CN117909905A - Hydrologic station multidimensional time sequence anomaly identification method and device considering hydrologic connection - Google Patents
Hydrologic station multidimensional time sequence anomaly identification method and device considering hydrologic connection Download PDFInfo
- Publication number
- CN117909905A CN117909905A CN202410176422.8A CN202410176422A CN117909905A CN 117909905 A CN117909905 A CN 117909905A CN 202410176422 A CN202410176422 A CN 202410176422A CN 117909905 A CN117909905 A CN 117909905A
- Authority
- CN
- China
- Prior art keywords
- hydrologic
- data
- monitoring
- image
- multidimensional
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000012544 monitoring process Methods 0.000 claims abstract description 152
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 98
- 230000002159 abnormal effect Effects 0.000 claims abstract description 16
- 239000002245 particle Substances 0.000 claims description 31
- 239000003550 marker Substances 0.000 claims description 30
- 238000013507 mapping Methods 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 5
- 230000006870 function Effects 0.000 claims description 3
- 230000005856 abnormality Effects 0.000 abstract description 9
- 238000012545 processing Methods 0.000 abstract description 5
- 230000000875 corresponding effect Effects 0.000 description 31
- 230000008901 benefit Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 239000004575 stone Substances 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The application discloses a hydrologic station multidimensional time sequence anomaly identification method and device considering hydrologic connection, and relates to hydrologic data processing technology, comprising searching markers of monitoring coverage range from corresponding video segments according to a received multidimensional hydrologic data sequence; extracting two frames of image data of video segment data contained in the received multidimensional hydrological data sequence, and calculating the water surface flow rate of the monitoring coverage area; matching the flow velocity information with the received flow velocity information to judge whether abnormality exists or not; calculating the water level of the monitoring coverage area from one frame of image data selected from video segment data in the received current multidimensional hydrological data sequence; and comparing the water level information with the received water level information to judge whether abnormality exists. The method of the application correlates the acquired hydrologic station data with the monitoring data, and realizes the abnormal identification of the hydrologic station monitoring data under the condition that any hydrologic station is provided with fewer data monitoring points.
Description
Technical Field
The application relates to the technical field of hydrologic data processing, in particular to a hydrologic station multidimensional time sequence anomaly identification method and device considering hydrologic connection.
Background
The multidimensional time series at the hydrologic station, including runoff and water level data of rivers, is the basis of hydrologic simulation and water resource analysis and evaluation of the river basin. Typically, the hydrologic data of the basin is obtained through a hydrologic yearbook issued by the relevant departments each year.
The prior art anomaly identification of data for a single hydrologic site can be accomplished by manual historical data analysis. However, for large watershed containing a plurality of branches, the processing amount of abnormal identification data of hydrologic data is large, and a plurality of data monitoring points are required to be arranged on each hydrologic station in a matched mode, so that on one hand, the labor force is consumed, and on the other hand, the monitoring cost of matched abnormal identification is increased.
Disclosure of Invention
The embodiment of the application provides a hydrologic station multidimensional time sequence abnormality identification method and device considering hydrologic connection, which are used for realizing abnormality identification of hydrologic station monitoring data under the condition that fewer data monitoring points are arranged on any hydrologic station by correlating acquired hydrologic station data with monitoring data, so that labor and equipment cost are reduced.
The embodiment of the application provides a hydrologic station multidimensional time sequence anomaly identification method considering hydrologic connection, which comprises the following steps:
Collecting a plurality of hydrologic data of the hydrologic station based on the monitoring point of the watershed hydrologic station, wherein the collected hydrologic data comprises water level information and flow rate information of the monitoring point, and any hydrologic data is provided with a time mark; and
For any hydrologic station, a monitoring camera is called to collect hydrologic video data in a corresponding flow field and a monitoring coverage range of the any hydrologic station;
extracting a required video segment from corresponding hydrological video data according to the time marks of the hydrological data acquired by each monitoring point, so that the extracted video segment comprises water surface data of a monitoring coverage area;
according to the time relation, combining the hydrologic data of the hydrologic station and the extracted video segment data to form a multidimensional hydrologic data sequence, and transmitting the multidimensional hydrologic data sequence to a central platform;
at the central platform, performing anomaly identification on the currently received multidimensional hydrological data sequence in the following manner:
searching markers of monitoring coverage areas from corresponding video segments according to the received multidimensional hydrological data sequences;
Extracting two successive frames of image data of video segment data contained in the received multidimensional hydrological data sequence, and calculating the water surface flow rate of the monitoring coverage area based on the two successive frames of image data and the marker;
Matching the calculated water surface flow velocity with flow velocity information in the received multidimensional hydrological data sequence to judge whether the acquired multidimensional hydrological data sequence is abnormal or not;
And
Selecting one frame of image data from video segment data in the received current multidimensional hydrological data sequence, and calculating the water level of a monitoring coverage area based on the marker;
And comparing the calculated water level of the monitoring coverage area with water level information in the received multidimensional hydrological data sequence to judge whether the collected multidimensional hydrological data sequence is abnormal or not.
Optionally, extracting the required video segment from the corresponding hydrologic video data according to the time stamp of the hydrologic data collected by each monitoring point includes:
and extracting video segments with specified duration from corresponding hydrological video data by taking time marks of hydrological data acquired by each monitoring point as precision references.
Optionally, searching for markers of the monitoring coverage from the corresponding video segment based on the received multi-dimensional hydrological data sequence includes:
Based on the time marks, respectively extracting a first monitoring image and a second monitoring image at two sides of the corresponding video segment;
and identifying the object in the first monitoring image, and selecting a fixed object in the identified object as a marker.
Optionally, the method further includes using the first monitoring image and the second monitoring image as extracted sequential two-frame image data, and calculating the water surface flow rate of the monitoring coverage area based on the sequential two-frame image data and the marker includes:
Extracting the boundary of the marker;
aligning the first monitoring image and the second monitoring image based on the boundary of the fixed object, and establishing a coordinate system based on the fixed object and the water flow direction;
selecting sub-images from the water surface areas of the first monitoring image and the second monitoring image based on the established coordinate system, and calculating the similarity between the selected sub-images so as to perform image pairing;
Calculating the water surface flow rate of the streamline corresponding to the two sub-images with the maximum similarity based on the image pairing result and the time difference between the first monitoring image and the second monitoring image;
And determining flow velocity information of the monitoring coverage area according to the calculated water surface flow velocity of the streamline corresponding to the two sub-images.
Optionally, selecting sub-images from the water surface areas of the first monitoring image and the second monitoring image, and calculating the similarity between the selected sub-images includes:
dividing a monitoring coverage range in the first monitoring image into a plurality of sub-images based on the established coordinate system, and taking any sub-image as a first area I;
In the second monitoring image, selecting a second area J in the water flow direction based on the position of any sub-image in the coordinate system, wherein the size of the second area J is the same as that of the first area I;
for particle I in the first region I, its similarity to particle J in the second region J is calculated to satisfy:
wherein, Is a particleAnd particlesThe degree of similarity between the two,As a first-order step function,、The thresholds for the distance deviation and the angle deviation respectively,、The distance and the link angle within the particle I and the interrogation zone I respectively,、The distance and the connection angle in the candidate particle J and the query region J are respectively N, M, and the particle number in the query region I and the query region J are respectively;
Repeatedly determining a second area J according to a preset stepping amount along the water flow direction and a preset offset amount perpendicular to the water flow direction, and calculating the particle similarity;
And for each particle of any sub-image, determining to construct a similar sub-image of any sub-image according to the similarity of each particle in the second monitoring image according to the calculated similarity.
Optionally, calculating the water surface flow rate of the monitoring coverage area based on the two frames of image data and the marker includes:
Judging whether offset occurs between any one sub-image and the corresponding similar sub-image along the direction perpendicular to the water flow or not based on the established coordinate system;
calculating a water surface flow rate based on a time difference between the first monitoring image and the second monitoring image and a flow distance in a water flow direction between any one of the sub-images and a corresponding similar sub-image in the absence of a vertical offset;
And if vertical offset occurs, determining a plurality of pixel points with the highest similarity between any sub-image and the corresponding similar sub-image, and performing curve fitting on the plurality of pixel points with the highest similarity to determine the flowing distance under the time difference between the first monitoring image and the second monitoring image, so as to calculate the water surface flow velocity.
Optionally, calculating the water level of the monitoring coverage area from the received frame of image data selected from the video segment data in the current multidimensional hydrological data sequence based on the marker includes:
Extracting one frame of image data from a prior multidimensional hydrological data sequence, and determining the water surface distance deviation between the prior extracted one frame of image data and the currently selected one frame of image data under the established coordinate system;
And matching the water surface distance deviation with a preset deviation mapping relation to determine a water level deviation value, wherein the deviation mapping relation is constructed in advance based on the position relation between the historical multidimensional hydrological data sequences at a plurality of previous moments and the marker.
The embodiment of the application also provides a hydrologic station multidimensional time series abnormality identification device considering hydrologic connection, which comprises a processor and a memory, wherein the memory is stored with a computer program, and the computer program realizes the steps of the hydrologic station multidimensional time series abnormality identification method considering hydrologic connection when being executed by the processor.
According to the embodiment of the application, the acquired hydrologic station data are associated with the monitoring data, so that the abnormal identification of the hydrologic station monitoring data is completed under the condition that fewer data monitoring points are set on any hydrologic station, and the labor and equipment cost is reduced.
The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to designate like parts throughout the figures. In the drawings:
fig. 1 is a basic flow diagram of a multi-dimensional time series anomaly identification method for a hydrologic station according to the present embodiment.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
The method of the embodiment of the application is used for establishing the association and comparison between the hydrologic data and the monitoring image data in the corresponding monitoring coverage range so as to realize the abnormal identification of the data of the monitoring points. The embodiment of the application provides a hydrologic station multidimensional time sequence anomaly identification method considering hydrologic connection, which is shown in fig. 1 and comprises the following steps:
In step S101, a plurality of hydrologic data of the hydrologic station is acquired based on a monitoring point of the watershed hydrologic station, wherein the acquired hydrologic data includes water level information and flow rate information of the monitoring point, and any hydrologic data has a time stamp. In a specific embodiment, any basin can comprise a plurality of hydrologic stations, and each hydrologic station can be provided with a corresponding sensor at each monitoring point, so that water level information and flow rate information of the monitoring points are collected through the sensors.
In step S102, for any hydrologic station, a monitoring camera is invoked to collect hydrologic video data in a corresponding flow field and a monitoring coverage range of the any hydrologic station. In a specific example, the camera device may be separately arranged, or any hydrologic station may be directly collected by using a monitoring camera of the hydrologic station, so that hydrologic video data in a monitoring coverage area may be a long-segment video or may be a part of video data intercepted with reference to a set time interval.
In step S103, according to the time stamp of the acquired hydrological data of each monitoring point, a required video segment is extracted from the corresponding hydrological video data, so that the extracted video segment includes the water surface data of the monitoring coverage area. In particular embodiments, the time stamp of the hydrographic data collected by the monitoring point may be a time stamp that is self-contained by the system, by which the desired video segment is extracted from the hydrographic video data, and in some embodiments the extracted video segment may be a short duration video segment, such as extracting only 1s, 5s of video data around the time stamp.
In step S104, the hydrologic data of the hydrologic station and the extracted video segment data are combined according to the time relationship to form a multidimensional hydrologic data sequence, and the multidimensional hydrologic data sequence is transmitted to the central platform. In specific implementation, the multidimensional hydrologic data sequences formed by combining the corresponding moments can be sent to a central platform for processing according to the specified time intervals.
At the central platform, performing anomaly identification on the currently received multidimensional hydrological data sequence in the following manner:
In step S105, markers of the monitoring coverage are searched for from the corresponding video segments according to the received multi-dimensional hydrological data sequence.
In step S106, two successive frames of image data of video segment data included in the received multidimensional hydrological data sequence are extracted, and based on the two successive frames of image data and the marker, a water surface flow rate of the monitoring coverage area is calculated. In a specific example, the water surface flow rate of the monitoring coverage area can be calculated by combining the front frame image and the rear frame image for matching, so that the water surface flow rate at the monitoring point can be further determined.
In step S107, the calculated water surface flow rate is matched with the flow rate information in the received multi-dimensional hydrological data sequence to determine whether the acquired multi-dimensional hydrological data sequence is abnormal.
In step S108, a frame of image data is selected from the video segment data in the received current multidimensional hydrological data sequence, and the water level of the monitoring coverage is calculated based on the marker. In a specific example, the proportional relationship of the hydrologic rise or fall can be determined based on the marker and the historical water level data without abnormality, so that the water level of the monitoring coverage is calculated based on the image data.
In step S109, the calculated water level of the monitoring coverage area is compared with the water level information in the received multidimensional hydrographic data sequence to determine whether the collected multidimensional hydrographic data sequence is abnormal.
The method of the embodiment of the application considers the relevance between the image data collected by monitoring and the actually measured hydrologic data, calculates a reference water level and flow rate range based on the image data, and carries out abnormal recognition on the water level and flow rate collected by the sensor through the calculated reference water level and flow rate range. The acquired hydrologic station data and the monitoring data are correlated by the method, so that the abnormal identification of the hydrologic station monitoring data can be completed under the condition that fewer data monitoring points are set in any hydrologic station, the accuracy of the hydrologic station sensing data is improved, and the labor and equipment cost is reduced.
In some embodiments, extracting the desired video segment from the corresponding hydrographic video data based on the time stamps of the hydrographic data collected at each monitoring point comprises:
and extracting video segments with specified duration from corresponding hydrological video data by taking time marks of hydrological data acquired by each monitoring point as precision references.
In some embodiments, searching for markers of monitoring coverage from corresponding video segments based on the received multi-dimensional sequence of hydrographic data comprises:
And respectively extracting a first monitoring image and a second monitoring image on two sides of the corresponding video segment based on the time mark. The first and second monitoring images are extracted for use as a later calculation of flow rate and water level.
And identifying the object in the first monitoring image, and selecting a fixed object in the identified object as a marker. In some specific examples, the marker may be a fixed object such as a stone pillar or the like in the coverage area, or a manually set marker such as a water line signboard or the like, and the marker may be fixed after first selection and used in subsequent processing and recognition.
In some embodiments, the method further comprises using the first and second monitoring images as extracted sequential two frames of image data, and calculating a water surface flow rate of the monitoring coverage area based on the sequential two frames of image data and the marker comprises:
extracting the boundary of the marker. In some examples, the boundary may be selected by pixel identification.
And aligning the first monitoring image and the second monitoring image based on the boundary of the fixed object, and establishing a coordinate system based on the fixed object and the water flow direction. In some examples, a fixed point may be selected on the marker based on the selected marker to establish a coordinate system, thereby completing subsequent image matching and computation processes based on the coordinate system.
Selecting sub-images from the water surface areas of the first monitoring image and the second monitoring image based on the established coordinate system, and calculating the similarity between the selected sub-images so as to perform image pairing. By pairing, for example, the moving amount of the sub-image in the first monitoring image in the front and the second monitoring image in the back can be determined, so that the flow velocity on the water flow line corresponding to the sub-image can be calculated.
And calculating the water surface flow rate of the streamline corresponding to the two sub-images with the maximum similarity based on the image pairing result and the time difference between the first monitoring image and the second monitoring image.
And determining flow velocity information of the monitoring coverage area according to the calculated water surface flow velocity of the streamline corresponding to the two sub-images.
In some embodiments, selecting sub-images from the water surface areas of the first and second monitoring images, and calculating the similarity between the selected sub-images includes:
dividing a monitoring coverage range in the first monitoring image into a plurality of sub-images based on the established coordinate system, and taking any sub-image as a first area I;
In the second monitoring image, selecting a second area J in the water flow direction based on the position of any sub-image in the coordinate system, wherein the size of the second area J is the same as that of the first area I;
for particle I in the first region I, its similarity to particle J in the second region J is calculated to satisfy:
wherein, Is a particleAnd particlesThe degree of similarity between the two,As a first-order step function,、The thresholds for the distance deviation and the angle deviation respectively,、The distance and the link angle within the particle I and the interrogation zone I respectively,、The distance and the connection angle in the candidate particle J and the query region J are respectively N, M, and the particle number in the query region I and the query region J are respectively;
the second region J is repeatedly determined in the water flow direction by a preset step amount and in the water flow direction perpendicular to the water flow direction by a preset offset amount, and the particle similarity is calculated. The adjustment of the stepping amount and the offset amount can be realized based on the established coordinate system, and the second area is repeatedly selected.
And for each particle of any sub-image, determining to construct a similar sub-image of any sub-image according to the similarity of each particle in the second monitoring image according to the calculated similarity. In a specific embodiment, according to the calculation result, a position where the similarity of each particle of any sub-image is maximum may be determined in the second monitoring image, so as to determine a similar region of any sub-image in the second monitoring image.
In some embodiments, calculating the water surface flow rate of the monitoring coverage area based on the two successive frames of image data and the marker comprises:
Based on the established coordinate system, whether the deviation occurs between any sub-image and the corresponding similar sub-image along the direction perpendicular to the water flow is judged. In the embodiment of the application, calculation is respectively performed for the condition that the water flow in the monitoring coverage area is a curve and a straight line, namely, whether the first area and the second area which are paired are offset based on the axis of the coordinate system is judged based on the established coordinate system.
In the absence of a vertical offset, a water surface flow rate is calculated based on a time difference between the first monitor image and the second monitor image, and a flow distance in a water flow direction between any one of the sub-images and the corresponding similar sub-image.
And if vertical offset occurs, determining a plurality of pixel points with the highest similarity between any sub-image and the corresponding similar sub-image, and performing curve fitting on the plurality of pixel points with the highest similarity to determine the flowing distance under the time difference between the first monitoring image and the second monitoring image, so as to calculate the water surface flow velocity.
After the flow rate for monitoring the coverage water surface range is calculated, the flow rate is corresponding to each monitoring point through the proportional relation, so that the flow rate at the monitoring point is calculated. In a specific example, a floating range may be configured for the calculated flow rate, so that, as a reference interval of the flow rate, if the flow rate at the monitoring point exceeds the flow rate interval, the flow rate data in the corresponding multidimensional hydrological data sequence is considered to be misaligned and abnormal.
In some embodiments, calculating the level of monitoring coverage from a selected frame of image data from video segment data in a received current sequence of multi-dimensional hydrologic data and based on the marker comprises:
Extracting one frame of image data from a prior multidimensional hydrological data sequence, and determining the water surface distance deviation between the prior extracted one frame of image data and the currently selected one frame of image data under the established coordinate system;
And matching the water surface distance deviation with a preset deviation mapping relation to determine a water level deviation value, wherein the deviation mapping relation is constructed in advance based on the position relation between the historical multidimensional hydrological data sequences at a plurality of previous moments and the marker. In some examples, the deviation mapping relationship may be established based on the anomaly-free monitoring image and water level data for the monitoring points. Similarly, a deviation range can be determined for the matched water level data, and when the water level data acquired by the water level sensor exceeds the deviation range, the corresponding multidimensional hydrological data sequence is considered to be abnormal.
According to the method provided by the embodiment of the application, the monitoring image is processed, the related hydrologic data is calculated to carry out abnormal recognition on the hydrologic data collected by the monitoring points, so that on one hand, the accuracy of the data collected by the hydrologic station monitoring points can be improved, and on the other hand, under the condition of collecting and monitoring the hydrologic data in a large range, the accuracy of the monitoring data can be further checked through the scheme of the application, the workload of manual judgment and screening is reduced, and the efficiency of collecting and recognizing the hydrologic data is improved.
The embodiment of the application also provides a hydrologic station multidimensional time series abnormality identification device considering hydrologic connection, which comprises a processor and a memory, wherein the memory is stored with a computer program, and the computer program realizes the steps of the hydrologic station multidimensional time series abnormality identification method considering hydrologic connection when being executed by the processor.
Furthermore, although exemplary embodiments have been described herein, the scope thereof includes any and all embodiments having equivalent elements, modifications, omissions, combinations (e.g., of the various embodiments across schemes), adaptations or alterations based on the present disclosure. And are not limited to the examples described in this specification or during the practice of the application, which examples are to be construed as non-exclusive.
The above description is intended to be illustrative and not restrictive. For example, the above-described examples (or one or more aspects thereof) may be used in combination with each other. For example, other embodiments may be used by those of ordinary skill in the art upon reading the above description.
The above embodiments are merely exemplary embodiments of the present disclosure, and those skilled in the art may make various modifications or equivalents to the present invention within the spirit and scope of the present disclosure, and such modifications or equivalents should also be construed as falling within the scope of the present invention.
Claims (8)
1. A hydrologic station multidimensional time series anomaly identification method considering hydrologic connection is characterized by comprising the following steps:
Collecting a plurality of hydrologic data of the hydrologic station based on the monitoring point of the watershed hydrologic station, wherein the collected hydrologic data comprises water level information and flow rate information of the monitoring point, and any hydrologic data is provided with a time mark; and
For any hydrologic station, a monitoring camera is called to collect hydrologic video data in a corresponding flow field and a monitoring coverage range of the any hydrologic station;
extracting a required video segment from corresponding hydrological video data according to the time marks of the hydrological data acquired by each monitoring point, so that the extracted video segment comprises water surface data of a monitoring coverage area;
according to the time relation, combining the hydrologic data of the hydrologic station and the extracted video segment data to form a multidimensional hydrologic data sequence, and transmitting the multidimensional hydrologic data sequence to a central platform;
at the central platform, performing anomaly identification on the currently received multidimensional hydrological data sequence in the following manner:
searching markers of monitoring coverage areas from corresponding video segments according to the received multidimensional hydrological data sequences;
Extracting two successive frames of image data of video segment data contained in the received multidimensional hydrological data sequence, and calculating the water surface flow rate of the monitoring coverage area based on the two successive frames of image data and the marker;
Matching the calculated water surface flow velocity with flow velocity information in the received multidimensional hydrological data sequence to judge whether the acquired multidimensional hydrological data sequence is abnormal or not;
And
Selecting one frame of image data from video segment data in the received current multidimensional hydrological data sequence, and calculating the water level of a monitoring coverage area based on the marker;
And comparing the calculated water level of the monitoring coverage area with water level information in the received multidimensional hydrological data sequence to judge whether the collected multidimensional hydrological data sequence is abnormal or not.
2. The method for identifying a multidimensional time series anomaly of a hydrologic station taking into account hydrologic connections according to claim 1, wherein extracting a required video segment from corresponding hydrologic video data according to the time stamp of the hydrologic data collected by each monitoring point comprises:
and extracting video segments with specified duration from corresponding hydrological video data by taking time marks of hydrological data acquired by each monitoring point as precision references.
3. The method for identifying a hydrologic station multidimensional time series anomaly in view of hydrologic connections of claim 2, wherein searching for markers of monitoring coverage from corresponding video segments based on the received multidimensional hydrologic data sequence includes:
Based on the time marks, respectively extracting a first monitoring image and a second monitoring image at two sides of the corresponding video segment;
and identifying the object in the first monitoring image, and selecting a fixed object in the identified object as a marker.
4. The method for identifying a multi-dimensional time series anomaly of a hydrologic station taking hydrologic connections into account of claim 3, further comprising using the first and second monitored images as extracted sequential two frames of image data, and calculating a water surface flow rate of a monitored coverage based on the sequential two frames of image data and the marker, comprising:
Extracting the boundary of the marker;
aligning the first monitoring image and the second monitoring image based on the boundary of the fixed object, and establishing a coordinate system based on the fixed object and the water flow direction;
selecting sub-images from the water surface areas of the first monitoring image and the second monitoring image based on the established coordinate system, and calculating the similarity between the selected sub-images so as to perform image pairing;
Calculating the water surface flow rate of the streamline corresponding to the two sub-images with the maximum similarity based on the image pairing result and the time difference between the first monitoring image and the second monitoring image;
And determining flow velocity information of the monitoring coverage area according to the calculated water surface flow velocity of the streamline corresponding to the two sub-images.
5. The method for identifying a multi-dimensional time series anomaly of a hydrologic station taking into account hydrologic connections according to claim 4, wherein selecting sub-images from the water surface areas of the first and second monitoring images and calculating the similarity between the selected sub-images comprises:
dividing a monitoring coverage range in the first monitoring image into a plurality of sub-images based on the established coordinate system, and taking any sub-image as a first area I;
In the second monitoring image, selecting a second area J in the water flow direction based on the position of any sub-image in the coordinate system, wherein the size of the second area J is the same as that of the first area I;
for particle I in the first region I, its similarity to particle J in the second region J is calculated to satisfy:
wherein/> Is particle/>And particle/>Similarity between/>As a first-order step function,/>、/>Threshold values for distance deviation and angle deviation, respectively,/>、/>Distance and line angle within particle I and query region I, respectively,/>、/>The distance and the connection angle in the candidate particle J and the query region J are respectively N, M, and the particle number in the query region I and the query region J are respectively;
Repeatedly determining a second area J according to a preset stepping amount along the water flow direction and a preset offset amount perpendicular to the water flow direction, and calculating the particle similarity;
And for each particle of any sub-image, determining to construct a similar sub-image of any sub-image according to the similarity of each particle in the second monitoring image according to the calculated similarity.
6. The method for identifying a multi-dimensional time series anomaly of a hydrologic station taking into account hydrologic connections according to claim 5, wherein calculating a water surface flow rate of a monitoring coverage area based on the two successive frames of image data and the marker comprises:
Judging whether offset occurs between any one sub-image and the corresponding similar sub-image along the direction perpendicular to the water flow or not based on the established coordinate system;
calculating a water surface flow rate based on a time difference between the first monitoring image and the second monitoring image and a flow distance in a water flow direction between any one of the sub-images and a corresponding similar sub-image in the absence of a vertical offset;
And if vertical offset occurs, determining a plurality of pixel points with the highest similarity between any sub-image and the corresponding similar sub-image, and performing curve fitting on the plurality of pixel points with the highest similarity to determine the flowing distance under the time difference between the first monitoring image and the second monitoring image, so as to calculate the water surface flow velocity.
7. The method for identifying a multi-dimensional time series anomaly of a hydrographic station taking into account hydrologic connections according to claim 4, wherein calculating a water level of a monitoring coverage area from a frame of image data selected from video segment data in a received current multi-dimensional hydrographic data sequence based on the marker comprises:
Extracting one frame of image data from a prior multidimensional hydrological data sequence, and determining the water surface distance deviation between the prior extracted one frame of image data and the currently selected one frame of image data under the established coordinate system;
And matching the water surface distance deviation with a preset deviation mapping relation to determine a water level deviation value, wherein the deviation mapping relation is constructed in advance based on the position relation between the historical multidimensional hydrological data sequences at a plurality of previous moments and the marker.
8. A hydrologic station multidimensional time series anomaly identification device taking into account hydrologic connections, comprising a processor and a memory, the memory having stored thereon a computer program which, when executed by the processor, performs the steps of the hydrologic station multidimensional time series anomaly identification method taking into account hydrologic connections as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410176422.8A CN117909905B (en) | 2024-02-08 | 2024-02-08 | Hydrologic station multidimensional time sequence anomaly identification method and device considering hydrologic connection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410176422.8A CN117909905B (en) | 2024-02-08 | 2024-02-08 | Hydrologic station multidimensional time sequence anomaly identification method and device considering hydrologic connection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117909905A true CN117909905A (en) | 2024-04-19 |
| CN117909905B CN117909905B (en) | 2024-06-21 |
Family
ID=90689506
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410176422.8A Active CN117909905B (en) | 2024-02-08 | 2024-02-08 | Hydrologic station multidimensional time sequence anomaly identification method and device considering hydrologic connection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117909905B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115100602A (en) * | 2022-06-30 | 2022-09-23 | 浙江金融职业学院 | Static water area monitoring method |
| CN115789527A (en) * | 2022-11-01 | 2023-03-14 | 江苏鸿利智能科技股份有限公司 | Analysis system and method based on water environment informatization treatment |
| WO2023174922A1 (en) * | 2022-03-16 | 2023-09-21 | Vortex.Io | Method and station for hydrological surveillance of a watercourse |
-
2024
- 2024-02-08 CN CN202410176422.8A patent/CN117909905B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023174922A1 (en) * | 2022-03-16 | 2023-09-21 | Vortex.Io | Method and station for hydrological surveillance of a watercourse |
| CN115100602A (en) * | 2022-06-30 | 2022-09-23 | 浙江金融职业学院 | Static water area monitoring method |
| CN115789527A (en) * | 2022-11-01 | 2023-03-14 | 江苏鸿利智能科技股份有限公司 | Analysis system and method based on water environment informatization treatment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117909905B (en) | 2024-06-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112906694B (en) | Reading correction system and method for transformer substation inclined pointer instrument image | |
| CN112102369B (en) | Autonomous inspection method, device, equipment and storage medium for water surface floating target | |
| CN108918526A (en) | A kind of chips defect detection method of flexibility IC package base plate line | |
| CN102768022B (en) | Tunnel surrounding rock deformation detection method adopting digital camera technique | |
| CN107729853B (en) | Automatic identification method suitable for narrow-scale pointer instrument of transformer substation | |
| CN111275960A (en) | Traffic road condition analysis method, system and camera | |
| CN111950330A (en) | Pointer instrument indicating number detection method based on target detection | |
| EP2927871A1 (en) | Method and device for calculating number of pedestrians and crowd movement directions | |
| CN109598947B (en) | Vehicle identification method and system | |
| CN105675626A (en) | Character defect detecting method of tire mold | |
| CN101937509A (en) | Automatic target identifying method based on image processing technology | |
| CN105718964B (en) | A kind of visible detection method of power transmission line damper | |
| CN107016353A (en) | A kind of method and system of variable resolution target detection and identification integration | |
| CN114639064B (en) | Water level identification method and device | |
| CN108592903B (en) | Vehicle geomagnetic matching positioning method based on road network geomagnetic reference library | |
| CN105307116A (en) | Distributed vision positioning system and method based on mobile robot | |
| CN112013921A (en) | Method, device and system for acquiring water level information based on water level gauge measurement image | |
| CN114486764A (en) | Agricultural non-point source pollution monitoring system based on full-spectrum water quality analyzer | |
| CN113239864A (en) | Route planning method of unmanned aerial vehicle suitable for agricultural investigation | |
| CN110674698B (en) | Remote sensing image rotating ship target detection method based on intensive subregion cutting | |
| CN117909905B (en) | Hydrologic station multidimensional time sequence anomaly identification method and device considering hydrologic connection | |
| CN114845260A (en) | Hydrology monitoring data acquisition system based on thing networking | |
| CN116469057A (en) | Environment monitoring system based on big data | |
| CN115934979A (en) | Soil erosion information management method and system based on multiple acquisition ends | |
| EP3913599A1 (en) | Sign position identification system and program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |