records in the continuous change sequence around the last time position of the target change segment to form a continuous record of the spatial position change of the floating target; Continuous position tracking is performed around the spatial position records of floating targets in a continuously changing sequence, and a target number record structure is established based on the changes in the spatial position of the floating targets, while simultaneously forming a spatial position trajectory record corresponding to the target number; Based on the target number record structure, the entry and exit records of floating targets in the monitoring section space are read, and the number of floating targets is organized and recorded around the entry and exit records. The records are integrated and recorded in a continuous sequence of changes, including target number records, spatial location trajectory records, entry records, and exit records, to form a structure for continuous monitoring and statistical recording of floating debris.

[0014] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention unifies the temporal processing of water surface change information and water flow change information, enabling the changes in the position of water surface targets and the changes in water flow to be correlated within the same time series. Based on this, a flow velocity distribution representation covering the monitoring section is constructed, allowing a direct correspondence between the motion changes of water surface targets and the corresponding water flow propulsion status. This enables the identification of whether water surface targets move with the water flow through continuous changes. In this way, the motion characteristics of floating targets and water flow characteristics form a unified reference relationship, distinguishing from interference changes caused by water ripples, floating vegetation, and water surface reflections at the level of motion patterns. This improves the accuracy of water surface target identification and makes the identification process of floating debris more stable.

[0015] This invention extracts target change segments that are consistent with water flow changes and continuously tracks and records the positions and frequency of identified floating targets within a continuous change sequence. This allows for a complete record of the movement trajectory of floating targets within the monitoring section. Simultaneously, it continuously counts floating targets entering the monitoring section, thereby achieving continuous monitoring and automatic statistics of floating debris. By uniformly recording the appearance process of floating targets within the monitoring section, the changes in the quantity of floating debris can be compiled into continuous statistical results in chronological order, providing stable data for aquatic environment management and enabling the floating debris monitoring process to operate continuously under unattended conditions. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0017] Figure 1 This is a flowchart of a multimodal intelligent monitoring method for floating debris according to the present invention. Detailed Implementation

[0018] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0019] This invention provides, for example Figure 1 The method for multimodal intelligent monitoring of floating debris, as shown, includes the following steps: Information on water surface changes and water flow changes at monitoring sections is acquired and organized in a uniform time sequence to construct a continuous change sequence reflecting the relationship between changes in the target water surface location and changes in water flow. By continuously acquiring and uniformly processing information on changes in water surface and flow conditions within the monitoring section, a continuous correlation record is formed between the changes in the target water surface location and the changes in flow within the same time dimension, thus creating a continuous change sequence that reflects the relationship between changes in the target water surface location and changes in flow. The specific implementation steps are as follows: Information on water surface and flow changes is continuously acquired within the river monitoring section area, and the acquired information is processed with a unified time stamp. In practice, monitoring cameras are installed along the riverbank, covering the entire monitoring section, to continuously record changes in the water surface. The cameras acquire images continuously at fixed time intervals, generating a complete water surface image at each acquisition point, thus forming a continuous image record arranged chronologically. Simultaneously, a water flow velocity measuring device is installed at the corresponding location on the monitoring section to continuously record changes in water flow velocity. The water flow velocity measuring device records the water flow velocity information at the current time point at the same time rhythm as the image acquisition points, and adds a corresponding time stamp to the recorded water flow velocity information. After completing the image and water flow velocity recordings, each water surface image data and the water flow velocity information recorded at the same time point are uniformly stamped, ensuring that both water surface and flow change information have a unified time stamp. This establishes a temporal correspondence between the two information sources, allowing subsequent data processing to be based on a unified time point.

[0020] After establishing a unified time stamp, the water surface change information and water flow change information are organized chronologically, ensuring that the water surface change process and the water flow change process are arranged in a corresponding time series. In practice, all water surface images acquired by monitoring cameras are arranged according to the time stamp, ensuring that each water surface image corresponds to a specific time node. Simultaneously, the water flow velocity information recorded by the water flow velocity measuring device is arranged according to the same time stamp, ensuring that each time node has corresponding water flow change information. During the organization process, a continuous time node sequence is established based on the time stamp, and water surface image data and corresponding water flow change information are recorded simultaneously at each time node, thus establishing a one-to-one correspondence between water surface change information and water flow change information on the same time axis. Through this organization method, a data sequence containing water surface state records and water flow state records is formed on a continuous time axis, enabling the water surface image records to be synchronized with the water flow change information at the same time node. This eliminates the offset effect of different information sources during the time recording process, ensuring that the water surface change records and water flow change records are consistent in the time dimension.

[0021] After forming a unified time-series data sequence, the target position changes of water surface changes in the continuous time series are recorded, ensuring a complete record of the positional changes of water surface targets within the monitoring section. In practice, the spatial position of each water surface image in the time series is recorded, and the coordinates of the water surface area in the image are calibrated, ensuring that each position in the water surface area corresponds to its actual position within the river monitoring section. After calibration, the target position is recorded for each time point in the water surface image, recording the target's coordinates within the monitoring section. The positional changes of the same target at adjacent time points are continuously recorded, forming a trajectory of the target's positional changes over consecutive time points. Simultaneously, the water flow change information at the corresponding time point is synchronously associated with the target position record, ensuring that each target position record includes the water flow change status at that time point. This creates a data recording structure where target position changes and water flow changes coexist in the continuous time series, allowing the target position change process to be expressed synchronously with the water flow change process.

[0022] After recording the target position changes, the water surface change information, target position change information, and flow change information from consecutive time points are integrated to form a continuous change sequence reflecting the relationship between target position changes and flow changes. In practice, a unified time point is used as the main thread, and the water surface image records, target position records, and flow change records for each time point are integrated and organized to ensure that each time point contains complete water surface and flow state information. The time points are connected sequentially to form a continuous change sequence, allowing the position change process of the water surface target in the monitoring section to be continuously expressed chronologically. Simultaneously, the correlation between target position change information and corresponding flow change information at each time point is preserved in the continuous change sequence, enabling the simultaneous reading of both the target position change state and the corresponding flow change state at any given time point. This creates a unified recording structure for the movement and flow changes of the water surface target in the monitoring section. By integrating the above methods, a continuous change sequence is formed that can simultaneously reflect the changes in the target position on the water surface and the changes in water flow. This allows subsequent steps to read the target position change information and the corresponding water flow change information at any time point based on the continuous change sequence, thus providing a unified data foundation for the subsequent construction of flow velocity distribution representation and floating debris identification and processing.

[0023] The water flow change information corresponding to each time position is read around the continuous change sequence, and the water flow change information is sorted horizontally and vertically in combination with the hydrodynamic characteristics of the monitoring section, so as to construct a velocity distribution expression covering the monitoring section in the continuous change sequence. Based on the existing continuous change sequence, the water flow change information corresponding to each time position in the continuous change sequence is further processed. Combined with the hydrodynamic characteristics of the river monitoring section, the water flow change information is processed both horizontally and vertically. This results in a velocity distribution representation covering the entire monitoring section within the continuous change sequence, enabling a unified representation of the water flow state at any location through the continuous change sequence, and providing a reference relationship for subsequent analysis of water surface target movement changes. The specific implementation steps are as follows: In this step, the water flow change information at each time point in the continuous change sequence is read sequentially, and a correspondence between the time point and the spatial location of the monitoring section is established. In the continuous change sequence formed in the previous steps, water surface change information and water flow change information are recorded at each time point. Following the time sequence of the continuous change sequence, the water flow change information corresponding to each time point is read sequentially, and the water flow velocity state at that time point is recorded. Simultaneously, during the reading of water flow change information, it is correlated with the spatial location in the monitoring section, ensuring that each water flow change record corresponds to the actual coordinates of the monitoring section. Specifically, the radar velocity measurement position in the monitoring section is used as the basic reference position. The actual single-point velocity vector collected at the radar velocity measurement position is recorded, and this single-point velocity vector is decomposed into horizontal and vertical components. The horizontal component represents the water flow velocity change in the river channel width direction, and the vertical component represents the water flow velocity change in the river channel flow direction. This establishes a correspondence between the time point, spatial location, and water flow change information in the continuous change sequence. In this way, a water flow change information recording structure with time sequence as the main line is formed in the continuous change sequence, so that each time position has a corresponding spatial water flow change record, thereby providing a basic data source for the subsequent construction of the expression of flow velocity distribution at monitoring sections.

[0024] After reading the water flow change information and establishing the correspondence between time and spatial locations, the water flow change information in the continuous change sequence is horizontally processed to form a continuous expression of the water flow change information within the horizontal range of the monitoring section. In the specific implementation process, the width direction of the monitoring section is used as the horizontal processing direction, the water flow change information recorded in the continuous change sequence is used as the basic data, and the actual single-point velocity vector collected at the radar velocity measurement location is used as the basic reference value for horizontal processing. Combined with the distance relationship between each spatial location of the monitoring section and the radar velocity measurement location, the water flow change information within the horizontal range of the monitoring section is extended and expressed. During the horizontal processing, the theoretical horizontal velocity expression at any location on the monitoring section is established using the following formula: In the formula, Indicates any location of the monitoring section The theoretical transverse velocity at a given location is a parameter used to describe the velocity component of the water flow along the width of the river channel. This represents the lateral component of the actual single-point flow velocity collected at the radar velocity measurement location. This lateral component indicates the motion state of the water flow in the direction of the river channel width at the radar velocity measurement location. This indicates the spatial coordinates of the radar velocity measurement location within the monitoring section. This represents the spatial coordinates of any location within the monitoring section. The effective radiation radius of radar velocity measurement is used to describe the spatial influence range of radar velocity measurement location on water flow changes in the surrounding area. This indicates the actual water depth at any location on the monitoring section; This represents the actual water depth at the radar velocity measurement location. The above formula allows for the expansion of single-point water flow change information collected outside the radar velocity measurement location within the transverse range of the monitoring section. This creates a water flow change record structure covering the transverse range of the monitoring section within a continuous change sequence, enabling theoretical transverse flow velocity information to be obtained at any transverse location of the monitoring section.

[0025] After completing the horizontal processing, the flow change information in the continuous variation sequence is further processed vertically to create a continuous expression of the flow change information along the flow direction of the monitoring section. In practice, the flow direction of the monitoring section is used as the vertical processing direction, the flow change information recorded in the continuous variation sequence is used as the basic data, and river hydrodynamic characteristic parameters are further introduced based on the horizontal processing results to create a unified expression of the flow change information within the longitudinal range of the monitoring section. During the vertical processing, the theoretical longitudinal flow velocity is established at any location on the monitoring section using the following formula: In the formula, It refers to any location of the monitoring section. The longitudinal theoretical flow velocity at that location, This represents the longitudinal component of the actual single-point flow velocity collected at the radar velocity measurement location. This longitudinal component represents the motion state of the water flow in the direction of the river flow. It represents the roughness coefficient of the river channel at any location on the monitoring section, and is used to reflect the influence of the riverbed on the state of water flow. This represents the river channel roughness coefficient at the radar velocity measurement location. This represents the gradient of the river surface, used to describe the changes in river flow along the direction of flow. Through the above formula, the single-point flow change information collected at the radar velocity measurement location can be extended and expressed within the longitudinal range of the monitoring section, combined with the river's hydrodynamic characteristics. This forms a flow change record structure covering the longitudinal range of the monitoring section within a continuous change sequence, allowing theoretical longitudinal flow velocity information to be obtained at any longitudinal location of the monitoring section.

[0026] After completing the horizontal and vertical processing, all flow change information in the continuous change sequence is integrated to construct a velocity distribution representation covering the entire monitoring section. In practice, the horizontal theoretical velocity information obtained through horizontal processing and the vertical theoretical velocity information obtained through vertical processing are combined and recorded in the continuous change sequence. This ensures that each time point in the continuous change sequence includes not only water surface and flow change information, but also horizontal and vertical theoretical velocity representations at any spatial location within the monitoring section. Through this integration method, horizontal and vertical theoretical velocity information can be obtained at any location within the monitoring section in the continuous change sequence, thus forming a velocity distribution representation structure covering the entire monitoring section. This allows the continuous change sequence to fully reflect the flow change state within the monitoring section and provides a stable reference basis for subsequent analysis of water surface target movement changes.

[0027] Based on the velocity distribution, the change process of the water surface target position in the continuous change sequence is read, and the change of the target's lateral motion and longitudinal motion are formed according to the change process of the water surface target position. After the aforementioned steps have resulted in a velocity distribution representation covering the monitoring section, this embodiment further organizes the spatial positional changes of the water surface target within the monitoring section based on the water surface change records in a continuous sequence. Combined with the already formed velocity distribution representation, this allows for the formation of an analytical record of the water surface target's motion at different spatial locations in both the lateral and longitudinal directions. This provides a basis for subsequent analysis of the water flow propulsion relationship. The specific implementation process is as follows.

[0028] First, water surface change records are read sequentially from the established continuous change sequence, and a record of water surface target location changes is established within the spatial coordinate framework of the velocity distribution. In the previous steps, the continuous change sequence already contains water surface change information arranged chronologically and velocity distribution information corresponding to the spatial locations. In this step, the chronological order of the continuous change sequence is used as the main reading guideline. Water surface change records at each time point are read, and the location of the water surface target is marked within the spatial coordinate range of the monitoring section. Specifically, the spatial coordinates within the monitoring section are used as a unified representation benchmark. When reading the water surface change records, the spatial coordinates of the water surface target at the current time point are recorded as follows: And record the spatial coordinates of the same water surface target at the next time position in the continuously changing sequence. By employing the above method, a record of the spatial position changes of a water surface target at different time points is generated within a continuous change sequence, enabling the movement of the water surface target within the monitoring section to be expressed through continuous spatial coordinate changes. Simultaneously, during the recording of the spatial position changes of the water surface target, the flow velocity distribution information corresponding to each spatial position is synchronously associated with the water surface target position record. This ensures that the water surface target position change record and the corresponding water flow state within the monitoring section form a unified expression, thereby establishing a spatial correspondence between the water surface target position change record and the flow velocity distribution expression within the continuous change sequence.

[0029] After recording the changes in the position of water surface targets, the spatial positional changes recorded in the continuous change sequence are processed to analyze lateral motion changes, enabling a unified representation of the water surface target's motion along the river channel width. In practice, the lateral direction in the spatial coordinates of the monitoring section is used as the processing direction. The spatial coordinates of water surface targets at adjacent positions in the continuous change sequence are compared and processed. By recording the lateral coordinate changes of the water surface targets in the continuous position records, the lateral motion changes of the water surface targets can be expressed through positional changes. To make the lateral motion changes quantifiable, the lateral coordinate changes of the water surface targets in the continuous change sequence are decomposed into velocity, and the actual lateral velocity of the target is represented by the following formula: In the formula, This indicates the location of the water surface target at any spatial position within the monitoring section. The actual lateral flow velocity at that location, This represents the lateral spatial coordinates of a surface target in the previous position record within a continuous sequence of changes. This indicates the lateral spatial coordinates of the water surface target in the subsequent position record. This represents the time interval between adjacent position records in a continuous change sequence. Using this method, the lateral positional changes of a water surface target within the monitoring section are transformed into lateral motion change records, thus enabling a unified representation of the water surface target's motion along the river channel width in a continuous change sequence. Through this method of expressing lateral motion change, the lateral movement of the water surface target at different spatial locations can be recorded in continuous numerical form, providing a basis for subsequent target motion change analysis.

[0030] After processing the lateral motion changes, the longitudinal motion changes of the water surface target positions in the continuous change sequence are further processed to ensure a unified representation of the water surface target's motion changes along the river flow direction. In practice, the longitudinal direction in the monitoring section's spatial coordinates is used as the processing direction. The spatial position changes of the water surface target recorded in the continuous change sequence are recorded longitudinally. By recording the longitudinal coordinate changes of the water surface target in the continuous position records, the motion changes of the water surface target along the river flow direction can be expressed through spatial position changes. To make the longitudinal motion changes quantifiable, the longitudinal coordinate changes of the water surface target in the continuous change sequence are decomposed into velocity, and the actual longitudinal velocity of the target is expressed using the following formula: In the formula, This indicates the location of the water surface target at any spatial position within the monitoring section. The actual longitudinal velocity at that point This represents the vertical spatial coordinates of a surface target in the previous position record within a continuous sequence of changes. This represents the longitudinal spatial coordinates of the water surface target in the subsequent position record. Through this representation, the longitudinal positional change of the water surface target within the monitoring section is transformed into a longitudinal motion change record, enabling the movement of the water surface target along the river flow direction to form a continuous table within a continuous sequence of changes. This longitudinal motion change table method allows the longitudinal movement of the water surface target at different spatial locations to be recorded in continuous numerical form, thus providing a basis for subsequent target motion change analysis.

[0031] After organizing the lateral and longitudinal motion changes, the records of water surface target position changes, lateral motion changes, and longitudinal motion changes in the continuous change sequence are integrated to form a complete representation of the water surface target's motion changes within the monitoring section. In practice, the actual lateral flow velocity obtained through the lateral motion change formula and the actual longitudinal flow velocity obtained through the longitudinal motion change formula are combined and recorded in the continuous change sequence. Each water surface target position change record simultaneously includes both lateral and longitudinal motion change information and is spatially correlated with the corresponding position in the flow velocity distribution expression. This allows the motion changes of the water surface target within the monitoring section to be uniformly expressed through lateral and longitudinal motion changes. Through this integration, a motion change expression structure is formed in the continuous change sequence, including records of water surface target position changes, lateral motion changes, and longitudinal motion changes. This allows the motion process of the water surface target within the monitoring section to be uniformly described through lateral and longitudinal motion changes, providing a complete foundation for subsequent analysis of water flow propagation relationships and identification of floating debris.

[0032] Based on the changes in the lateral and longitudinal motion of the target, the corresponding positions of the velocity distribution are returned. The relationship between the changes in the lateral and longitudinal motion of the target and the water flow propulsion is compared and organized, and the target change segments that are consistent with the changes in the water flow are extracted. Based on the existing records of the target's lateral and longitudinal motion changes, the spatial correlation between the target's motion changes and the flow velocity distribution within the monitoring section is analyzed. This process compares and organizes the relationship between target motion changes and water flow propulsion, ensuring a unified representation of the target's motion changes in a continuous sequence and the corresponding water flow changes. This allows for the extraction of target change segments that are consistent with water flow changes from the continuous sequence. Through this process, disturbances such as water ripples, floating vegetation, and water reflections can be distinguished from the target change records, enabling the independent recording of the target's actual movement with the water flow. The specific implementation steps are as follows: Based on the changes in the target's lateral and longitudinal motion, the corresponding positions in the velocity distribution are retrieved, and the spatial correlation of the water surface target's position within the monitoring section is organized. In the preceding steps, the spatial position changes of the water surface target at different time locations have been recorded in the continuous change sequence, simultaneously forming records of the target's lateral and longitudinal motion changes. In this step, the spatial position of the water surface target recorded in the continuous change sequence is used as the basic reference position. The corresponding water flow change record in the velocity distribution is searched, thereby establishing a spatial correspondence between the target's motion changes and the water flow changes. In specific implementation, the spatial position of the water surface target in the continuous change sequence is recorded as the position coordinates in the monitoring section, and the corresponding lateral and longitudinal velocity information is read from the velocity distribution. This ensures that each target motion change record can correspond to the water flow change record at the same spatial location, thus forming a comparative record structure in the continuous change sequence that includes information on the target's lateral motion changes, target longitudinal motion changes, and corresponding water flow changes.

[0033] To make the spatial correspondence clearer, a record table of the correspondence between the target motion changes and the water flow changes can be established in the continuous change sequence, as shown below.

[0034] Table 1: Record of Target Motion Changes and Corresponding Water Flow Changes In the aforementioned record table, each time sequence number represents a time position in a continuous change sequence; the horizontal and vertical coordinates of the target position represent the spatial position of the water surface target within the monitoring section; the changes in the target's horizontal and vertical motion represent the motion changes of the water surface target between consecutive time positions; and the corresponding horizontal and vertical water flow changes represent the water flow change state corresponding to that spatial position read from the flow velocity distribution representation. By establishing this corresponding record table, a clear spatial correspondence is formed between the target motion changes and water flow changes, thus providing a basic data structure for subsequent comparison and processing.

[0035] After establishing the spatial correspondence, the relationship between the target's lateral and longitudinal motion changes and the water flow propulsion is compared and organized to ensure a continuous expression of the relationship between the target's motion and the water flow in a continuous sequence. In practice, the time sequence within the continuous sequence is used as the main thread. Each time point in the record table is organized, comparing the target's lateral motion changes with the corresponding lateral water flow changes, and simultaneously comparing the target's longitudinal motion changes with the corresponding longitudinal water flow changes, ensuring a corresponding record of target motion and water flow changes at each time point. This organization method establishes a continuous correlation between the target's motion and water flow changes, creating a complete motion change comparison structure within the continuous sequence. This structure provides a clear visual representation of the changing relationship between the target's motion and the direction of water flow propulsion during continuous movement.

[0036] To make the comparison and organization process clearer, a target motion change comparison table can be further established in the continuous change sequence, as shown below.

[0037] Table 2: Comparison Table of Target Motion Changes In the aforementioned comparison table, by recording the relationship between changes in the target's lateral movement and the corresponding lateral water flow, and simultaneously recording the relationship between changes in the target's longitudinal movement and the corresponding longitudinal water flow, a continuous comparative record of the relationship between target movement changes and water flow propulsion can be formed. This comparison table can visually present the movement changes of a water surface target in a continuous sequence, allowing the changes in the relationship between target movement changes and water flow propulsion to be expressed in a continuous recording manner.

[0038] After completing the analysis and comparison of motion changes, the target motion change records in the continuous change sequence are segmented to identify target change segments consistent with water flow changes. In practice, based on the continuous time sequence numbers in the target motion change comparison table, continuous time series where the lateral and longitudinal motion change comparison results are simultaneously consistent are organized. These continuous time series records are divided into target change segments, ensuring that the motion changes of the water surface target within these segments are considered consistent with the water flow propulsion relationship. Simultaneously, the time positions where the comparison results deviate are separated from these segments, ensuring that each target change segment contains only continuous change records consistent with water flow changes. This segmentation method creates target change segment records within the continuous change sequence, allowing the motion changes of the water surface target in the monitoring section to be expressed in a segmented form, thus providing a clear segmented structure for subsequent target identification.

[0039] After dividing the target change segments, all extracted target change segments are uniformly recorded in a continuous change sequence. This ensures that each target change segment includes complete records of spatial position changes, lateral movement changes, longitudinal movement changes, and corresponding water flow changes. This recording method clearly presents the movement changes of surface targets in different time periods within the continuous change sequence, and establishes an independent recording structure for target change segments that align with water flow changes. This approach allows for clear identification of the movement of surface targets as they are propelled by the water flow within the continuous change sequence, providing a complete data foundation for subsequent floating debris identification and continuous monitoring and treatment. Furthermore, the combination of the recording table and the segment organization structure clearly expresses the relationship between target movement changes and water flow changes within the continuous change sequence, enabling a continuous presentation of the surface target movement process through temporal sequence and spatial positional relationships.

[0040] Based on the target change segment, continue to read the subsequent changes in the continuous change sequence, continuously track the location and count the number of times the identified floating target is sorted out, and complete the continuous monitoring and statistics of floating garbage; After recording the target change segment, subsequent changes in the continuous change sequence are read based on this segment. By continuously recording the spatial position changes of floating targets, the movement of floating targets within the monitoring section can form a continuous trajectory. Simultaneously, the number of floating targets entering the monitoring section is compiled, thereby achieving continuous monitoring and statistics of floating debris. The entire process revolves around a continuous change sequence, recording the positional changes of water surface targets within the monitoring section in chronological order, and continuously compiling this data to form a complete monitoring record. The specific implementation steps are as follows: Based on the extracted target change segments, subsequent changes in the continuous change sequence are read to create a continuous record of the floating target's motion within the monitoring section. The target change segment, already formed in the previous steps, records the continuous positional changes of the water surface target within the monitoring section and its motion in sync with the water flow. In this step, the last time position of this target change segment is used as the starting reading position, and the water surface change records are read sequentially within the continuous change sequence. During the reading process, the water surface image records at each time position are observed one by one, the location of the water surface target is identified within the spatial range of the monitoring section, and the target's spatial position at the current time position is connected to the last recorded position of the target change segment. This allows the spatial positional changes of the target in subsequent time positions to be continuously expressed in relation to the positional records in the original target change segment. In this way, the motion of the water surface target in the continuous change sequence can be continuously extended, thus forming a continuous record of target positional changes in the time dimension, enabling the movement trajectory of the floating target within the monitoring section to be expressed through continuous recording.

[0041] After completing the subsequent reading of the continuous change sequence, the spatial position changes of the floating target within the monitoring section are continuously tracked, ensuring a complete record of the floating target's trajectory. In practice, the spatial position of the water surface target is recorded at each time point, using the temporal order of the continuous change sequence as the main thread. The spatial position of the current time point is then correlated with the spatial position of the previous time point, forming a continuous trajectory of the floating target's spatial position changes across different time points. To ensure clear recording of the floating target's position changes within the monitoring section, each floating target is assigned a unique target number, and all spatial position changes corresponding to that target number are uniformly recorded within the continuous change sequence. This method creates a continuous trajectory record within the continuous change sequence, including the target number, target spatial position, and time position. This allows the movement of the floating target within the monitoring section to be organized according to a unified number, preventing confusion between different floating targets during recording and ensuring that the trajectory of each floating target is recorded independently within the continuous change sequence.

[0042] After continuous position tracking is completed, the frequency of appearances of floating targets within the monitoring section is compiled to create a continuous statistical record of the amount of floating debris in the monitoring section. In practice, the established target numbers are used as the basis for compilation, and the entry and exit records of each floating target within the monitoring section are compiled. When a floating target first appears within the monitoring section's spatial range in the continuous change sequence, that time position is recorded as the target's entry time, and a new target record is added to the statistical record. Subsequently, the position change trajectory of the floating target within the monitoring section continues to be recorded in the continuous change sequence. When the floating target moves outside the spatial boundary of the monitoring section, that time position is recorded as the target's exit time, and continuous position recording for that target number ceases. Through this compilation method, a correspondence is established between the entry and exit records of each floating target within the monitoring section, allowing the frequency of floating target appearances to be compiled and statistically analyzed using target numbers, and enabling the amount of floating debris in the monitoring section to be statistically recorded in chronological order.

[0043] After tracking and recording the number of times floating targets are observed, the monitoring records in the continuous change sequence are integrated to form a complete record structure for the continuous monitoring and statistics of floating debris. In practice, the records of target number, target spatial location changes, target entry time, and target departure time in the continuous change sequence are uniformly organized, allowing the movement of each floating target within the monitoring section to be fully recorded. Simultaneously, the number of occurrences of all floating targets is accumulated, enabling a continuous statistical record of the amount of floating debris within the monitoring section over a continuous time period. Through this integration, a monitoring record structure is formed in the continuous change sequence, including records of floating target spatial location changes, floating target frequency records, and continuous time change records. This allows for the continuous recording and statistics of changes in floating debris within the monitoring section in chronological order, thus completing the continuous monitoring and statistics process for floating debris and enabling the continuous recording of changes in floating debris within the monitoring section.

[0044] This invention unifies the temporal processing of water surface change information and water flow change information, enabling the changes in the position of water surface targets and the changes in water flow to be correlated within the same time series. Based on this, a flow velocity distribution representation covering the monitoring section is constructed, allowing a direct correspondence between the motion changes of water surface targets and the corresponding water flow propulsion status. This enables the identification of whether water surface targets move with the water flow through continuous changes. In this way, the motion characteristics of floating targets and water flow characteristics form a unified reference relationship, distinguishing from interference changes caused by water ripples, floating vegetation, and water surface reflections at the level of motion patterns. This improves the accuracy of water surface target identification and makes the identification process of floating debris more stable.

[0045] This invention extracts target change segments that are consistent with water flow changes and continuously tracks and records the positions and frequency of identified floating targets within a continuous change sequence. This allows for a complete record of the movement trajectory of floating targets within the monitoring section. Simultaneously, it continuously counts floating targets entering the monitoring section, thereby achieving continuous monitoring and automatic statistics of floating debris. By uniformly recording the appearance process of floating targets within the monitoring section, the changes in the quantity of floating debris can be compiled into continuous statistical results in chronological order, providing stable data for aquatic environment management and enabling the floating debris monitoring process to operate continuously under unattended conditions.

[0046] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A multimodal intelligent monitoring method for floating debris, characterized in that, Includes the following steps: Information on water surface changes and water flow changes at monitoring sections is acquired and organized in a uniform time sequence to construct a continuous change sequence reflecting the relationship between changes in the target water surface location and changes in water flow. The water flow change information corresponding to each time position is read around the continuous change sequence, and the water flow change information is sorted horizontally and vertically in combination with the hydrodynamic characteristics of the monitoring section, so as to construct a velocity distribution expression covering the monitoring section in the continuous change sequence. Based on the velocity distribution, the change process of the water surface target position in the continuous change sequence is read, and the change of the target's lateral motion and longitudinal motion are formed according to the change process of the water surface target position. Based on the changes in the lateral and longitudinal motion of the target, the corresponding positions of the velocity distribution are returned. The relationship between the changes in the lateral and longitudinal motion of the target and the water flow propulsion is compared and organized, and the target change segments that are consistent with the changes in the water flow are extracted. Based on the target change segment, the subsequent changes in the continuous change sequence are read, and the identified floating targets are continuously tracked and the number of times are sorted out to complete the continuous monitoring and statistics of floating garbage.

2. The method for multimodal intelligent monitoring of floating debris according to claim 1, characterized in that, The steps for obtaining water level and flow change information at the monitoring section and constructing a continuous change sequence are as follows: Acquire information on water surface changes and water flow changes within the monitoring section area, and perform unified time-identification processing on the information on water surface changes and water flow changes; Based on a unified time identifier, the information on water surface changes and water flow changes is organized in chronological order to form a time series record in which the information on water surface changes and water flow changes correspond one-to-one. Based on the time series records, water surface change information is read and spatial location is recorded to establish the position coordinate record of the water surface target in the monitoring section, and the water surface target position change trajectory is formed by combining the time series records; By integrating water surface change information, water surface target position change information, and water flow change information from time series records around the trajectory of water surface target position change, a continuous change sequence reflecting the relationship between water surface target position change and water flow change is formed.

3. The method for multimodal intelligent monitoring of floating debris according to claim 2, characterized in that, The process of constructing a velocity distribution representation covering the monitoring section based on a continuously changing sequence is as follows: Read the water flow change information corresponding to each time position in the continuous change sequence, and establish the correspondence between the water flow change information and the spatial position of the monitoring section around the time position. At the same time, use the radar velocity measurement position as the reference position to record the actual single-point flow velocity vector, and form a water flow change recording structure with horizontal and vertical components. The information on water flow changes in a continuous sequence is organized laterally based on the water flow change record structure, and a theoretical expression of the lateral flow velocity within the lateral range of the monitoring section is formed based on the lateral component at the radar velocity measurement location and the distance relationship between the spatial locations of the monitoring section. Based on the transverse theoretical velocity expression, the water flow change information in the continuous change sequence is longitudinally organized, and combined with the river roughness coefficient, water depth information and river water surface gradient to form a longitudinal theoretical velocity expression within the longitudinal range of the monitoring section. The theoretical flow velocity expressions for the transverse and longitudinal directions are combined and recorded in a continuous sequence to form a flow velocity distribution expression covering the spatial range of the monitoring section.

4. The method for multimodal intelligent monitoring of floating debris according to claim 3, characterized in that, The transverse and longitudinal theoretical velocity expressions are recorded in a continuous change sequence according to their corresponding time and spatial locations. They form a joint recording structure of transverse and longitudinal theoretical velocity information around the spatial range of the monitoring section, thus forming a complete velocity distribution expression in the continuous change sequence.

5. The method for multimodal intelligent monitoring of floating debris according to claim 3, characterized in that, The steps for expressing the lateral and longitudinal motion changes of the target based on velocity distribution are as follows: Read the water surface change records in the continuous change sequence, and record the coordinates of the water surface target position around the spatial coordinates of the monitoring section. At the same time, combine the flow velocity distribution to establish the correspondence between the water surface target position change records and the spatial position of the monitoring section. The system records the changes in the position of a target on the water surface by reading the spatial coordinates of the target at adjacent time points, and then generates a record of the target's lateral movement based on the changes in its lateral spatial coordinates. Based on the record of changes in the position of the water surface target, the spatial coordinates of the water surface target at adjacent time positions are read, and a record of the longitudinal motion change of the target is formed based on the changes in the longitudinal spatial coordinates; The records of lateral and longitudinal motion changes around the target are combined in a continuous sequence of changes and combined with the velocity distribution to form a structure for expressing the motion changes of the water surface target.

6. The method for multimodal intelligent monitoring of floating debris according to claim 5, characterized in that, The records of the target's lateral and longitudinal motion changes are linked and organized through a continuous sequence of changes, forming a continuous motion trajectory record around the target's position coordinates on the water surface. At the same time, the correspondence between the target's motion changes on the water surface and the spatial position of the monitoring section is established by combining the flow velocity distribution.

7. The method for multimodal intelligent monitoring of floating debris according to claim 5, characterized in that, The steps for extracting the target change segment that is consistent with the water flow change are as follows: Read the target's lateral and longitudinal motion change records in the continuous change sequence, and obtain the corresponding lateral and longitudinal water flow change records in the velocity distribution expression around the target's spatial position on the water surface, forming a spatial correspondence record between the target's motion change and the water flow change. The lateral movement changes of the target and the lateral water flow changes are compared and organized based on the spatial correspondence records. At the same time, the longitudinal movement changes of the target and the longitudinal water flow changes are compared and organized to form a motion change comparison record in a continuous change sequence. Based on the motion change comparison record, read the time sequence record in the continuous change sequence, and form the target motion change segment record around the horizontal motion change record and the vertical motion change record; The records of the target movement change section are integrated into a continuous change sequence to form a target change section that is consistent with the water flow change.

8. The method for multimodal intelligent monitoring of floating debris according to claim 7, characterized in that, The steps for continuous monitoring and statistics of floating debris are as follows: Read the water surface change records corresponding to the target change segment in the continuous change sequence, and continue to read the water surface change records in the continuous change sequence around the last time position of the target change segment to form a continuous record of the spatial position change of the floating target; Continuous position tracking is performed around the spatial position records of floating targets in a continuously changing sequence, and a target number record structure is established based on the changes in the spatial position of the floating targets, while simultaneously forming a spatial position trajectory record corresponding to the target number; Based on the target number record structure, the entry and exit records of floating targets in the monitoring section space are read, and the number of floating targets is organized and recorded around the entry and exit records. The records are integrated and recorded in a continuous sequence of changes, including target number records, spatial location trajectory records, entry records, and exit records, to form a structure for continuous monitoring and statistical recording of floating debris.

Images