A method and system for monitoring pollutant flow in upstream and downstream of a river

Abstract

The present invention provides a method and system for monitoring the flow of pollutants based on the upstream and downstream of a river, and relates to the technical field of river pollutant monitoring. The method comprises: S1: obtaining a current monitoring image set arranged from upstream to downstream in the order of rivers; S2: traversing the monitoring image set, and sequentially identifying pollutants in the first river image. When pollutants are identified in the first river image, the traversal is stopped and the first river image is marked as a pollutant source image, and then the process goes to step S3; S3: taking screenshots of a first video to be identified at a second screenshot frequency to obtain a plurality of second river images; S4: sequentially identifying pollutants in the plurality of second river images within a time range, and judging and obtaining the pollutant flow corresponding to the pollutants identified in the pollutant source image according to the pollutant identification result; S5: arranging and displaying the monitoring images in the flow of pollutants from upstream to downstream in the order of rivers.

Description

Technical Field

[0001] The present invention relates to the technical field of river pollutant monitoring, and in particular to a method and system for monitoring pollutant flow in upstream and downstream of a river. Background Art

[0002] At present, in the field of river pollutant monitoring, river patrol personnel are generally used to patrol designated rivers. Due to many limitations of manual operation, the following problems often occur during the patrol process:

[0003] (1) Although river patrol officers patrol frequently and have a relatively high rate of problem discovery, river emergencies may still occur during non-patrol periods, which makes it impossible for river patrol officers to discover and report related problems in a timely manner;

[0004] (2) River patrol officers often encounter problems beyond their professional and technical capabilities, such as being unable to identify water quality safety risks on site. After reporting problems to higher levels through telephone or other means, the process of obtaining the final judgment result is long, and most of the time it takes is about 30 to 60 minutes, which means the handling time is too long;

[0005] (3) There is insufficient application of monitoring data from existing sensing devices in the river section. Generally, the monitoring data of the current sensing device is obtained. When the pollutant value in the monitoring data exceeds the threshold, the sensing device will issue a single alarm message, causing the back-end control personnel to process the current sensing device without further analysis, so that the back-end control personnel cannot grasp the comprehensive information. Summary of the invention

[0006] The purpose of the present invention is to provide a method and system for monitoring the flow of pollutants based on the upstream and downstream of a river. The method mainly monitors the flow of pollutants in a river photographed by a river camera at a monitoring station based on the upstream and downstream sequence of the river. When pollutants are monitored, the flow of pollutants can be quickly and accurately formed through the upstream and downstream sequence of the river. The pollutant flow is a monitoring image corresponding to the monitoring stations arranged from upstream to downstream according to the sequence of the river. The pollutant flow is displayed so that the user can quickly and logically switch the monitoring screen according to the direction of the river when patrolling the river online, and can simulate the flow of pollutants in actual river patrols online.

[0007] In order to solve the above technical problems, the present invention adopts the following solutions:

[0008] A method for monitoring pollutant flow in the upstream and downstream of a river, the method specifically comprising the following steps:

[0009] S1: obtaining a current monitoring image set arranged from upstream to downstream in river sequence, wherein the monitoring image set includes a plurality of first site image sets corresponding to monitoring sites, wherein the first site image set is composed of a plurality of first river images arranged from early to late in chronological order, and the first river images are obtained by the monitoring site by taking screenshots of a first camera video to be identified at a first screenshot frequency;

[0010] S2: Traverse the monitoring image set, identify pollutants in the first river image in turn, and when pollutants are identified in the first river image, stop traversing and mark the first river image as a pollutant source image, and then go to step S3;

[0011] S3: extracting a first camera video to be identified of a corresponding monitoring site according to the pollutant source image, and taking screenshots of the first camera video to be identified at a second screenshot frequency to obtain a plurality of second river images arranged in chronological order from morning to night, wherein the second screenshot frequency is greater than the first screenshot frequency;

[0012] S4: locating a plurality of second river images within the time range according to the time range corresponding to the pollutant source image, and sequentially identifying pollutants on the plurality of second river images within the time range, and determining and obtaining the pollutant flow corresponding to the pollutant identified in the pollutant source image according to the pollutant identification result;

[0013] S5: Obtain monitoring images corresponding to monitoring sites arranged from upstream to downstream in the order of the river according to the flow of pollutants, and display the monitoring images arranged from upstream to downstream in the order of the river.

[0014] Furthermore, the first camera video to be identified is the river surface captured by the monitoring site through its own river camera within a preset time period, so that the first river image is the river surface captured at the screenshot time.

[0015] Furthermore, the S4 specifically includes the following steps:

[0016] S41: extracting the screenshot time corresponding to the pollutant source image, and expanding the preset time period based on the screenshot time to obtain the time range corresponding to the pollutant source image;

[0017] S42: Locate multiple second river images within the time range, and identify pollutants in the second river images in chronological order from early to late. When pollutants are identified in the second river images, record the total number of times the pollutants are identified, and determine whether the total number meets the condition of continuous occurrence and exceeding a preset threshold. If so, go to step S43; if not, go to step S44;

[0018] S43: generating a pollutant flow path corresponding to the pollutant identified in the pollutant source image according to the current monitoring image set and the pollutant source image;

[0019] S44: Locate the first site image set corresponding to the pollutant source image in the current monitoring image set, traverse each first site image set downward with the first site image set, and obtain the pollutant image corresponding to each first site image set through the screenshot time corresponding to the pollutant source image, and generate the pollutant flow corresponding to the pollutant identified in the pollutant source image according to the pollutant source image and the pollutant image.

[0020] Furthermore, the S43 specifically includes the following steps:

[0021] S431: Using the pollutant source image as the monitoring image corresponding to the first monitoring site in the pollutant flow;

[0022] S432: Locate the first site image set corresponding to the pollutant source image in the current monitoring image set, and use the first site image set to traverse each first site image set downward, and use the screenshot time corresponding to the pollutant source image as the starting time of the pollutant flow, and then filter down a first river image in turn according to the starting time of the pollutant flow, so that the pollutant source image is sequentially connected to the filtered multiple first river images arranged in chronological order from early to late, to form a pollutant flow.

[0023] Furthermore, the S44 specifically includes the following steps:

[0024] S441: Using the pollutant source image as the monitoring image corresponding to the first monitoring site in the pollutant flow;

[0025] S442: locate the first site image set corresponding to the pollutant source image in the current monitoring image set, traverse each first site image set downward with the first site image set, take the screenshot time corresponding to the pollutant source image as the starting time of the pollutant flow, and then identify the pollutants for the first site images in each first site image set in turn according to the starting time of the pollutant flow, and determine in turn whether the first site images in each first site image set contain the pollutants identified in the pollutant source image, if yes, go to step S443, if not, go to step S444;

[0026] S443: extracting the first site image corresponding to the first site image set as a pollutant image, and arranging the pollutant image from upstream to downstream according to the river sequence and adding it to the pollutant flow;

[0027] S444: Obtain the screenshot time of the last pollutant image arranged in the current pollutant flow, extract the second camera video to be identified in the monitoring site corresponding to the first site image set by the screenshot time, screenshot the second camera video to be identified by the third screenshot frequency to obtain the third river image, and then identify pollutants in the third river image in turn, and determine whether the third river image contains pollutants identified in the pollutant source image in turn. If so, use the third river image as the pollutant image, and arrange the pollutant images from upstream to downstream in river order and add them to the pollutant flow, wherein the third screenshot frequency is greater than the first screenshot frequency.

[0028] Furthermore, the pollutant flow is composed of monitoring images corresponding to monitoring stations arranged from upstream to downstream in river order, and the monitoring images of the pollutant flow are also arranged from early to late in chronological order. A monitoring image of the pollutant flow is composed of a pollutant source image and a first river image or of a pollutant source image and a pollutant image.

[0029] Furthermore, the second camera video to be identified is the river surface within a preset time period obtained by photographing the monitoring site through its own river camera starting from the screenshot time of the last arranged pollutant image, and the river camera is a camera fixed by the monitoring site to photograph the river surface.

[0030] Furthermore, in S1, a current monitoring image set arranged from upstream to downstream in river order is obtained, and each first site image set in the monitoring image set undergoes image preprocessing, and the image preprocessing includes: preprocessing the image obtained by the screenshot, removing abnormal points in the image, and compressing the image into a single-channel grayscale image, and using the Laplace method to convolve and score the image. When the score reaches a preset value, the image is determined to be valid, and the image is used as the first river image and added to the first site image set at the screenshot time.

[0031] A pollutant flow monitoring system based on the upstream and downstream of a river, comprising:

[0032] A monitoring image set acquisition module: used to obtain a current monitoring image set arranged from upstream to downstream in river sequence, wherein the monitoring image set includes a plurality of first site image sets corresponding to monitoring sites, wherein the first site image set is composed of a plurality of first river images arranged from early to late in chronological order, and the first river image is obtained by the monitoring site by taking a screenshot of a first camera video to be identified at a first screenshot frequency;

[0033] The first river image judgment module is used to traverse the monitoring image set, identify pollutants in the first river image in turn, and when pollutants are identified in the first river image, stop traversal and mark the first river image as a pollutant source image;

[0034] A second river image acquisition module: used to extract a first camera video to be identified of a corresponding monitoring site according to the pollutant source image, and to screenshot the first camera video to be identified at a second screenshot frequency to obtain a plurality of second river images arranged in chronological order from morning to night, wherein the second screenshot frequency is greater than the first screenshot frequency;

[0035] Pollutant flow calculation module: used to locate multiple second river images within the time range according to the time range corresponding to the pollutant source image, and sequentially identify pollutants in the multiple second river images within the time range, and determine and obtain the pollutant flow corresponding to the pollutants identified in the pollutant source image according to the pollutant identification results;

[0036] Pollutant flow display module: used to obtain monitoring images corresponding to monitoring sites arranged from upstream to downstream in river sequence according to pollutant flow, and to display the monitoring images arranged from upstream to downstream in river sequence.

[0037] Furthermore, the pollutant flow calculation module specifically includes the following modules:

[0038] Time extension module: used to extract the screenshot time corresponding to the pollutant source image, and expand the preset time period with the screenshot time to obtain the time range corresponding to the pollutant source image;

[0039] The second river image judgment module is used to locate multiple second river images within a time range, and to identify pollutants in the second river images in chronological order from early to late. When pollutants are identified in the second river images, the total number of times the pollutants are identified is recorded, and it is determined whether the total number meets the condition of continuous occurrence and exceeds a preset threshold;

[0040] A first pollutant flow generation module: used to generate a pollutant flow corresponding to the pollutant identified in the pollutant source image according to the current monitoring image set and the pollutant source image;

[0041] The second pollutant flow generation module is used to locate the first site image set corresponding to the pollutant source image in the current monitoring image set, traverse each first site image set downward with the first site image set, and obtain the pollutant image corresponding to each first site image set through the screenshot time corresponding to the pollutant source image, and generate the pollutant flow corresponding to the pollutant identified in the pollutant source image according to the pollutant source image and the pollutant image.

[0042] Beneficial effects of the present invention:

[0043] The present invention provides a pollutant flow monitoring method and system based on the upstream and downstream of a river. The method mainly monitors the pollutant flow of a river photographed by a river camera at a monitoring station based on the upstream and downstream sequence of the river. When pollutants are monitored, the pollutant flow can be quickly and accurately formed through the upstream and downstream sequence of the river. The pollutant flow is a monitoring image corresponding to the monitoring stations arranged from upstream to downstream according to the sequence of the river. The pollutant flow is displayed so that the user can quickly and logically switch the monitoring screen according to the direction of the river when patrolling the river online, and can simulate the pollutant flow in the actual river patrol online. In addition, the real-time status can be updated through the upstream and downstream sequence of the river camera, so that the user does not need to visit the site in person, but can fully grasp the distribution and current status of the river and its related facilities, thereby significantly improving the efficiency of information acquisition. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a schematic flow chart of a method for monitoring pollutant flow in the upstream and downstream of a river in Example 1 of the present invention;

[0045] Figure 2 is a schematic diagram of the structure of the monitoring image set in Example 1 of the present invention;

[0046] Figure 3 This is a schematic diagram of intensively capturing a time period T in Example 1 of the present invention;

[0047] Figure 4 In Example 1 of the present invention, based on Figure 2 Schematic diagram of pollutant flow in the monitoring image set;

[0048] Figure 5 This is a schematic diagram of a page for conducting an online simulation of a river patrol process on a platform in Example 1 of the present invention. DETAILED DESCRIPTION

[0049] 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, rather than all the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present invention and its application or use. 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.

[0050] The relative arrangement of components and steps, the numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present invention unless specifically stated otherwise.

[0051] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0052] Additionally, descriptions of well-known structures, functions, and configurations may be omitted for clarity and conciseness.One of ordinary skill in the art will recognize that various changes and modifications may be made to the examples described herein without departing from the spirit and scope of the present disclosure.

[0053] Technologies, methods, and apparatus known to ordinary technicians in the relevant field may not be discussed in detail, but where appropriate, such technologies, methods, and apparatus should be considered part of the authorization specification.

[0054] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0055] The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments:

[0056] Example 1

[0057] like Figure 1 As shown, a method for monitoring pollutant flow in the upstream and downstream of a river, the method specifically comprises the following steps:

[0058] S1: obtaining a current monitoring image set arranged from upstream to downstream in river sequence, wherein the monitoring image set includes a plurality of first site image sets corresponding to monitoring sites, wherein the first site image set is composed of a plurality of first river images arranged from early to late in chronological order, and the first river images are obtained by the monitoring site by taking screenshots of a first camera video to be identified at a first screenshot frequency;

[0059] S2: Traverse the monitoring image set, identify pollutants in the first river image in turn, and when pollutants are identified in the first river image, stop traversing and mark the first river image as a pollutant source image, and then go to step S3;

[0060] S3: extracting a first camera video to be identified of a corresponding monitoring site according to the pollutant source image, and taking screenshots of the first camera video to be identified at a second screenshot frequency to obtain a plurality of second river images arranged in chronological order from morning to night, wherein the second screenshot frequency is greater than the first screenshot frequency;

[0061] S4: locating a plurality of second river images within the time range according to the time range corresponding to the pollutant source image, and sequentially identifying pollutants on the plurality of second river images within the time range, and determining and obtaining the pollutant flow corresponding to the pollutant identified in the pollutant source image according to the pollutant identification result;

[0062] S5: Obtain monitoring images corresponding to monitoring sites arranged from upstream to downstream in the order of the river according to the flow of pollutants, and display the monitoring images arranged from upstream to downstream in the order of the river.

[0063] Preferably, the first camera video to be identified is the river surface captured by the monitoring site through its own river camera within a preset time period, so that the first river image is the river surface captured at the screenshot time.

[0064] Preferably, the S4 specifically includes the following steps:

[0065] S41: extracting the screenshot time corresponding to the pollutant source image, and expanding the preset time period based on the screenshot time to obtain the time range corresponding to the pollutant source image;

[0066] S42: Locate multiple second river images within the time range, and identify pollutants in the second river images in chronological order from early to late. When pollutants are identified in the second river images, record the total number of times the pollutants are identified, and determine whether the total number meets the condition of continuous occurrence and exceeding a preset threshold. If so, go to step S43; if not, go to step S44;

[0067] S43: generating a pollutant flow path corresponding to the pollutant identified in the pollutant source image according to the current monitoring image set and the pollutant source image;

[0068] S44: Locate the first site image set corresponding to the pollutant source image in the current monitoring image set, traverse each first site image set downward with the first site image set, and obtain the pollutant image corresponding to each first site image set through the screenshot time corresponding to the pollutant source image, and generate the pollutant flow corresponding to the pollutant identified in the pollutant source image according to the pollutant source image and the pollutant image.

[0069] Preferably, the S43 specifically includes the following steps:

[0070] S431: Using the pollutant source image as the monitoring image corresponding to the first monitoring site in the pollutant flow;

[0071] S432: Locate the first site image set corresponding to the pollutant source image in the current monitoring image set, and use the first site image set to traverse each first site image set downward, and use the screenshot time corresponding to the pollutant source image as the starting time of the pollutant flow, and then filter down a first river image in turn according to the starting time of the pollutant flow, so that the pollutant source image is sequentially connected to the filtered multiple first river images arranged in chronological order from early to late, to form a pollutant flow.

[0072] Preferably, the S44 specifically includes the following steps:

[0073] S441: Using the pollutant source image as the monitoring image corresponding to the first monitoring site in the pollutant flow;

[0074] S442: locate the first site image set corresponding to the pollutant source image in the current monitoring image set, traverse each first site image set downward with the first site image set, take the screenshot time corresponding to the pollutant source image as the starting time of the pollutant flow, and then identify the pollutants for the first site images in each first site image set in turn according to the starting time of the pollutant flow, and determine in turn whether the first site images in each first site image set contain the pollutants identified in the pollutant source image, if yes, go to step S443, if not, go to step S444;

[0075] S443: extracting the first site image corresponding to the first site image set as a pollutant image, and arranging the pollutant image from upstream to downstream according to the river sequence and adding it to the pollutant flow;

[0076] S444: Obtain the screenshot time of the last pollutant image arranged in the current pollutant flow, extract the second camera video to be identified in the monitoring site corresponding to the first site image set by the screenshot time, screenshot the second camera video to be identified by the third screenshot frequency to obtain the third river image, and then identify pollutants in the third river image in turn, and determine whether the third river image contains pollutants identified in the pollutant source image in turn. If so, use the third river image as the pollutant image, and arrange the pollutant images from upstream to downstream in river order and add them to the pollutant flow, wherein the third screenshot frequency is greater than the first screenshot frequency.

[0077] Preferably, the pollutant flow is composed of monitoring images corresponding to monitoring stations arranged from upstream to downstream in river order, and the monitoring images of the pollutant flow are also arranged in chronological order from early to late. A monitoring image of the pollutant flow is composed of a pollutant source image and a first river image or of a pollutant source image and a pollutant image.

[0078] Preferably, the second camera video to be identified is the river surface within a preset time period obtained by photographing the monitoring site through its own river camera starting from the screenshot time of the last arranged pollutant image, and the river camera is a camera fixed by the monitoring site to photograph the river surface.

[0079] Preferably, in S1, a current monitoring image set arranged from upstream to downstream in river order is obtained, and each first site image set in the monitoring image set undergoes image preprocessing, and the image preprocessing includes: preprocessing the image obtained by screenshot, removing abnormal points in the image, and compressing the image into a single-channel grayscale image, using the Laplace method to perform convolution scoring on the image, and when the score reaches a preset value, the image is judged to be valid, and the image is used as the first river image, and is added to the first site image set at the screenshot time.

[0080] Based on the above principles, the present invention is further described:

[0081] The present invention is mainly to solve the problems existing in river patrol in actual situations. Although the patrol frequency of river patrol personnel is high and the problem discovery rate is relatively high, there will still be sudden incidents in the river during the non-patrol period, which makes it impossible for river patrol personnel to discover and report related problems in time. Therefore, at present, multiple monitoring stations can be set up for rivers, and the river surface can be monitored by fixing river cameras on the monitoring stations. When the monitoring station detects pollutants, it is reported through the monitoring station. At this time, the reported information is generally used to notify the staff that there are pollutants in the river section, so that the staff can control and handle the monitoring station, and it cannot provide further support data for control to the control personnel. Therefore, the present invention aims to simulate the river patrol process, monitor the flow of pollutants based on the upstream and downstream sequence of the river, and enable users to conduct online river patrols on the platform through the flow of pollutants, and can quickly and logically switch the monitoring screen according to the direction of water flow. It not only simulates the actual scenario of on-site river patrol, but also breaks through the geographical limitations, greatly improving the flexibility and efficiency of river patrol. At the same time, it integrates map functions, which can clearly display the specific locations of video monitoring points, water quality monitoring stations and pollution sources, and provide real-time status updates, so that users do not need to visit the site in person, but can fully understand the distribution and current status of the river and its related facilities, which significantly improves the efficiency of information acquisition.

[0082] In this embodiment, the upstream and downstream order of the monitoring sites can be calculated based on the locations of the monitoring sites and the distribution of river flow directions, and the upstream and downstream order of the river cameras can be established according to the upstream and downstream order of the monitoring sites. Based on the upstream and downstream order of the river cameras, the current monitoring image set arranged from upstream to downstream in the river order can be obtained. Specifically, if the upstream and downstream order of the current monitoring sites is monitoring site n1, monitoring site n2, monitoring site n3, and monitoring site n4 arranged from upstream to downstream in the river order, then the upstream and downstream order of the river cameras is established based on the upstream and downstream order of the current monitoring sites to obtain the current monitoring image set arranged from upstream to downstream in the river order, such as Figure 2 As shown, the current monitoring image set is arranged from upstream to downstream in the order of the river as the first site image set x1, the first site image set x2, the first site image set x3, and the first site image set x4, and the first site image set x1, the first site image set x2, the first site image set x3, and the first site image set x4 correspond to the monitoring site n1, the monitoring site n2, the monitoring site n3, and the monitoring site n4 respectively, and each first site image set includes 4 first river images arranged in chronological order from early to late, such as the first site image set x1 is composed of the first river image x11, the first river image x12, the first river image x13, and the first river image x14, and so on.

[0083] In order to ensure the validity of the images in the monitoring image set, the images used for pollutant identification in the monitoring image set can be preprocessed in advance in the present invention. Specifically, when the monitoring site takes a screenshot of the first camera video to be identified at the first screenshot frequency, multiple screenshots can be obtained, and the binary images of the screenshots are analyzed, and the RGB pixel threshold rgb_range = ((0, 130, 0), (5, 140, 5)) is used to remove the green dots in the picture and the abnormal green obstruction pixels caused by the growth of plants at the location where the river camera is arranged; based on the image information after removing the green dots and plant obstructions, the screenshot is compressed into a single-channel grayscale image to improve the accuracy of subsequent image evaluation, and then the Laplace method is used to perform convolution calculation and scoring on the image to obtain detail_score, and min_score=250 is set. If detail_score>min_score, it means that the screenshot is valid, and the screenshot is used as the first river image, and is added to the first site image set of the monitoring site at the screenshot time. By analogy, the first site image set of the monitoring sites in the upstream and downstream order can be obtained in turn. At this time, the first site image set of the monitoring sites in the upstream and downstream order is the screenshot that has been pre-processed within the same preset time period; if detail_score<=min_score, it means that the screenshot has a flowery screen, no picture, or a missing screenshot, then the first screenshot frequency can be slightly adjusted to re-screenshot and analyze the first camera video to be identified.

[0084] Then, the current monitoring image set is traversed, and pollutants are identified for the first river image in turn. Specifically, the first river image x11, the first river image x12, the first river image x13, and the first river image x14 in the first site image set x1 are traversed first, and pollutants are identified for the first river image x11, the first river image x12, the first river image x13, and the first river image x14 in turn. Then, the first river image x21, the first river image x22, the first river image x23, and the first river image x24 in the first site image set x2 are traversed, and so on. Pollutant identification can be achieved for all river images in the monitoring image set, which is used to preliminarily determine whether there are pollutants in the current river, which is equivalent to monitoring the river surface by fixing a river camera on the monitoring site. When the monitoring site detects pollutants, it is reported through the monitoring site.

[0085] In the present invention, when the monitoring station detects pollutants, it is necessary to obtain the camera video located downstream based on the upstream and downstream sequence of the river camera, and further judge the camera video located downstream to form the flow of pollutants. The flow of pollutants is used to simulate the discovery of pollutants in the current river patrol process, and the pollutants are further tracked, which is convenient for staff to patrol the river online and further control the river.

[0086] Since monitoring stations in rivers are generally also equipped with pollutant interception devices, when the present invention detects pollutants at the monitoring station, it can display the flow of different pollutants based on the upstream and downstream sequence of the river camera, so that staff can intuitively and clearly know the flow of pollutants when patrolling the river online.

[0087] Specifically, if pollutants are identified in the first river image x12 during the traversal of the current monitoring image set, the traversal is stopped and the first river image x12 is marked as a pollutant source image. Then, the first camera video to be identified of the corresponding monitoring site n1 is extracted again according to the pollutant source image, and the first camera video to be identified is screenshotted at the second screenshot frequency to obtain multiple second river images arranged in chronological order from early to late. At this time, the second screenshot frequency is greater than the first screenshot frequency, which means that the number of second river images obtained by the second screenshot frequency is greater than the number of first river images obtained by the first screenshot frequency.

[0088] Furthermore, since the second screenshot frequency and the first screenshot frequency are both for taking screenshots of the first camera video to be identified, the first camera video to be identified is screenshotted again by the second screenshot frequency, and pollutants are identified on the second river image, so it can be determined whether the pollutants in the pollutant source image are intercepted by the monitoring site. If they are intercepted, then the pollutants can be obtained from the video taken by the river camera for a long time to show that the pollutants exist on the surface of the river; if they are not intercepted, then the pollutants can be obtained from the video taken by the river camera after a period of time to show that the pollutants will not exist on the surface of the river. Therefore, the present invention selects a second screenshot frequency having a screenshot frequency greater than the first screenshot frequency to intensively take screenshots of the first camera video to be identified at the monitoring site n1, and according to the screenshot time of the pollutant source image, a preset time period is extended to obtain the time range corresponding to the pollutant source image, and then intensively taking screenshots of the video within the preset time can reduce the amount of data processing, which is equivalent to performing multiple screenshot processing between the pollution source image and the first river image x13 and between the first river image x11 and the pollution source image, such as Figure 3 As shown, the preset time period is extended with the screenshot time t of the pollution source image, and extended to the time period T between the first river image x11 and the first river image x13, to ensure that the screenshot time t of the pollution source image is included in the extended time period, and then the video in the time period T is densely screenshotted through the second screenshot frequency to obtain multiple second river images. The multiple second river images can then be used to determine whether the existing pollutants are intercepted by the monitoring station.

[0089] Specifically, pollutants are identified for the second river image in chronological order from morning to night. When pollutants are identified in the second river image, the total number of times the pollutants are identified is recorded, and a determination is made as to whether the total number of times satisfies the condition of continuous occurrence and exceeding a preset threshold. If so, it indicates that the pollutants in the pollutant source image are intercepted by the monitoring station; if not, it indicates that the pollutants in the pollutant source image are not intercepted by the monitoring station, indicating that the pollutants have swum to the downstream monitoring station in the river. Then, the first station image set corresponding to the pollutant source image is located according to the current monitoring image set, and each first station image set is traversed downward with the first station image set, that is, whether the pollutant exists in the downstream monitoring station in the upstream and downstream order.

[0090] Since each first site image set in the current monitoring image set is obtained according to the same screenshot frequency within the same time period, the number of first river images and the screenshot time in the first site image set are the same, and a time sequence traversal can be performed downstream according to the screenshot time in the pollutant source image. The second camera video to be identified of the downstream monitoring site is obtained according to the upstream and downstream sequence, wherein the second camera video to be identified can be the first camera video to be identified, and the camera video to be identified required for the downstream monitoring site is judged according to the position of the pollutant source image in the first site image set.

[0091] Then, the screenshot time corresponding to the pollutant source image is used as the starting time of the pollutant flow, and then the pollutants in each first site image set are identified in turn according to the starting time of the pollutant flow. Specifically, the first river image x12 in the first site image set x1 is used as the current pollutant source image, and the first river image x23 and the first river image x24 in the first site image set x2 are identified in turn according to the screenshot time corresponding to the current pollutant source image. If the first river image x23 is identified as containing the pollutants identified in the pollutant source image, the first river image x23 in the first site image set x2 is extracted as the pollutant image, and the pollutant image Arrange them from upstream to downstream in the order of rivers and add them to the pollutant flow, that is, the current pollutant flow is represented by the first river image x12-first river image x23. Then, pollutants are identified according to the first river image x33 and the first river image x34 in the first site image set x3. If the first river image x34 is found to contain pollutants identified in the pollutant source image, the first river image x34 in the first site image set x3 is extracted as the pollutant image, and the pollutant image is arranged from upstream to downstream in the order of rivers and added to the pollutant flow, that is, the current pollutant flow is represented by the first river image x12-first river image x23-first river image x34, as shown in FIG. Figure 4 As shown;

[0092] If no river pollutants are identified in the first river image in the first site image set after traversing the first site image set, it may be because the first screenshot frequency is set to a small value, resulting in the first river image in the first site image set not showing river pollutants. The present invention also provides a method of taking screenshots of the second camera video to be identified at a third screenshot frequency to obtain a third river image, i.e., dense screenshots, and then sequentially identifying pollutants in the third river image to form a flow of pollutants.

[0093] In summary, when the present invention simulates the river patrol process through the platform, a one-key river patrol can be performed based on the upstream and downstream sequence of the camera. The user can quickly and logically switch the monitoring screen according to the direction of the water flow, realizing the online simulation of the river patrol process, such as Figure 5 As shown, the flow of pollutants can be monitored based on the upstream and downstream order of the river. When one-key river patrol is clicked, the monitoring images corresponding to the monitoring stations arranged from upstream to downstream in the order of the river in the flow of pollutants are displayed, and the monitoring images are also arranged in chronological order from morning to night, so that targeted river patrol monitoring of pollutants can be achieved.

[0094] Example 2

[0095] A pollutant flow monitoring system based on the upstream and downstream of a river, comprising:

[0096] A monitoring image set acquisition module: used to obtain a current monitoring image set arranged from upstream to downstream in river sequence, wherein the monitoring image set includes a plurality of first site image sets corresponding to monitoring sites, wherein the first site image set is composed of a plurality of first river images arranged from early to late in chronological order, and the first river image is obtained by the monitoring site by taking a screenshot of a first camera video to be identified at a first screenshot frequency;

[0097] The first river image judgment module is used to traverse the monitoring image set, identify pollutants in the first river image in turn, and when pollutants are identified in the first river image, stop traversal and mark the first river image as a pollutant source image;

[0098] A second river image acquisition module: used to extract a first camera video to be identified of a corresponding monitoring site according to the pollutant source image, and to screenshot the first camera video to be identified at a second screenshot frequency to obtain a plurality of second river images arranged in chronological order from morning to night, wherein the second screenshot frequency is greater than the first screenshot frequency;

[0099] Pollutant flow calculation module: used to locate multiple second river images within the time range according to the time range corresponding to the pollutant source image, and sequentially identify pollutants in the multiple second river images within the time range, and determine and obtain the pollutant flow corresponding to the pollutants identified in the pollutant source image according to the pollutant identification results;

[0100] Pollutant flow display module: used to obtain monitoring images corresponding to monitoring sites arranged from upstream to downstream in river sequence according to pollutant flow, and to display the monitoring images arranged from upstream to downstream in river sequence.

[0101] Preferably, the pollutant flow calculation module specifically includes the following modules:

[0102] Time extension module: used to extract the screenshot time corresponding to the pollutant source image, and expand the preset time period with the screenshot time to obtain the time range corresponding to the pollutant source image;

[0103] The second river image judgment module is used to locate multiple second river images within a time range, and to identify pollutants in the second river images in chronological order from early to late. When pollutants are identified in the second river images, the total number of times the pollutants are identified is recorded, and it is determined whether the total number meets the condition of continuous occurrence and exceeds a preset threshold;

[0104] A first pollutant flow generation module: used to generate a pollutant flow corresponding to the pollutant identified in the pollutant source image according to the current monitoring image set and the pollutant source image;

[0105] The second pollutant flow generation module is used to locate the first site image set corresponding to the pollutant source image in the current monitoring image set, traverse each first site image set downward with the first site image set, and obtain the pollutant image corresponding to each first site image set through the screenshot time corresponding to the pollutant source image, and generate the pollutant flow corresponding to the pollutant identified in the pollutant source image according to the pollutant source image and the pollutant image.

[0106] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. According to the technical essence of the present invention, within the spirit and principles of the present invention, any simple modification, equivalent replacement and improvement made to the above embodiment still falls within the protection scope of the technical solution of the present invention.

Claims

1. A method for monitoring pollutant flow in the upstream and downstream of a river, characterized in that: The method specifically comprises the following steps: S1: obtaining a current monitoring image set arranged from upstream to downstream in river sequence, wherein the monitoring image set includes a plurality of first site image sets corresponding to monitoring sites, wherein the first site image set is composed of a plurality of first river images arranged from early to late in chronological order, and the first river images are obtained by the monitoring site by taking screenshots of a first camera video to be identified at a first screenshot frequency; S2: Traverse the monitoring image set, identify pollutants in the first river image in turn, and when pollutants are identified in the first river image, stop traversing and mark the first river image as a pollutant source image, and then go to step S3; S3: extracting a first camera video to be identified of a corresponding monitoring site according to the pollutant source image, and taking screenshots of the first camera video to be identified at a second screenshot frequency to obtain a plurality of second river images arranged in chronological order from morning to night, wherein the second screenshot frequency is greater than the first screenshot frequency; S4: locating a plurality of second river images within the time range according to the time range corresponding to the pollutant source image, and sequentially identifying pollutants on the plurality of second river images within the time range, and determining and obtaining the pollutant flow corresponding to the pollutant identified in the pollutant source image according to the pollutant identification result; S5: Obtain monitoring images corresponding to monitoring sites arranged from upstream to downstream in the order of the river according to the flow of pollutants, and display the monitoring images arranged from upstream to downstream in the order of the river.

2. A method for monitoring pollutant flow in upstream and downstream of a river according to claim 1, characterized in that: The first camera video to be identified is the river surface captured by the monitoring site through its own river camera within a preset time period, so that the first river image is the river surface captured at the screenshot time.

3. A method for monitoring pollutant flow in upstream and downstream of a river according to claim 1, characterized in that: The S4 specifically includes the following steps: S41: extracting the screenshot time corresponding to the pollutant source image, and expanding the preset time period based on the screenshot time to obtain the time range corresponding to the pollutant source image; S42: Locate multiple second river images within the time range, and identify pollutants in the second river images in chronological order from early to late. When pollutants are identified in the second river images, record the total number of times the pollutants are identified, and determine whether the total number meets the condition of continuous occurrence and exceeding a preset threshold. If so, go to step S43; if not, go to step S44; S43: generating a pollutant flow path corresponding to the pollutant identified in the pollutant source image according to the current monitoring image set and the pollutant source image; S44: Locate the first site image set corresponding to the pollutant source image in the current monitoring image set, traverse each first site image set downward with the first site image set, and obtain the pollutant image corresponding to each first site image set through the screenshot time corresponding to the pollutant source image, and generate the pollutant flow corresponding to the pollutant identified in the pollutant source image according to the pollutant source image and the pollutant image.

4. A method for monitoring pollutant flow in upstream and downstream of a river according to claim 3, characterized in that: The S43 specifically includes the following steps: S431: Using the pollutant source image as the monitoring image corresponding to the first monitoring site in the pollutant flow; S432: Locate the first site image set corresponding to the pollutant source image in the current monitoring image set, and use the first site image set to traverse each first site image set downward, and use the screenshot time corresponding to the pollutant source image as the starting time of the pollutant flow, and then filter down a first river image in turn according to the starting time of the pollutant flow, so that the pollutant source image is sequentially connected to the filtered multiple first river images arranged in chronological order from early to late, to form a pollutant flow.

5. The method for monitoring pollutant flow in the upstream and downstream of a river according to claim 3, characterized in that: The S44 specifically includes the following steps: S441: Using the pollutant source image as the monitoring image corresponding to the first monitoring site in the pollutant flow; S442: locate the first site image set corresponding to the pollutant source image in the current monitoring image set, traverse each first site image set downward with the first site image set, take the screenshot time corresponding to the pollutant source image as the starting time of the pollutant flow, and then identify the pollutants for the first site images in each first site image set in turn according to the starting time of the pollutant flow, and determine in turn whether the first site images in each first site image set contain the pollutants identified in the pollutant source image, if yes, go to step S443, if not, go to step S444; S443: extracting the first site image corresponding to the first site image set as a pollutant image, and arranging the pollutant image from upstream to downstream according to the river sequence and adding it to the pollutant flow; S444: Obtain the screenshot time of the last pollutant image arranged in the current pollutant flow, extract the second camera video to be identified in the monitoring site corresponding to the first site image set by the screenshot time, screenshot the second camera video to be identified by the third screenshot frequency to obtain the third river image, and then identify pollutants in the third river image in turn, and determine whether the third river image contains pollutants identified in the pollutant source image in turn. If so, use the third river image as the pollutant image, and arrange the pollutant images from upstream to downstream in river order and add them to the pollutant flow, wherein the third screenshot frequency is greater than the first screenshot frequency.

6. A method for monitoring pollutant flow in upstream and downstream of a river according to claim 3, characterized in that: The pollutant flow is composed of monitoring images corresponding to monitoring stations arranged from upstream to downstream in river order, and the monitoring images of the pollutant flow are also arranged from early to late in chronological order. A monitoring image of the pollutant flow is composed of a pollutant source image and a first river image or of a pollutant source image and a pollutant image.

7. A method for monitoring pollutant flow in the upstream and downstream of a river according to claim 5, characterized in that: The second camera video to be identified is the river surface within a preset time period obtained by photographing the monitoring site through its own river camera starting from the screenshot time of the last arranged pollutant image, and the river camera is a camera fixedly used by the monitoring site to photograph the river surface.

8. The method for monitoring pollutant flow in the upstream and downstream of a river according to claim 1, characterized in that: In S1, a current monitoring image set arranged from upstream to downstream in river order is obtained, and each first site image set in the monitoring image set undergoes image preprocessing, which includes: preprocessing the image obtained by screenshot, removing abnormal points in the image, and compressing the image into a single-channel grayscale image, and using the Laplace method to perform convolution scoring on the image. When the score reaches a preset value, the image is determined to be valid, and the image is used as the first river image and added to the first site image set at the screenshot time.

9. A pollutant flow monitoring system based on the upstream and downstream of a river, characterized in that: include: A monitoring image set acquisition module: used to obtain a current monitoring image set arranged from upstream to downstream in river sequence, wherein the monitoring image set includes a plurality of first site image sets corresponding to monitoring sites, wherein the first site image set is composed of a plurality of first river images arranged from early to late in chronological order, and the first river image is obtained by the monitoring site by taking a screenshot of a first camera video to be identified at a first screenshot frequency; The first river image judgment module is used to traverse the monitoring image set, identify pollutants in the first river image in turn, and when pollutants are identified in the first river image, stop traversal and mark the first river image as a pollutant source image; A second river image acquisition module: used to extract a first camera video to be identified of a corresponding monitoring site according to the pollutant source image, and to screenshot the first camera video to be identified at a second screenshot frequency to obtain a plurality of second river images arranged in chronological order from morning to night, wherein the second screenshot frequency is greater than the first screenshot frequency; Pollutant flow calculation module: used to locate multiple second river images within the time range according to the time range corresponding to the pollutant source image, and sequentially identify pollutants in the multiple second river images within the time range, and determine and obtain the pollutant flow corresponding to the pollutants identified in the pollutant source image according to the pollutant identification results; Pollutant flow display module: used to obtain monitoring images corresponding to monitoring sites arranged from upstream to downstream in river sequence according to pollutant flow, and to display the monitoring images arranged from upstream to downstream in river sequence.

10. A pollutant flow monitoring system based on upstream and downstream of a river according to claim 9, characterized in that: The pollutant flow calculation module specifically includes the following modules: Time extension module: used to extract the screenshot time corresponding to the pollutant source image, and expand the preset time period with the screenshot time to obtain the time range corresponding to the pollutant source image; The second river image judgment module is used to locate multiple second river images within a time range, and to identify pollutants in the second river images in chronological order from early to late. When pollutants are identified in the second river images, the total number of times the pollutants are identified is recorded, and it is determined whether the total number meets the condition of continuous occurrence and exceeds a preset threshold; A first pollutant flow generation module: used to generate a pollutant flow corresponding to the pollutant identified in the pollutant source image according to the current monitoring image set and the pollutant source image; The second pollutant flow generation module is used to locate the first site image set corresponding to the pollutant source image in the current monitoring image set, traverse each first site image set downward with the first site image set, and obtain the pollutant image corresponding to each first site image set through the screenshot time corresponding to the pollutant source image, and generate the pollutant flow corresponding to the pollutant identified in the pollutant source image according to the pollutant source image and the pollutant image.

Images