A method and system for monitoring ice based on video monitoring

Abstract

The present application discloses an ice monitoring method and system based on video monitoring, which involves river channel monitoring and image processing technology, including: pre-dividing the basin into multiple sub-basin segments according to the basin relationship based on the geographical elevation of the basin; obtaining video data of the basin in the partial area of ​​the sub-basin segment where ice may appear according to weather data; identifying the river bank boundary and river surface width in the intercepted image data; extracting a pixel range of a preset proportion of the river surface width on both sides of the river bank boundary as a possible ice starting area according to the identified river bank boundary; judging whether there is ice growth on the river bank boundary; and continuously monitoring the basin when it is determined that there is ice growth, and monitoring the growth and flow of ice based on the flow area. The present application accurately monitors and analyzes the generation and flow of ice based on the video monitoring data of the basin, so as to effectively respond to ice disasters in real time.

Description

Technical Field

[0001] The present application relates to the field of river channel monitoring and image processing technology, and in particular to an ice monitoring method and system based on video monitoring. Background Art

[0002] Ice is the solid formed by water condensing at 0℃ or below 0℃, and ice accumulation is icy. Ice can be divided into three categories: hydrous ice, sedimentary ice, and glacial ice. Hydrous ice is formed by the direct freezing of water, including river ice, lake ice, and sea ice.

[0003] When the air temperature continues to drop and the water temperature reaches or slightly below 0℃, the river begins to freeze on the shore and in the water. River freezing is freezing in moving water, which is different from lake freezing. Lake freezing is limited to the surface of the water body, and the deep water body still maintains a temperature above 0℃. Due to the turbulent mixing of water flow in rivers, the heat loss of water bodies almost occurs simultaneously in the entire water body. Therefore, rivers not only form thin ice and shore ice on the surface of the water, but also form underwater ice in the water and on the riverbed.

[0004] For example, the Yellow River Basin in my country will enter the ice flood period in winter. Ice will flow along the entire Yellow River, and the length of the ice-flowing river section is about 400 kilometers. The ice phenomenon may affect river traffic, water resource utilization, etc. Therefore, real-time and accurate monitoring and analysis of the Yellow River is an important task of ice monitoring in the Yellow River Basin in order to effectively respond to potential ice disasters. Summary of the invention

[0005] The embodiments of the present application provide an ice monitoring method and system based on video surveillance, which are used to accurately monitor and analyze the generation and flow of ice based on video surveillance data of a watershed, so as to effectively respond to ice disasters in real time.

[0006] The present application embodiment provides an ice monitoring method based on video monitoring, including:

[0007] Based on the geographical elevation of the basin, the basin is divided into multiple sub-basin sections in sequence according to the basin relationship, so that the geographical elevation of each sub-basin section is within the set elevation difference range, and the sub-basin sections in sequence have similar temperature difference distribution;

[0008] According to the weather data, for a partial area in the sub-basin section where ice may appear, video data of the watershed in the partial area is obtained;

[0009] For the acquired video data, according to the time relationship, the image data at the designated time is intercepted, and the river bank boundary and the river width in the intercepted image data are identified;

[0010] According to the identified river bank boundary, a pixel range of a preset proportion of the river surface width on both sides of the river bank boundary is extracted as a possible ice starting area, and other areas in the river surface are used as flow areas;

[0011] For the possible ice initiation area, a preset river bank image without ice is used for matching to determine whether there is ice growth at the river bank boundary;

[0012] If ice growth is confirmed, the basin will be monitored continuously, and the growth and flow of ice will be monitored based on the flow area.

[0013] Optionally, identifying a riverbank boundary in the captured image data includes:

[0014] Identifying a first pixel range covered by a water area in the image data, and adding a disturbed color interval with set transparency to the first pixel range to obtain a second pixel range, wherein the disturbed pixel range is determined based on the color interval of the shadow and riverbed area;

[0015] Outside the boundary of the first pixel range covered by the water area, a continuous boundary is determined based on the second pixel range to use the continuous boundary as the river bank boundary.

[0016] Optionally, identifying the width of a river in the captured image data includes:

[0017] Select any reference point on the river bank boundary, and determine the opposite bank reference point with the closest pixel distance on the opposite bank boundary based on the reference point; and

[0018] Select multiple reference points based on any of the reference points according to the set pixel spacing, and determine the opposite bank reference point with the closest pixel distance corresponding to the side bank boundary;

[0019] Fitting the midpoints of the connecting lines between the reference point and the opposite bank reference point, and between each reference point and the corresponding opposite bank reference point;

[0020] The similarity between the fitted curve and the river bank boundary is calculated, and when the similarity meets a preset threshold, the river width is determined based on the reference point and the reference point on the opposite bank, and the fitted curve is used as a river streamline.

[0021] Optionally, before matching the possible ice initiation area using a preset riverbank image without ice, the method further includes:

[0022] Extracting the river bank boundary of the preset river bank image in the absence of ice;

[0023] matching the identified riverbank boundary with the riverbank boundary of the intercepted image data;

[0024] Select a reference position from the matched partial area;

[0025] The reference position of the captured image data is color balanced based on the reference position of the river bank image without ice, so that the color deviation of the reference position is within a set range after color balancing.

[0026] Optionally, matching the possible ice initiation area with a preset riverbank image without ice includes:

[0027] For the matched partial area, the area outside the preset pixel distance from the river bank boundary is eliminated;

[0028] Segment the removed area;

[0029] Determine a third pixel range contained in a sub-region of the intercepted image data and a color deviation between the segmented sub-regions;

[0030] When the third pixel range includes preset ice edge pixels and the color deviation between corresponding sub-areas exceeds the deviation threshold, it is determined that ice is growing on the river bank boundary.

[0031] Optionally, it also includes continuous monitoring of river bank boundaries to determine if there is ice growth; and,

[0032] intercepting image data from the subsequently acquired video data;

[0033] In the river surface area in the subsequent image data, pixel values ​​in the first pixel range are removed to determine whether ice appears in the river surface area;

[0034] When it is determined that ice is present in the river surface area, the flow velocity of the ice is calculated based on the river flow line and a comparison image with subsequent image data at a specified time interval.

[0035] Optionally, calculating the flow velocity of ice based on the river flow line and a comparison image with subsequent image data at a specified time interval includes:

[0036] In the subsequent image data, determining the position of the ice closest to the stream line of the river;

[0037] Based on the river flow line, searching for the determined ice position in the comparison image at a specified time interval to find a second position in the flow direction;

[0038] According to the second position and the ice position, fitting is performed with reference to the river streamline to obtain a flow distance;

[0039] The flow velocity of the ice is determined according to the flow distance and the time interval.

[0040] Optionally, when it is determined that ice slush appears in the river surface area, and ice slush is included, it also includes:

[0041] In the river surface area in the subsequent image data, the coverage area of ​​the ice is determined according to the area after the pixel values ​​of the first pixel range are removed;

[0042] Based on the determined coverage area, the ice coverage is estimated according to the empirical model of ice.

[0043] An embodiment of the present application further proposes an ice monitoring system based on video monitoring, including a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the steps of the ice monitoring method based on video monitoring as described above are implemented.

[0044] The method of the embodiment of the present application accurately monitors and analyzes the generation and flow of ice and icicles based on the video surveillance data of the watershed, so as to effectively respond to ice and icicle disasters in real time.

[0045] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0047] Figure 1 The following is a schematic diagram of the basic process of the ice monitoring method based on video monitoring in this embodiment. DETAILED DESCRIPTION

[0048] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0049] The present application embodiment proposes an ice monitoring method based on video monitoring, such as Figure 1 As shown, the following steps are included:

[0050] In step S101, the basin is divided into a plurality of sub-basin sections in sequence according to the basin relationship based on the geographical elevation of the basin in advance, so that the geographical elevation of each sub-basin section is within the set elevation difference range, and the sequential sub-basin sections have similar temperature difference distribution. For example, if the elevation interval of a river section in the basin is relatively flat, the entire section can be converted into the same sub-basin section, and for elevation differences higher than the set value, such as 60m, they are divided into different sub-basin sections, so that the elevation distribution in the same sub-basin section is relatively uniform, and the corresponding temperature difference in the same sub-basin section is within the set range, such as a deviation of 0.4°C.

[0051] In some examples, the temperature has a significant impact on the conditions for the occurrence of ice. In step S102 of the present application, based on weather data, video data of the partial area of ​​the sub-basin segment where ice may appear is obtained, while data acquisition or processing is not performed for other sub-basin segments, thereby reducing the amount of data calculations.

[0052] In step S103, the image data at the specified time is intercepted for the acquired video data according to the time relationship, and the river bank boundary and the river width in the intercepted image data are identified. For example, the image data at the specified time t is intercepted, and the specified time t can be set according to actual needs. In subsequent examples, the flow rate of ice can be determined based on the t± time interval.

[0053] In step S104, based on the identified river bank boundary, a pixel range of a preset ratio of the river width on both sides of the river bank boundary is extracted as a possible ice starting area, and other areas in the river surface are used as flow areas. In subsequent examples, the flow rate of ice can be determined based on the river streamlines in the flow area, and a pixel range of a preset ratio is extracted or intercepted as a possible ice starting area, thereby discarding other river bank areas to reduce the amount of image processing operations.

[0054] In step S105, the possible ice initiation area is matched with a preset river bank image without ice to determine whether there is ice growing on the river bank boundary.

[0055] In step S106, when it is determined that ice is growing, the watershed is continuously monitored, and the growth and flow of ice are monitored based on the flow area.

[0056] The method of the embodiment of the present application accurately monitors and analyzes the generation and flow of ice and icicles based on the video surveillance data of the watershed, so as to effectively respond to ice and icicle disasters in real time.

[0057] In some embodiments, identifying a riverbank boundary in the captured image data includes:

[0058] A first pixel range covered by water in the image data is identified, and a disturbed color interval with a set transparency is added to the first pixel range to obtain a second pixel range, wherein the disturbed pixel range is determined based on the color interval of the shadow and the riverbed area. In some examples, for example, a river bank reflection and a partial color of the riverbed are added, and a certain transparency, such as 50%-70%, is set to simulate the river bank reflection and the transparent part of the riverbed, so as to better identify the river bank boundary.

[0059] Outside the boundary of the first pixel range covered by the water area, a continuous boundary is determined based on the second pixel range to use the continuous boundary as the river bank boundary.

[0060] In some embodiments, identifying the width of a river in the captured image data includes:

[0061] Select any reference point on the river bank boundary, and determine the opposite bank reference point with the closest pixel distance on the opposite bank boundary based on the reference point; and

[0062] Based on any of the reference points, multiple reference points are selected according to the set pixel spacing, and the opposite bank reference point with the closest pixel distance is determined at the side bank boundary. That is, based on the reference point, a line segment is drawn to the opposite bank, and the reference point of the river bank corresponding to the shortest line segment is selected as the opposite bank reference point.

[0063] Fit the midpoints of the connecting lines between the reference point and the opposite bank reference point, and between each reference point and the corresponding opposite bank reference point. After repeatedly selecting multiple reference points near the reference point, the center line of the river channel can be determined based on the midpoints of the shortest line segments, and the center line is most likely the main river channel streamline.

[0064] The similarity between the fitted curve and the river bank boundary is calculated, and when the similarity meets a preset threshold, the width of the river surface is determined based on the reference point and the corresponding reference point on the opposite bank, and the fitted curve is used as the river streamline. In the embodiment of the present application, by calculating the similarity between the fitted curve and the river bank boundary, the fitting effect of the river streamline can be further improved, the shortest line segment deviation in some river flow sections can be avoided, and the pixel distance between the reference point and the corresponding reference point on the opposite bank is further used as the river surface width.

[0065] In some embodiments, before matching the possible ice initiation area using a preset riverbank image without ice, the method further includes:

[0066] The river bank boundary of the preset river bank image in the absence of ice and icicles may be extracted by the aforementioned method, and the extraction may be completed based on historical image data and applied in subsequent monitoring.

[0067] The identified riverbank boundary is matched with the riverbank boundary of the captured image data.

[0068] A reference position is selected from the matching partial area, that is, in the embodiment of the present application, a color reference position is selected only for the matching partial area. In a specific example, a partial area with a relatively fixed color value can be selected as the reference position.

[0069] The reference position of the intercepted image data is color balanced based on the reference position of the riverbank image without ice, so that the color deviation of the reference position is within a set range after color balancing. For example, if the color deviation between the reference positions is greater than a set deviation threshold, the overall image brightness contrast of the intercepted image data can be adjusted based on the reference position, so that the color deviation is controlled within the set range.

[0070] In some embodiments, matching the possible ice initiation area using a preset riverbank image without ice includes:

[0071] For the matched partial area, the area outside the preset pixel distance from the river bank boundary is eliminated. For example, the area of ​​the pixel range with the preset ratio to the river width is combined, the corresponding distance is used as the preset pixel distance, and the area outside the pixel distance is eliminated, and the remaining area is the possible ice starting area.

[0072] The removed areas are segmented. In some examples, the initial generation of ice is not uniform along the river bank, and there is a certain degree of randomness. In an embodiment of the present application, the possible ice starting area is segmented, and the segmented sub-images are locally compared to determine whether ice is generated.

[0073] Determine the third pixel range contained in the sub-region of the intercepted image data and the color deviation between the segmented sub-regions. Further, after color balancing, theoretically, for the sub-region without ice ridges, the difference from the third pixel range of the preset riverbank image without ice ridges is not too large, while the sub-region with ice ridges has pixels with a large deviation from the third pixel range.

[0074] Based on this, when the third pixel range contains preset ice edge pixels and the color deviation between corresponding sub-areas exceeds the deviation threshold, it is determined that ice is growing on the river bank boundary.

[0075] In some embodiments, it further includes continuously monitoring the river bank boundary after determining whether there is ice growth; and

[0076] intercepting image data from the subsequently acquired video data;

[0077] In the river surface area in the subsequent image data, the pixel values ​​of the first pixel range are removed to determine whether icicles appear in the river surface area. That is, after determining that icicles grow on the river bank boundary, continuous monitoring is performed, and continuous monitoring can also be performed during the thawing period after the river surface is frozen. For the acquired video data, the image information is extracted, and the pixel values ​​of the first pixel range of the water surface are removed from the river surface area. The remaining part of the river surface is most likely an ice ridge area, that is, icicles appear.

[0078] When it is determined that ice is present in the river surface area, the flow velocity of the ice is calculated based on the river flow line and a comparison image with subsequent image data at a specified time interval.

[0079] In some embodiments, calculating the flow velocity of ice based on the river flow line and the comparison image with the subsequent image data at a specified time interval includes:

[0080] In the subsequent image data, the position of the ice floes closest to the river flow line is determined. For example, when the river flow line includes the position of the ice floes, the position is directly determined or selected.

[0081] Based on the river flow line, the determined ice position is searched in the comparison image at a specified time interval to find a second position in the flow direction.

[0082] According to the second position and the ice position, fitting is performed with reference to the river streamline to obtain the flow distance. In some examples, by finding the position of the ice after a certain interval, the pixel distance moved by the ice in the interval time can be determined.

[0083] The flow velocity of the ice is determined according to the flow distance and the time interval, and can be further mapped to the display distance in combination with the pixel distance, so as to calculate the flow velocity of the ice, thereby realizing the monitoring of the flow of ice.

[0084] In some embodiments, when it is determined that ice slush appears in the river surface area, and ice slush is included, the method further includes:

[0085] In the river surface area in the subsequent image data, the coverage area of ​​the ice is determined according to the area after the pixel values ​​of the first pixel range are eliminated. For example, the coverage area of ​​each piece of ice can be determined separately.

[0086] Based on the determined coverage area and the empirical model of ice, the coverage of ice is estimated. In some examples, the empirical model can be combined to determine the volume of ice exposed above the water surface under normal circumstances and the coverage area on the water surface, thereby establishing a mapping relationship. The amount of ice coverage on the river surface can be estimated based on the determined coverage area.

[0087] An embodiment of the present application further proposes an ice monitoring system based on video monitoring, including a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the steps of the ice monitoring method based on video monitoring as described above are implemented.

[0088] In addition, although exemplary embodiments have been described herein, the scope includes any and all embodiments based on the present disclosure with equivalent elements, modifications, omissions, combinations (e.g., various embodiments intersecting schemes), adaptations or changes. It is not limited to the examples described in this specification or during the implementation of this application, and its examples will be interpreted as non-exclusive.

[0089] The above description is intended to be illustrative rather than restrictive. For example, the above examples (or one or more of them) can be used in combination with each other. For example, those of ordinary skill in the art may use other embodiments when reading the above description.

[0090] The above embodiments are merely exemplary embodiments of the present disclosure. Those skilled in the art may make various modifications or equivalent substitutions to the present invention within the essence and protection scope of the present disclosure, and such modifications or equivalent substitutions shall also be deemed to fall within the protection scope of the present invention.

Claims

1. An ice monitoring method based on video monitoring, characterized in that: include: Based on the geographical elevation of the basin, the basin is divided into multiple sub-basin sections in sequence according to the basin relationship, so that the geographical elevation of each sub-basin section is within the set elevation difference range, and the sub-basin sections in sequence have similar temperature difference distribution; According to the weather data, for a partial area in the sub-basin section where ice may appear, video data of the watershed in the partial area is obtained; For the acquired video data, according to the time relationship, the image data at the designated time is intercepted, and the river bank boundary and the river width in the intercepted image data are identified; According to the identified river bank boundary, a pixel range of a preset proportion of the river surface width on both sides of the river bank boundary is extracted as a possible ice starting area, and other areas in the river surface are used as flow areas; For the possible ice initiation area, a preset river bank image without ice is used for matching to determine whether there is ice growth at the river bank boundary; If ice growth is confirmed, the basin will be monitored continuously, and the growth and flow of ice will be monitored based on the flow area.

2. The ice monitoring method based on video monitoring according to claim 1, characterized in that: Identifying the riverbank boundaries in the captured image data includes: Identifying a first pixel range covered by a water area in the image data, and adding a disturbed color interval with set transparency to the first pixel range to obtain a second pixel range, wherein the disturbed color interval is determined based on a color interval of a shadow and a riverbed area; Outside the boundary of the first pixel range covered by the water area, a continuous boundary is determined based on the second pixel range to use the continuous boundary as the river bank boundary.

3. The ice monitoring method based on video monitoring according to claim 2, characterized in that: The river width in the captured image data is identified as follows: Select any reference point on the river bank boundary, and determine the opposite bank reference point with the closest pixel distance on the opposite bank boundary based on the reference point; and Based on any of the reference points, multiple reference points are selected according to a set pixel spacing, and a reference point on the opposite bank with the closest pixel distance is determined at the side bank boundary; Fitting the midpoints of the connecting lines between the reference point and the opposite bank reference point, and between each reference point and the corresponding opposite bank reference point; The similarity between the fitted curve and the river bank boundary is calculated, and when the similarity meets a preset threshold, the river width is determined based on the reference point and the reference point on the opposite bank, and the fitted curve is used as a river streamline.

4. The ice monitoring method based on video monitoring according to claim 3, characterized in that: Before matching the possible ice starting area with a preset riverbank image without ice, the method further includes: Extracting the river bank boundary of the preset river bank image in the absence of ice; matching the identified riverbank boundary with the riverbank boundary of the intercepted image data; Select a reference position from the matched partial area; The reference position of the captured image data is color balanced based on the reference position of the river bank image without ice, so that the color deviation of the reference position is within a set range after color balancing.

5. The ice monitoring method based on video monitoring according to claim 4, characterized in that: For the possible ice initiation area, matching is performed using a preset riverbank image without ice, including: For the matched partial area, the area outside the preset pixel distance from the river bank boundary is eliminated; Segment the removed area; Determine a third pixel range contained in a sub-region of the intercepted image data and a color deviation between the segmented sub-regions; When the third pixel range includes preset ice edge pixels and the color deviation between corresponding sub-areas exceeds the deviation threshold, it is determined that ice is growing on the river bank boundary.

6. The ice monitoring method based on video monitoring according to claim 5, characterized in that: It also includes continuous monitoring of river bank boundaries to determine if there is ice growth; as well as, intercepting image data from the subsequently acquired video data; In the river surface area in the subsequent image data, pixel values ​​in the first pixel range are removed to determine whether ice appears in the river surface area; When it is determined that ice is present in the river surface area, the flow velocity of the ice is calculated based on the river flow line and a comparison image with subsequent image data at a specified time interval.

7. The ice monitoring method based on video monitoring according to claim 6, characterized in that: Calculating the flow velocity of ice based on the river flow line and the comparison image with the subsequent image data at a specified time interval includes: In the subsequent image data, determining the position of the ice closest to the stream line of the river; Based on the river flow line, searching for the determined ice position in the comparison image at a specified time interval to find a second position in the flow direction; According to the second position and the ice position, fitting is performed with reference to the river streamline to obtain a flow distance; The flow velocity of the ice is determined according to the flow distance and the time interval.

8. The ice monitoring method based on video monitoring according to claim 6, characterized in that: When ice and icicles are detected in the river area, including floating ice, the following shall also be included: In the river surface area in the subsequent image data, the coverage area of ​​the ice is determined according to the area after the pixel values ​​of the first pixel range are removed; Based on the determined coverage area, the ice coverage is estimated according to the empirical model of ice.

9. An ice monitoring system based on video surveillance, characterized in that: It comprises a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the steps of the ice monitoring method based on video monitoring as described in any one of claims 1 to 8 are implemented.

Images