Marine oil spill quantity estimation method and system based on image depth analysis and medium

Abstract

The invention discloses a marine oil spill quantity estimation method and system based on image depth analysis and a medium, and relates to the technical field of image data analysis, and the system comprises a time period analysis module, an image data processing module and a diffusion analysis estimation module. The output end of the time period analysis module is connected with the input end of the image data processing module; and the output end of the image data processing module is connected with the input end of the diffusion analysis and estimation module. Aiming at still water experiment analysis of marine oil spill, a brand new analysis processing direction is provided, under different characteristic conditions, motion characteristic, rule and trajectory analysis in a related complex dynamic condition environment is achieved, prediction diffusion range estimation of marine oil spill is formed, and therefore influences are accurately judged and decisions are made in advance.

Description

Technical Field

[0001] The present invention relates to the technical field of image data analysis, and in particular to a method, system and medium for estimating the amount of marine oil spill based on image depth analysis. Background Art

[0002] With the development of the oil industry and the maritime transport industry, marine oil spill accidents continue to occur, becoming a huge hidden danger threatening the marine environment and social development. When an oil spill occurs in the ocean, satellite remote sensing technology or drone aerial photography is usually used to collect oil spill image information in the sea area where the accident occurred. This information includes the specific location coordinates of the accident, the area of ​​the oil spill, etc. However, due to the influence of many factors such as weather, oil spill accidents themselves have the characteristics of suddenness and uncertainty, and the collected real-time information is of limited help in the detection and impact prediction of oil spills.

[0003] In current research, the analysis of marine oil spills generally remains at the static water experiment stage, lacking specific feature analysis, such as the movement characteristics, laws and trajectories of surface oil spills under complex dynamic conditions such as wind, waves, and currents. However, there are many factors affecting marine oil spills, and the cross-combination of different influencing factors can easily affect the uncertainty of diffusion, making it impossible to accurately determine the impact area of ​​marine oil spills. Summary of the invention

[0004] The purpose of the present invention is to provide a method, system and medium for estimating the amount of oil spilled in the ocean based on image depth analysis, so as to solve the problems raised in the prior art.

[0005] To achieve the above object, the present invention provides the following technical solution: a method for estimating the amount of marine oil spill based on image depth analysis, the method comprising the following steps:

[0006] S1. Set a fixed time period, obtain each batch of marine oil spill image data under historical data, and select according to the fixed time period;

[0007] S2. Based on the selected image data, mark the initial image data and mark the center point on the initial image data;

[0008] S3, establishing an angle ray at the center point, setting a segmentation angle, wherein the segmentation angle can divide 360 ​​degrees into equal parts, extending the angle ray according to the segmentation angle, and recording the intersection point of the ray with the edge point of the oil spill in the image data as the focus data on the angle ray;

[0009] S4, acquiring the marine oil spill image data of the same batch based on the angle ray, taking the center point as the coordinate origin, forming a coordinate system of the image plane, acquiring the focal data coordinates of the image data in each fixed time period, and acquiring the feature data of the image data in each fixed time period;

[0010] S5. Based on the feature data of the image data in each fixed time period and the change in the coordinates of the focus data of the image data in each adjacent fixed time period, a marine oil spill diffusion estimation model is formed to analyze the predicted diffusion range of the marine oil spill under different features.

[0011] According to the above technical solution, in step S1, each batch refers to the same marine oil spill incident;

[0012] The fixed time period refers to setting a fixed time period, and selecting image data at the current time point every fixed time period.

[0013] According to the above technical solution, in step S2, the initial image data refers to the first image data of an ocean oil spill incident;

[0014] The center point refers to the minimum rotated rectangle circumscribed by the irregular outline of the marine oil spill image presented in the initial image data. The center point of the minimum rotated rectangle is used as the center point of the mark. The subsequent marine oil spill image data of the same batch all use the center point of the initial image data.

[0015] According to the above technical solution, in steps S3-S5, it also includes:

[0016] Under historical data, take any batch of marine oil spill image data and fill in feature data, which include: wind speed, ocean current, temperature and salinity;

[0017] Take any characteristic data as the research feature, record it as research feature A, and build the laboratory environment:

[0018] The laboratory environment includes: a water storage device for containing liquid, a wave maker, a flow maker, a fan and an oil outlet arranged in the water storage device, an oil supply device connected to the oil outlet, a plurality of image acquisition devices uniformly distributed above the water pool for collecting images of oil spills in the water pool, and a computer connected to the wave maker, the flow maker, the fan, the oil supply device and the image acquisition device;

[0019] Take any batch of marine oil spill image data to form a data set, which includes: [θ, d 1 ,d 2 ,…,d n 、T 0 ], where θ represents the angle between the center point and the positive direction of the x-axis in the coordinate system with the center point as the origin; d 1 ,d 2 ,…,d n Respectively represent the diffusion distances in the 1st to nth fixed time periods; the diffusion distance refers to the distance between the focal coordinates corresponding to the two time endpoints of a fixed time period; T 0 Represents feature data;

[0020] With θ, T 0 As initial input, keep other characteristic data except research characteristic A unchanged, and adjust the laboratory environment: the adjustment includes gradually increasing or decreasing research characteristic A;

[0021] Form different data sets in laboratory environment: [θ, t 1 ,t 2 ,…,t n , T 0 ], where t 1 ,t 2 ,…,t n They represent the diffusion distances in the 1st to nth fixed time periods under the tth group of data respectively; in the first fixed time period, the difference of the research feature A is used as the horizontal axis, and the corresponding diffusion distance data difference is used as the vertical axis to form (A i -A 0 , d i -d 1 ), where A i represents the value of the research feature A for the ith adjustment, d i represents the diffusion distance in the first fixed time period of the formation of the i-th regulation; A 0 Refers to the value of the research feature A under the initial input; i ranges from 1 to M, where M represents the total number of adjustments;

[0022] Based on the M points formed, the variation function of the current batch of marine oil spill image data is fitted: y = k 1 *a+b, where k 1 represents the slope of the variation function, y represents the diffusion distance; a represents the difference of the research feature A; b represents the constant term;

[0023] For historical data, the variation functions of all batches of marine oil spill image data under the first fixed time period are solved respectively; the maximum and minimum slope values ​​of all variation functions formed, and the maximum and minimum values ​​of the constant terms are taken to form the function interval under the first fixed time period.

[0024] According to the above technical solution, it also includes:

[0025] Two sets of historical marine oil spill image data with similar research features A are obtained, and the similar research features A include:

[0026] Take any two sets of historical marine oil spill image data, calculate the absolute value of the difference of other features under the premise that the research feature A is different, and sum up all the absolute values, and take the two sets of data with the smallest absolute value sum as the two sets of historical marine oil spill image data with similar research feature A;

[0027] In the first fixed time period of two sets of historical marine oil spill image data with similar research characteristics A, the difference of research characteristics A is used as the horizontal coordinate, and the corresponding diffusion distance data difference is used as the vertical coordinate to form coordinate scatter points, and an arbitrary straight line is constructed in the coordinate system, and the straight line satisfies the function interval of the slope and the constant term in the first fixed time period;

[0028] Among the satisfied straight lines, the straight line with the largest number of coordinate scatter points is selected as the change function of the research feature A under the first fixed time period. If there are several straight lines that pass through the largest and identical number of coordinate scatter points, the straight line with the smallest sum of the distances from all coordinate scatter points to the straight line is selected as the change function of the research feature A under the first fixed time period.

[0029] According to the above technical solution, it also includes:

[0030] Calculate and form the variation function of each research feature under each fixed time period;

[0031] Acquire the initial image data, analyze the predicted diffusion range of marine oil spills under different characteristics in different fixed time periods based on the change function of each research feature formed in each fixed time period, and feed back to the administrator port.

[0032] A marine oil spill estimation system based on image depth analysis, the system comprises: a time period analysis module, an image data processing module and a diffusion analysis estimation module;

[0033] The time period analysis module is used to set a fixed time period, obtain each batch of marine oil spill image data under historical data, and select according to the fixed time period; the image data processing module marks the initial image data based on the selected image data, and marks the center point on the initial image data; at the same time, an angle ray is established on the center point, and a segmentation angle is set, the segmentation angle can divide 360 ​​degrees into equal parts, the angle ray is extended according to the segmentation angle, and the intersection point of the ray with the edge point of the oil spill amount in the image data is recorded as the focal point data on the angle ray; based on the angle ray, the marine oil spill image data under the same batch is obtained, and the center point is used as the coordinate origin to form a coordinate system of the image plane, and the focal point data coordinates of the image data under each fixed time period are obtained, and the feature data of the image data under each fixed time period are obtained; the diffusion analysis estimation module forms a marine oil spill diffusion estimation model based on the feature data of the image data under each fixed time period and the change of the focal point data coordinates of the image data under each adjacent fixed time period, and analyzes the predicted diffusion range of the marine oil spill under different characteristics;

[0034] The output end of the time period analysis module is connected to the input end of the image data processing module; the output end of the image data processing module is connected to the input end of the diffusion analysis estimation module.

[0035] According to the above technical solution, the image data processing module includes a center point processing unit and a feature data analysis unit;

[0036] The center point processing unit is used to mark the center point on the initial image data; the center point refers to the minimum rotation rectangle circumscribed by the irregular outline of the marine oil spill image presented in the initial image data, and the center point of the minimum rotation rectangle is used as the center point of the mark. The subsequent marine oil spill image data of the same batch all use the center point of the initial image data, and at the same time, an angle ray is established on the center point to set the segmentation angle; the feature data analysis unit obtains the marine oil spill image data of the same batch based on the angle ray, takes the center point as the coordinate origin, forms a coordinate system of the image plane, obtains the focus data coordinates of the image data in each fixed time period, and obtains the feature data of the image data in each fixed time period; the feature data includes: wind speed, ocean current, temperature and salinity;

[0037] The output end of the central point processing unit is connected to the input end of the characteristic data analysis unit.

[0038] According to the above technical solution, the diffusion analysis and estimation module includes a marine oil spill diffusion estimation model analysis unit and a feedback unit;

[0039] The marine oil spill diffusion estimation model analysis unit forms a marine oil spill diffusion estimation model based on the feature data of the image data in each fixed time period and the focus data coordinate changes of the image data in each adjacent fixed time period, and analyzes the predicted diffusion range of the marine oil spill under different features; the feedback unit is connected to the administrator port to form a predicted diffusion range map and feed it back to the administrator port.

[0040] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for estimating the amount of oil spilled in the ocean based on image depth analysis

[0041] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention proposes a new analysis and processing direction for static water experimental analysis of marine oil spills, realizes the analysis of movement characteristics, laws and trajectories under complex dynamic conditions under different characteristic conditions, forms an estimate of the predicted diffusion range of marine oil spills, thereby accurately judging the impact and making decisions in advance. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic flow chart of the method for estimating the amount of marine oil spill based on image depth analysis of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] Example: Figure 1 As shown, the present invention provides a method for estimating the amount of marine oil spill based on image depth analysis, the method comprising:

[0045] A fixed time period is set to obtain each batch of marine oil spill image data under historical data, and the data is selected according to the fixed time period; each batch refers to the same marine oil spill incident; the fixed time period refers to setting a fixed duration and selecting the image data at the current time point every fixed duration.

[0046] In this embodiment, the unit is minutes, and 0.5 minutes is recorded as a fixed time period. Then, there is a batch of marine oil spill image data, and a piece of marine oil spill image data is taken every 0.5 minutes from the initial diffusion of the marine oil spill;

[0047] Based on the selected image data, the initial image data is marked, and the center point is marked on the initial image data; the initial image data refers to the first image data in a marine oil spill incident; in this embodiment, it refers to the first image data at the beginning of the initial diffusion of the marine oil spill, and the irregular contour of the first image data is marked in the OpenCV software, and the minimum rotation rectangle circumscribed to the irregular contour is used, and the center point of the minimum rotation rectangle is used as the center point of the mark, and the subsequent marine oil spill image data of the same batch all use the center point of the initial image data;

[0048] An angle ray is established at the center point, and a segmentation angle is set, wherein the segmentation angle can divide 360 ​​degrees into equal parts, and the angle ray is extended according to the segmentation angle, and the intersection point of the ray with the edge point of the oil spill in the image data is recorded as the focal data on the angle ray; the center point is used as the ray vertex, and 1 degree is used as the segmentation angle to form a ray, and the intersection point of the other end of the ray with the irregular contour is recorded as the focal data; based on the angle ray, the marine oil spill image data of the same batch is obtained, and the center point is used as the coordinate origin to form a coordinate system of the image plane, and the focal data coordinates of the image data in each fixed time period are obtained, and the characteristic data of the image data in each fixed time period are obtained;

[0049] Under historical data, take any batch of marine oil spill image data and fill in feature data, which include: wind speed, ocean current, temperature and salinity;

[0050] Take any characteristic data as the research feature, recorded as research feature A. For example, take wind speed as an example to construct the laboratory environment:

[0051] The laboratory environment includes: a water storage device for containing liquid, a wave maker, a flow maker, a fan and an oil outlet arranged in the water storage device, an oil supply device connected to the oil outlet, a plurality of image acquisition devices uniformly distributed above the water pool for collecting images of oil spills in the water pool, and a computer connected to the wave maker, the flow maker, the fan, the oil supply device and the image acquisition device;

[0052] Take any batch of marine oil spill image data to form a data set, which includes: [θ, d 1 d 2 ,…,d n , T 0 ], where θ represents the angle between the center point and the positive direction of the x-axis in the coordinate system with the center point as the origin; d 1 d 2 ,…,d n Respectively represent the diffusion distances in the 1st to nth fixed time periods; the diffusion distance refers to the distance between the focal coordinates corresponding to the two time endpoints of a fixed time period; T 0 Representative characteristic data; in this embodiment, strong wind 10.8m / s is taken as the research characteristic;

[0053] With θ, T 0 As initial input, keep other characteristic data except research characteristic A unchanged, and adjust the laboratory environment: the adjustment includes gradually increasing or decreasing research characteristic A;

[0054] Form different data sets in laboratory environment: [θ, t 1 ,t 2 ,…,t n , T 0 ], where t 1 ,t 2 ,…,t n They represent the diffusion distances in the 1st to nth fixed time periods under the tth group of data respectively; in the first fixed time period, the difference of the research feature A is used as the horizontal axis, and the corresponding diffusion distance data difference is used as the vertical axis to form (A i -A 0 , d i -d 1 ), where A i represents the value of the research feature A for the ith adjustment, d i represents the diffusion distance in the first fixed time period of the formation of the i-th regulation; A 0 Refers to the value of the research feature A under the initial input; i ranges from 1 to M, where M represents the total number of adjustments;

[0055] Based on the M points formed, the variation function of the current batch of marine oil spill image data is fitted: y = k 1 *a+b, where k 1 represents the slope of the variation function, y represents the diffusion distance; a represents the difference of the research feature A; b represents the constant term;

[0056] For historical data, the variation functions of all batches of marine oil spill image data under the first fixed time period are solved respectively; the maximum and minimum slope values ​​of all variation functions formed, and the maximum and minimum values ​​of the constant terms are taken to form the function interval under the first fixed time period.

[0057] Two sets of historical marine oil spill image data with similar research features A are obtained, and the similar research features A include:

[0058] Take any two sets of historical marine oil spill image data, calculate the absolute value of the difference of other features under the premise that the research feature A is different, and sum up all the absolute values, and take the two sets of data with the smallest absolute value sum as the two sets of historical marine oil spill image data with similar research feature A;

[0059] In the first fixed time period of two sets of historical marine oil spill image data with similar research characteristics A, the difference of research characteristics A is used as the horizontal coordinate, and the corresponding diffusion distance data difference is used as the vertical coordinate to form coordinate scatter points, and an arbitrary straight line is constructed in the coordinate system, and the straight line satisfies the function interval of the slope and the constant term in the first fixed time period;

[0060] Among the satisfied straight lines, the straight line with the largest number of coordinate scatter points is selected as the change function of the research feature A under the first fixed time period. If there are several straight lines that pass through the largest and identical number of coordinate scatter points, the straight line with the smallest sum of the distances from all coordinate scatter points to the straight line is selected as the change function of the research feature A under the first fixed time period.

[0061] Calculate and form the variation function of each research feature under each fixed time period;

[0062] Acquire the initial image data, and form a marine oil spill diffusion estimation model based on the feature data of the image data in each fixed time period and the change in the focal data coordinates of the image data in each adjacent fixed time period. Analyze and form the predicted diffusion range of marine oil spills under different features in different fixed time periods, and feed back to the administrator port.

[0063] In this embodiment, a marine oil spill estimation system based on image depth analysis is also provided, the system comprising: a time period analysis module, an image data processing module and a diffusion analysis estimation module;

[0064] The time period analysis module is used to set a fixed time period, obtain each batch of marine oil spill image data under historical data, and select according to the fixed time period; the image data processing module marks the initial image data based on the selected image data, and marks the center point on the initial image data; at the same time, an angle ray is established on the center point, and a segmentation angle is set, the segmentation angle can divide 360 ​​degrees into equal parts, the angle ray is extended according to the segmentation angle, and the intersection point of the ray with the edge point of the oil spill amount in the image data is recorded as the focal point data on the angle ray; based on the angle ray, the marine oil spill image data under the same batch is obtained, and the center point is used as the coordinate origin to form a coordinate system of the image plane, and the focal point data coordinates of the image data under each fixed time period are obtained, and the feature data of the image data under each fixed time period are obtained; the diffusion analysis estimation module forms a marine oil spill diffusion estimation model based on the feature data of the image data under each fixed time period and the change of the focal point data coordinates of the image data under each adjacent fixed time period, and analyzes the predicted diffusion range of the marine oil spill under different characteristics;

[0065] The output end of the time period analysis module is connected to the input end of the image data processing module; the output end of the image data processing module is connected to the input end of the diffusion analysis estimation module.

[0066] The image data processing module includes a center point processing unit and a feature data analysis unit;

[0067] The center point processing unit is used to mark the center point on the initial image data; the center point refers to the minimum rotation rectangle circumscribed by the irregular outline of the marine oil spill image presented in the initial image data, and the center point of the minimum rotation rectangle is used as the center point of the mark. The subsequent marine oil spill image data of the same batch all use the center point of the initial image data, and at the same time, an angle ray is established on the center point to set the segmentation angle; the feature data analysis unit obtains the marine oil spill image data of the same batch based on the angle ray, takes the center point as the coordinate origin, forms a coordinate system of the image plane, obtains the focus data coordinates of the image data in each fixed time period, and obtains the feature data of the image data in each fixed time period; the feature data includes: wind speed, ocean current, temperature and salinity;

[0068] The output end of the central point processing unit is connected to the input end of the characteristic data analysis unit.

[0069] The diffusion analysis and estimation module includes an ocean oil spill diffusion estimation model analysis unit and a feedback unit;

[0070] The marine oil spill diffusion estimation model analysis unit forms a marine oil spill diffusion estimation model based on the feature data of the image data in each fixed time period and the focus data coordinate changes of the image data in each adjacent fixed time period, and analyzes the predicted diffusion range of the marine oil spill under different features; the feedback unit is connected to the administrator port to form a predicted diffusion range map and feed it back to the administrator port.

[0071] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for estimating the amount of oil spilled in the ocean based on image depth analysis

[0072] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A method for estimating the amount of marine oil spills based on image depth analysis, characterized in that: The method comprises the following steps: S1. Set a fixed time period, obtain each batch of marine oil spill image data under historical data, and select according to the fixed time period; S2. Based on the selected image data, mark the initial image data and mark the center point on the initial image data; S3, establishing an angle ray at the center point, setting a segmentation angle, wherein the segmentation angle can divide 360 ​​degrees into equal parts, extending the angle ray according to the segmentation angle, and recording the intersection point of the ray with the edge point of the oil spill in the image data as the focus data on the angle ray; S4, acquiring the marine oil spill image data of the same batch based on the angle ray, taking the center point as the coordinate origin, forming a coordinate system of the image plane, acquiring the focal data coordinates of the image data in each fixed time period, and acquiring the feature data of the image data in each fixed time period; S5. Based on the feature data of the image data in each fixed time period and the change in the coordinates of the focus data of the image data in each adjacent fixed time period, a marine oil spill diffusion estimation model is formed to analyze the predicted diffusion range of the marine oil spill under different features.

2. The method for estimating the amount of oil spilled in the ocean based on image depth analysis according to claim 1, characterized in that: In step S1, each batch refers to the same marine oil spill incident; The fixed time period refers to setting a fixed time period, and selecting image data at the current time point every fixed time period.

3. The method for estimating the amount of oil spilled in the ocean based on image depth analysis according to claim 1, characterized in that: In step S2, the initial image data refers to the first image data of an ocean oil spill incident; The center point refers to the minimum rotated rectangle circumscribed by the irregular outline of the marine oil spill image presented in the initial image data. The center point of the minimum rotated rectangle is used as the center point of the mark. The subsequent marine oil spill image data of the same batch all use the center point of the initial image data.

4. The method for estimating the amount of oil spilled in the ocean based on image depth analysis according to claim 1, characterized in that: In steps S3-S5, it also includes: Under historical data, take any batch of marine oil spill image data and fill in feature data, which include: wind speed, ocean current, temperature and salinity; Take any characteristic data as the research feature, record it as research feature A, and build the laboratory environment: The laboratory environment includes: a water storage device for containing liquid, a wave maker, a flow maker, a fan and an oil outlet arranged in the water storage device, an oil supply device connected to the oil outlet, a plurality of image acquisition devices uniformly distributed above the water pool for collecting images of oil spills in the water pool, and a computer connected to the wave maker, the flow maker, the fan, the oil supply device and the image acquisition device; Take any batch of marine oil spill image data to form a data set, which includes: [θ, d1, d2, ..., d n , T0], where θ represents the angle between the positive direction of the x-axis and the coordinate system with the center point as the origin; d1, d2, …, d n Respectively represent the diffusion distances in the 1st to nth fixed time periods; the diffusion distance refers to the distance between the focal coordinates corresponding to the two time endpoints of a fixed time period; T0 represents the characteristic data; Taking θ and T0 as initial inputs, keeping other characteristic data except the research characteristic A unchanged, adjusting the laboratory environment: the adjustment includes gradually increasing or decreasing the research characteristic A; Form different data sets in the laboratory environment: [θ, t1, t2, …, t n , T0], where t1, t2, …, t n They represent the diffusion distances in the 1st to nth fixed time periods under the tth group of data respectively; in the first fixed time period, the difference of the research feature A is used as the horizontal axis, and the corresponding diffusion distance data difference is used as the vertical axis to form (A i -A0,d i -d1), where A i represents the value of the research feature A for the ith adjustment, d i represents the diffusion distance in the first fixed time period of the formation of the i-th adjustment; A0 refers to the value of the research feature A under the initial input; i belongs to 1 to M, where M represents the total number of adjustments; Based on the M points formed, the variation function of the current batch of marine oil spill image data is fitted: y=k1*a+b, where k1 represents the slope of the variation function, y represents the diffusion distance; a represents the difference of the research feature A; and b represents the constant term; For historical data, the variation functions of all batches of marine oil spill image data under the first fixed time period are solved respectively; the maximum and minimum slope values ​​of all variation functions formed, and the maximum and minimum values ​​of the constant terms are taken to form the function interval under the first fixed time period.

5. The method for estimating the amount of oil spilled in the ocean based on image depth analysis according to claim 4, characterized in that: Also includes: Two sets of historical marine oil spill image data with similar research features A are obtained, and the similar research features A include: Take any two sets of historical marine oil spill image data, calculate the absolute value of the difference of other features under the premise that the research feature A is different, and sum up all the absolute values, and take the two sets of data with the smallest absolute value sum as the two sets of historical marine oil spill image data with similar research feature A; In the first fixed time period of two sets of historical marine oil spill image data with similar research characteristics A, the difference of research characteristics A is used as the horizontal coordinate, and the corresponding diffusion distance data difference is used as the vertical coordinate to form coordinate scatter points, and an arbitrary straight line is constructed in the coordinate system, and the straight line satisfies the function interval of the slope and the constant term in the first fixed time period; Among the satisfied straight lines, the straight line with the largest number of coordinate scatter points is selected as the change function of the research feature A under the first fixed time period. If there are several straight lines that pass through the largest and identical number of coordinate scatter points, the straight line with the smallest sum of the distances from all coordinate scatter points to the straight line is selected as the change function of the research feature A under the first fixed time period.

6. The method for estimating the amount of oil spilled in the ocean based on image depth analysis according to claim 5, characterized in that: Also includes: Calculate and form the variation function of each research feature under each fixed time period; Acquire the initial image data, analyze the predicted diffusion range of marine oil spills under different characteristics in different fixed time periods based on the change function of each research feature formed in each fixed time period, and feed back to the administrator port.

7. Marine oil spill estimation system based on image depth analysis, characterized by: The system includes: a time period analysis module, an image data processing module and a diffusion analysis estimation module; The time period analysis module is used to set a fixed time period, obtain each batch of marine oil spill image data under historical data, and select according to the fixed time period; the image data processing module marks the initial image data based on the selected image data, and marks the center point on the initial image data; at the same time, an angle ray is established on the center point, and a segmentation angle is set, the segmentation angle can divide 360 ​​degrees into equal parts, the angle ray is extended according to the segmentation angle, and the intersection point of the ray with the edge point of the oil spill amount in the image data is recorded as the focal point data on the angle ray; based on the angle ray, the marine oil spill image data under the same batch is obtained, and the center point is used as the coordinate origin to form a coordinate system of the image plane, and the focal point data coordinates of the image data under each fixed time period are obtained, and the feature data of the image data under each fixed time period are obtained; the diffusion analysis estimation module forms a marine oil spill diffusion estimation model based on the feature data of the image data under each fixed time period and the change of the focal point data coordinates of the image data under each adjacent fixed time period, and analyzes the predicted diffusion range of the marine oil spill under different characteristics; The output end of the time period analysis module is connected to the input end of the image data processing module; the output end of the image data processing module is connected to the input end of the diffusion analysis estimation module.

8. The marine oil spill estimation system based on image depth analysis according to claim 7, characterized in that: The image data processing module includes a center point processing unit and a feature data analysis unit; The center point processing unit is used to mark the center point on the initial image data; the center point refers to the minimum rotation rectangle circumscribed by the irregular outline of the marine oil spill image presented in the initial image data, and the center point of the minimum rotation rectangle is used as the center point of the mark. The subsequent marine oil spill image data of the same batch all use the center point of the initial image data, and at the same time, an angle ray is established on the center point to set the segmentation angle; the feature data analysis unit obtains the marine oil spill image data of the same batch based on the angle ray, takes the center point as the coordinate origin, forms a coordinate system of the image plane, obtains the focus data coordinates of the image data in each fixed time period, and obtains the feature data of the image data in each fixed time period; The characteristic data include: wind speed, ocean current, temperature and salinity; The output end of the central point processing unit is connected to the input end of the characteristic data analysis unit.

9. The marine oil spill estimation system based on image depth analysis according to claim 7, characterized in that: The diffusion analysis and estimation module includes an ocean oil spill diffusion estimation model analysis unit and a feedback unit; The marine oil spill diffusion estimation model analysis unit forms a marine oil spill diffusion estimation model based on the feature data of the image data in each fixed time period and the focus data coordinate changes of the image data in each adjacent fixed time period, and analyzes the predicted diffusion range of the marine oil spill under different features; the feedback unit is connected to the administrator port to form a predicted diffusion range map and feed it back to the administrator port.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for estimating the amount of marine oil spill based on image depth analysis as described in any one of claims 1 to 6 is implemented.

Images