Method for identifying and predicting fishing vessel dangers based on fusion data

Abstract

The present invention relates to a method for identifying and predicting fishing boat danger based on fused data, which belongs to the field of image data processing and geographic information system, and comprises the following steps: S1. acquiring fishing area grid forecast data and fishing boat position data; S2. generating a fishing area operation visualization image according to the acquired data; S3. detecting fishing boats in dangerous fishing areas based on fishing area image processing and analysis; S4. acquiring a fishing boat list, triggering an early warning system and notifying management personnel; the present invention has the advantages of generating a fishing area operation dynamic image by fusing fishing area, forecast, fishing boat position and operation condition data, efficiently retrieving vessels in dangerous fishing areas based on image detection technology, avoiding freeze and delay problems caused by large quantities of fishing boats and fishing area space query traversal, converting GIS spatial analysis problems into image processing problems, greatly improving recognition efficiency, and timely triggering an early warning system.

Description

Technical Field

[0001] The invention relates to a method for identifying and predicting fishing boat danger based on fused data, and belongs to the field of image data processing and geographic information system. Background Art

[0002] Marine fishery production has always been a high-risk industry. How to accurately dispatch fishing boats in strong wind warning weather is a key link in ensuring safe production of fisheries. Accurate marine meteorological forecasts can accurately command and dispatch fishing boats operating at sea to leave dangerous sea areas with high wind levels, without all returning to port to avoid the wind. Fishermen have significantly increased the time and sea area for sea operations.

[0003] As my country's marine forecasting capabilities become increasingly higher in temporal and spatial resolution, it provides the possibility for refined scheduling of marine fishing vessels in windy weather. The present invention mainly analyzes dangerous chart areas efficiently based on refined meteorological grid forecast data, annotates charts and identifies dangerous ships, thereby improving the accuracy of fishing vessel scheduling in windy weather. Summary of the invention

[0004] In order to overcome the defects of the existing technology of dangerous fishing boat operations in windy weather, such as insufficient automation, low accuracy, limited real-time performance, great influence by human factors, and low spatial retrieval efficiency, the present invention provides a method for identifying and predicting fishing boat dangers based on fused data. The technical solution of the present invention is:

[0005] A method for identifying and predicting fishing vessel danger based on fused data comprises the following steps:

[0006] S1. Obtaining grid forecast data of fishing areas and fishing vessel location data;

[0007] S2. Generate a visual image of fishing area operations based on the acquired data;

[0008] S3. Detect fishing boats in dangerous fishing areas based on fishing area image processing and analysis;

[0009] S4. Obtain a list of fishing vessels, trigger the early warning system and notify management personnel;

[0010] The step S2 is specifically as follows:

[0011] (2-1) Based on Leaflet and Vue development, first load and render the fishing area data in geojson format to generate a fishing area grid layer; load the real-time location of the fishing boat and display it as a triangle symbol layer; Leaflet provides the GeoJSON layer function to load and render geographic information data, including fishing area boundaries and dangerous area data;

[0012] Use the L.geoJSON method to render the loaded fishing area data into a Leaflet layer;<l-marker>and <l-icon>The component is used to display the triangular symbol of the fishing boat's location, which is rendered in green. The location of the boat is dynamically displayed by looping through the fishing boat array. In order to integrate the Leaflet map component in the Vue application, the VueLeaflet library is used. The VueLeaflet library provides a Vue component that enables the use of Leaflet map functions in Vue projects.

[0013] (2-2) Label the fishing area grid layer and use Leaflet's style function to define the style of each geographic feature. If the wind speed is greater than 10.8m / s, it will be marked as blue; if the wind speed is greater than 17.2m / s, it will be marked as yellow; if the wind speed is greater than 24.5m / s, it will be marked as orange; if the wind speed is greater than 32.6m / s, it will be marked as red. The above colored fishing areas are dangerous fishing areas. At this point, users can intuitively see the real-time distribution of dangerous fishing areas and whether fishing boats are in dangerous fishing areas.

[0014] (2-3) Overlay the layers of step (2-2) to generate and export a new image; at the same time, generate matching image coordinates corresponding to the fishing boat coordinates, use the Leaflet-image plug-in to render the Leaflet map into a Canvas, and then export the Canvas as an image file; the L.imageOverlay method is used to export the map into a Canvas, and accepts a callback function to process the export result; convert the Canvas into base64-encoded PNG image data;

[0015] Use the latLngToContainerPoint method to convert geographic coordinates to container pixel coordinates, that is, pixel coordinates in the map view, and store them in the geo_coordinates list;

[0016] The specific formula for converting the geographic coordinates of a fishing vessel into image coordinates is:

[0017]

[0018] Among them, (x image ,y image ) represents the coordinates in the image, φ and λ represent the latitude and longitude of the fishing boat, respectively.

[0019] In the step S1, the weather forecast data of the target fishing area is the data in the Shandong Province Marine Meteorological Zoning Forecast Service Platform, specifically: obtaining the Shandong Province fishing area zoning forecast grid data from the platform according to the httpPost security interface method, with the reporting time being 8:00 a.m. every day, supporting forecasts for the next 11 days, including 3 hours, 6 hours, 12 hours, and 24 hours intervals, and the element selected is gusts; then accessing the real-time AIS coordinate data of the fishing vessel from the domestic ship information network platform through the TCP encryption interface.

[0020] The step S3 is specifically as follows:

[0021] (3-1) Image color segmentation: Based on the blue, yellow, orange, and red color markings of dangerous fishing areas, color threshold segmentation is performed to separate dangerous fishing areas from the background. First, use cv2.imread to load the map image data generated by Leaflet from a file or memory. In color_thresholds, define the upper and lower bounds of the colors red, orange, yellow, and blue for subsequent color threshold segmentation. Convert the image to HSV or other color spaces to better segment the colors. Create a blank image to store the segmentation results. Perform pixel-level traversal and processing on the loaded image data. Check the color of the current pixel to determine whether it is within the defined color threshold range.

[0022] Use the cv2.inRange() function to create a color threshold mask to filter out pixels that meet the conditions; copy the pixels segmented by the color threshold to the result image to form a segmented image; then set the pixels of the dangerous fishing area to 255 and the background to 0, and use it as a mask for the original image to perform mask processing and only keep the dangerous fishing area;

[0023] (3-2) Fishing boat shape detection: perform contour detection on each color area, filter out contours that may be triangular symbols, apply shape matching algorithm to the filtered contours, compare them with the predefined triangle template, and determine whether they meet the shape characteristics of the triangle symbol; use the image processing library OpenCV to load the image generated by the mask, and perform color threshold segmentation based on the green color of the fishing boat symbol to extract the binary image of the green area, that is, the black and white image, and use OpenCV's contour detection function findContours() to find all contours in the binary image; apply shape feature descriptors to each detected contour for filtering, use the contour approximation algorithm cv2.approxPolyDP for matching, and further use the shape template matching algorithm cv2.matchShapes to compare the filtered contours with the predefined triangle template to determine whether they meet the shape characteristics of the triangle symbol;

[0024] (3-3) Position output: For the successfully matched triangle symbol, output its center point coordinates as the position of the fishing boat, and store these center point coordinates in the triangle_coordinates list, and then match the fishing boat code according to the image coordinates;

[0025] The specific formula for outputting the center coordinates of the successfully matched triangle symbol as the location of the fishing boat is:

[0026]

[0027] in, Represents the coordinates of the center point of the successfully matched contour, and N represents the number of vertices on the contour.

[0028] The step S4 is specifically as follows: forming a list of fishing boats for gale dispatch based on the screened fishing boat information, sending warning information with one click, dispatching operations in gale weather, and informing the fishing boats to return to the port or anchor in time.

[0029] The advantages of the present invention are: by integrating fishing areas, forecasts, fishing boat locations, and operating condition data, dynamic images of fishing area operations are generated, and vessels in dangerous fishing areas are efficiently retrieved based on image detection technology, thereby avoiding the jamming and delay problems caused by large-scale fishing boats and fishing area spatial query traversal, and converting GIS spatial analysis problems into image processing problems, which can greatly improve recognition efficiency and trigger the early warning system in time.

[0030] Image processing technology is used to automatically identify dangerous fishing areas and vessels, screen the list of vessels to be dispatched, and issue early warning information. It has the advantages of automation, speed, and accuracy. This method can be widely used in fishing vessel dispatch systems in coastal areas to improve monitoring efficiency and accuracy and provide strong support for fishery disasters. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0032] The present invention will be further described below in conjunction with specific embodiments, and the advantages and features of the present invention will become clearer as the description proceeds. However, these embodiments are exemplary only and do not constitute any limitation to the scope of the present invention. It should be understood by those skilled in the art that the details and forms of the technical solution of the present invention may be modified or replaced without departing from the spirit and scope of the present invention, but these modifications and replacements all fall within the scope of protection of the present invention.

[0033] See also Figure 1 The present invention relates to a method for identifying and predicting fishing vessel dangers based on fused data, comprising the following steps:

[0034] S1. Obtain fishing area grid forecast data and fishing vessel location data; S2. Generate a visual image of fishing area operations based on the acquired data; S3. Detect fishing vessels in dangerous fishing areas based on fishing area image processing and analysis; S4. Obtain a list of fishing vessels, trigger the early warning system and notify management personnel.

[0035] In the step S1, the weather forecast data of the target fishing area is the data in the Shandong Province Marine Meteorological Zoning Forecast Service Platform, specifically: obtaining the Shandong Province fishing area zoning forecast grid data from the platform according to the httpPost security interface method, with the reporting time being 8:00 a.m. every day, supporting forecasts for the next 11 days, including 3 hours, 6 hours, 12 hours, and 24 hours intervals, and the element selected is gusts; then accessing the real-time AIS coordinate data of the fishing vessel from the domestic ship information network platform through the TCP encryption interface.

[0036] D weather = {W(t i ), G(t i )|ti∈{3h,6h,12h,24h}};

[0037] Among them, D weather represents the meteorological data set, W(t i ) and G(t i ) indicates t i wind speed and direction;

[0038] D AIS ={(φ j ,λ j , t j )|j∈ships};

[0039] D AIS represents the fishing boat location dataset, φ j and λ j denote the latitude and longitude of the j-th fishing boat, t j Indicates a timestamp.

[0040] The step S2 is specifically as follows:

[0041] (2-1) Based on Leaflet and Vue development, first load and render the fishing area data in geojson format to generate a fishing area grid layer; load the real-time location of the fishing boat and display it as a triangle symbol layer. Leaflet provides a powerful GeoJSON layer function that can load and render geographic information data, including fishing area boundaries and dangerous area data. Use the L.geoJSON (data, options) method to render the loaded fishing area data into a Leaflet layer; <l-marker>and <l-icon>The component is used to display the triangular symbol of the fishing boat position, which is rendered in green. Multiple ship positions are dynamically displayed by looping through the fishing boat array. In order to integrate the Leaflet map component in the Vue application, the VueLeaflet library is used. The VueLeaflet library provides a Vue component that enables the use of Leaflet map functions in Vue projects.

[0042] (2-2) Label the fishing area grid layer and use Leaflet's style function to define the style of each geographic feature, such as fill color, border color, and transparency. If the wind speed is greater than 10.8m / s, it will be marked as blue; if the wind speed is greater than 17.2m / s, it will be marked as yellow; if the wind speed is greater than 24.5m / s, it will be marked as orange; if the wind speed is greater than 32.6m / s, it will be marked as red. The above colored fishing areas are dangerous fishing areas. At this point, users can intuitively see the real-time distribution of dangerous fishing areas and whether fishing boats are in dangerous fishing areas.

[0043] (2-3) Overlay the layers in step (2-2) to generate and export a new image; at the same time, generate matching image coordinates corresponding to the fishing boat coordinates. Use the leaflet-image plug-in to render the Leaflet map into a Canvas, and then export the Canvas as an image file;

[0044] The L.imageOverlay method is used to export the map to Canvas and accepts a callback function to process the export result; canvas.toDataURL('image / png') converts Canvas into base64-encoded PNG image data.

[0045] Use the latLngToContainerPoint method to convert geographic coordinates to pixel coordinates in the map view and store them in the geo_coordinates list;

[0046] Convert the fishing boat geographic coordinates to image coordinates:

[0047]

[0048] Among them, (x image ,y image ) represents the coordinates in the image, φ and λ represent the latitude and longitude of the fishing boat, respectively.

[0049] The step S3 is specifically as follows:

[0050] (3-1) Image color segmentation: According to the color mark of the dangerous fishing area (blue, yellow, orange, red), color threshold segmentation is performed to separate the dangerous fishing area from the background. First, use cv2.imread to load the map image data generated by Leaflet from the file or memory, define the upper and lower bounds of each color (red, orange, yellow, blue) in color_thresholds for subsequent color threshold segmentation, convert the image to HSV or other color space to better segment the color, create a blank image to store the segmented results, traverse and process the loaded image data at the pixel level, check the color of the current pixel to determine whether it is within the defined color threshold range. Use the cv2.inRange() function to create a color threshold mask to filter out pixels that meet the conditions. Copy the pixels segmented by the color threshold to the result image to form a segmented image. Then set the pixels of the dangerous fishing area to 255 and the background to 0, use them as the mask of the original image, and mask the original image to keep only the dangerous fishing area.

[0051]

[0052] Among them, C(x,y) represents the color mark of the position (x,y), and V(x,y) represents the wind speed of the position (x,y).

[0053] (3-2) Fishing boat shape detection: contour detection is performed on each color area to screen out the contours that may be triangular symbols. The shape matching algorithm is applied to the screened contours and compared with the predefined triangle template to determine whether they meet the shape characteristics of the triangle symbol. The image generated by the mask is loaded using the image processing library OpenCV, and color threshold segmentation is performed based on the color (green) of the fishing boat symbol to extract the binary image (i.e., black and white image) of the green area, thereby further focusing on the number of shape matching targets. The contour detection function findContours() of OpenCV is used to find all contours in the binary image. The shape feature descriptor is applied to each detected contour for screening, and the contour approximation algorithm cv2.approxPolyDP is used for matching. The screened contours are further compared with the predefined triangle template using the shape template matching algorithm cv2.matchShapes to determine whether they meet the shape characteristics of the triangle symbol.

[0054] (3-3) Position output: For the successfully matched triangle symbol, output its center point coordinates as the location of the fishing boat, and store these center point coordinates in the triangle_coordinates list. Then, match the fishing boat code according to the image coordinates; output the center coordinates of the successfully matched triangle symbol as the location of the fishing boat:

[0055]

[0056] in, Represents the coordinates of the center point of the successfully matched contour, and N represents the number of vertices on the contour.

[0057] The step S4 is specifically as follows: forming a list of fishing boats for gale dispatch based on the screened fishing boat information, sending warning information with one click, dispatching operations in gale weather, and informing the fishing boats to return to the port or anchor in time.

[0058] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for identifying and predicting fishing vessel dangers based on fusion data, characterized in that: The following steps are involved: S1. Obtaining grid forecast data of fishing areas and fishing vessel location data; S2. Generate a visual image of fishing area operations based on the acquired data; S3. Detect fishing boats in dangerous fishing areas based on fishing area image processing and analysis; S4. Obtain a list of fishing vessels, trigger the early warning system and notify management personnel; The step S2 is specifically as follows: (2-1) Developed based on Leaflet and Vue, first load and render the fishing area data in geojson format to generate a fishing area grid layer; load the real-time location of the fishing boat and display it as a triangle symbol layer; Leaflet provides the GeoJSON layer function to load and render geographic information data, including fishing area boundaries and dangerous area data; Use the L.geoJSON method to render the loaded fishing area data into a Leaflet layer; <l-marker>and <l-icon> The component is used to display the triangular symbol of the fishing boat's location, which is rendered in green. The location of the boat is dynamically displayed by looping through the fishing boat array. In order to integrate the Leaflet map component in the Vue application, the VueLeaflet library is used. The VueLeaflet library provides a Vue component that enables the use of Leaflet map functions in Vue projects. (2-2) Label the fishing area grid layer and use Leaflet's style function to define the style of each geographic feature. If the wind speed is greater than 10.8m / s, it will be marked as blue; if the wind speed is greater than 17.2m / s, it will be marked as yellow; if the wind speed is greater than 24.5m / s, it will be marked as orange; if the wind speed is greater than 32.6m / s, it will be marked as red. The above colored fishing areas are dangerous fishing areas. At this point, users can intuitively see the real-time distribution of dangerous fishing areas and whether fishing boats are in dangerous fishing areas. (2-3) Overlay the layers of step (2-2) to generate and export a new image; at the same time, generate matching image coordinates corresponding to the fishing boat coordinates, use the Leaflet-image plug-in to render the Leaflet map into Canvas, and then export the Canvas as an image file; The L.imageOverlay method is used to export the map to Canvas and accept a callback function to process the export result; convert the Canvas to base64-encoded PNG image data; Use the latLngToContainerPoint method to convert the fishing boat geographic coordinates into pixel coordinates in the map view and store them in the geo_coordinates list; The specific formula for converting the geographic coordinates of a fishing vessel into image coordinates is: ； Among them, (x image， y image ) represents the coordinates in the image, φ and λ represent the latitude and longitude of the fishing boat, respectively.

2. The method for identifying and predicting fishing vessel danger based on fusion data according to claim 1 is characterized in that: In the step S1, the weather forecast data of the target fishing area is the data in the Shandong Province Marine Meteorological Zoning Forecast Service Platform, specifically: the Shandong Province fishing area zoning forecast grid data is obtained from the platform according to the httpPost security interface method, the starting time is 8 am every day, and the forecast for the next 11 days is supported, including 3 hours, 6 hours, 12 hours, and 24 hours. The element selected is gust, and then the real-time AIS coordinate data of the fishing vessel is accessed from the domestic ship information network platform through the TCP encryption interface.

3. The method for identifying and predicting fishing vessel danger based on fusion data according to claim 1 is characterized in that: The step S3 is specifically as follows: (3-1) Image color segmentation: According to the blue, yellow, orange, and red color markings of dangerous fishing areas, color threshold segmentation is performed to separate dangerous fishing areas from the background. First, use cv2.imread to load the map image data generated by Leaflet from a file or memory. In color_thresholds, define the upper and lower bounds of the colors red, orange, yellow, and blue for subsequent color threshold segmentation. Convert the image to HSV or other color spaces to better segment the colors. Create a blank image to store the segmentation results. Perform pixel-level traversal and processing on the loaded image data. Check the color of the current pixel to determine whether it is within the defined color threshold range. Use the cv2.inRange() function to create a color threshold mask to filter out pixels that meet the conditions; copy the pixels segmented by the color threshold to the result image to form a segmented image; then set the pixels of the dangerous fishing area to 255 and the background to 0, and use them as a mask for the original image to perform mask processing and only keep the dangerous fishing area; (3-2) Fishing boat shape detection: perform contour detection on each color area, filter out contours that may be triangular symbols, apply shape matching algorithm to the filtered contours, compare them with the predefined triangle template, and determine whether they meet the shape characteristics of the triangle symbol; use the image processing library OpenCV to load the image generated by the mask, and perform color threshold segmentation based on the green color of the fishing boat symbol to extract the binary image of the green area, that is, the black and white image, and use OpenCV's contour detection function findContours() to find all contours in the binary image; apply shape feature descriptors to each detected contour for filtering, use the contour approximation algorithm cv2.approxPolyDP for matching, and further use the shape template matching algorithm cv2.matchShapes to compare the filtered contours with the predefined triangle template to determine whether they meet the shape characteristics of the triangle symbol; (3-3) Position output: for the successfully matched triangle symbol, output its center point coordinates as the location of the fishing boat, and store these center point coordinates in the triangle_coordinates list, and then match the fishing boat code according to the image coordinates; The specific formula for outputting the center coordinates of the successfully matched triangle symbol as the location of the fishing boat is: ;in, Represents the coordinates of the center point of the successfully matched contour, and N represents the number of vertices on the contour.

4. The method for identifying and predicting fishing vessel danger based on fusion data according to claim 1 is characterized in that: The step S4 specifically comprises: forming a list of fishing boats for high wind dispatch according to the screened fishing boat information, sending early warning information with one key, dispatching operations in high wind weather, and informing the fishing boats to return to the port or anchor in time.

Images