A statistical method for fishery resources in a marine ranch based on image recognition technology

Abstract

The present invention relates to the technical field of statistical techniques for fishery resources in marine pastures, and particularly to a method for statistically calculating fishery resources in marine pastures based on image recognition technology. The present invention conducts bathymetric topographic surveys through multi-beam and constructs an underwater three-dimensional topographic model, calculates the amount of artificial reefs placed according to the underwater three-dimensional topographic model and the bathymetric topographic survey water depth map; then selects the edge of the demonstration area and artificial reefs, and uses image recognition technology to statistically calculate the fishery resources in the marine pasture per unit reef body, completing the statistical calculation of fishery resources in the marine pasture based on image recognition technology; the present invention makes an exploratory attempt to use image recognition technology to measure the changes in fishery resources in marine pastures and the evaluation method, obtains the changes in the resources of nekton and demersal eggs in the marine pasture, and provides data support for the construction of marine pastures.

Description

Technical Field

[0001] The present invention relates to the technical field of statistical techniques for fishery resources in marine ranches, and particularly to a method for statistically calculating fishery resources in a marine ranch based on image recognition technology. Background Art

[0002] Countries around the world attach great importance to the protection and conservation of water area resources. Most coastal areas adopt methods such as building marine ranches to protect the ecological environment and increase fishery resources. A "marine ranch" refers to a certain sea area where large-scale fishery facilities and systematic management systems are adopted, and the natural marine ecological environment is utilized to gather artificially released economic marine organisms. Just like grazing cattle and sheep on land, planned and purposeful sea ranching of marine resources such as fish, shrimps, shellfish, and algae is carried out.

[0003] The main purposes of building marine ranches are twofold: firstly, to increase the yield of certain economic varieties or the fish yield of the entire sea area to ensure the stable and continuous growth of fishery resources; secondly, while utilizing marine resources, focus on protecting the marine ecosystem to achieve sustainable ecological fishery.

[0004] Currently, when evaluating the effects of marine ranches, it mainly focuses on on-site investigations of the sea area environment and comparative analysis of the community structures of organisms and fishery resources. However, this comparative analysis generally only involves data comparison between individual indicators. Measuring and evaluating the changes in fishery resources in a marine ranch is an important evaluation method for a marine ranch; currently, there is no method for measuring and evaluating the changes in fishery resources in a marine ranch. Summary of the Invention

[0005] The technical problem to be solved by the present invention is how to complete the measurement and evaluation of the changes in fishery resources in a marine ranch.

[0006] To solve the above problems, the present invention provides a method for statistically calculating fishery resources in a marine ranch based on image recognition technology. Through multi-beam bathymetric surveys, an underwater three-dimensional terrain model is constructed, and the artificial reef placement quantity is calculated according to the underwater three-dimensional terrain model and the multi-beam bathymetric survey depth chart; then, the edge of the demonstration area and artificial reefs are selected, and image recognition technology is used to statistically calculate the fishery resources per unit reef body in the marine ranch, thus completing the statistical calculation of fishery resources in the marine ranch based on image recognition technology;

[0007] It includes the following steps:

[0008] (1) Construct an underwater three-dimensional terrain model;

[0009] (2) Draw a multi-beam bathymetric survey depth chart;

[0010] (3) Calculate the artificial reef placement quantity;

[0011] (4) Set up a demonstration area and conduct video collection of fishery resources in the marine ranch within the demonstration area. Calculate the improvement rate of fishery resources in the marine ranch by the frequency of occurrence of fishery resources in the video collection.

[0012] Furthermore, step (3) for the amount of artificial reefs placed includes two parts; one part is the measured volume of fishing reefs on the upper part of the seabed surveyed by a multibeam sounding system; the other part is the estimated volume of settled fishing reefs based on the ground settlement and distribution area caused by the placement of fishing reefs; the sum of the two parts is the total volume of fishing reefs placed.

[0013] Furthermore, in the said step (2), the multibeam bathymetric chart specifically is, according to the requirements of the "Code for Hydrographic Survey of Water Transport Engineering" JTS131 - 2012, the water depth data is based on the theoretical lowest tide level, the mapping uses the CGCS2000 coordinate system, and the drawing is compiled using "AUTOCAD2004".

[0014] Furthermore, the fishery resources in the marine ranch include the nekton and the amount of demersal eggs in the marine ranch.

[0015] Furthermore, the video collection method in step (4) is as follows:

[0016] (1) Set up survey stations within the demonstration area and control stations outside the demonstration area;

[0017] (2) Take video photos and screenshots of the fishery resources in the artificial reef areas of the survey stations within the demonstration area and the control stations outside the demonstration area;

[0018] (3) Count the frequency of occurrence of fishery resources based on the video screenshots;

[0019] (4) Calculate the frequency of occurrence of fishery resources in different years based on the frequency of occurrence in step (3), and calculate the improvement rate of fishery resources in the marine ranch.

[0020] Furthermore, the calculation method for the improvement ratio of the amount of demersal eggs is as follows:

[0021] (1) Select 5 sites where the edge of the demonstration area intersects with the diagonal, and respectively select 1 artificial reef with a complete individual for fixed-point observation;

[0022] (2) Use the underwater videography method to collect the number of egg masses per unit reef body by video, and at the same time bring the egg masses attached to shellfish - like marine organisms back to the laboratory for rinsing and counting;

[0023] (3) Count the number of fish egg masses per unit reef body and the number of fish eggs per unit reef body, so as to estimate the improvement ratio of fish egg biological resources in the demonstration area in two years.

[0024] The beneficial effects of the present invention are as follows: The present invention conducts bathymetric topographic surveys through multi-beams and constructs an underwater three-dimensional topographic model, and calculates the amount of artificial reefs to be placed based on the underwater three-dimensional topographic model and the bathymetric map of the multi-beam topographic survey; then, the edge of the demonstration area and artificial reefs are selected, and image recognition technology is used to count the fishery resources in the unit reef body of the marine ranch, completing the statistics of the fishery resources in the marine ranch based on image recognition technology; the present invention explores and attempts to use image recognition technology to measure the changes in fishery resources in the marine ranch and the evaluation method, and obtains the changes in the resources of nekton and demersal eggs in the marine ranch, providing data support for the construction of the marine ranch. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0026] Figure 1 It is a local 2D effect diagram near the artificial reef body;

[0027] Figure 2 It is a local 3D effect diagram near the artificial reef body;

[0028] Figure 3 It is a 2D effect diagram of the distribution of artificial reef bodies in the entire survey area;

[0029] Figure 4 It is a local bathymetric map near the artificial reef body;

[0030] Figure 5 It is the distribution of survey stations;

[0031] Figure 6 It is a screenshot of the underwater video of larvae in the demonstration area in 2020;

[0032] Figure 7 It is a screenshot of the underwater video of larvae in the demonstration area in 2020;

[0033] Figure 8 It is a screenshot of the underwater video of larvae in the demonstration area in 2020;

[0034] Figure 9 It is a screenshot of the underwater video of larvae in the demonstration area in 2020;

[0035] Figure 10 It is a screenshot of the underwater video of larvae in the demonstration area in 2020;

[0036] Figure 11 It is a screenshot of the underwater video of larvae in the demonstration area in 2020;

[0037] Figure 12 Screenshot of the larval fish underwater video 1 in the demonstration area in 2021;

[0038] Figure 13 Screenshot of the larval fish underwater video 2 in the demonstration area in 2021;

[0039] Figure 14 Screenshot of the larval fish underwater video 3 in the demonstration area in 2021;

[0040] Figure 15 Screenshot of the larval fish underwater video 4 in the demonstration area in 2021;

[0041] Figure 16 Screenshot of the larval fish underwater video 5 in the demonstration area in 2021;

[0042] Figure 17 Screenshot of the larval fish underwater video 6 in the demonstration area in 2021;

[0043] Figure 18 Distribution of fish egg sampling sites in the demonstration area;

[0044] Figure 19 Screenshot of the fish egg underwater video in the demonstration area in 2020;

[0045] Figure 20 Fish eggs collected underwater in the demonstration area in 2020;

[0046] Figure 21 Fish egg count in the demonstration area in 2020;

[0047] Figure 22 Screenshot of the fish egg underwater video in the demonstration area in 2021;

[0048] Figure 23 Fish eggs collected underwater in the demonstration area in 2021;

[0049] Figure 24 Fish egg count in the demonstration area in 2021. Detailed implementation manner

[0050] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0051] Taking the demonstration area located in Jiayi Marine Ranch in Changdao, Yantai as an example, Shandong Zhengyuan Digital City Construction Co., Ltd. undertook the assessment work after the reef throwing in the artificial reef area of Changdao Jiayi Seafood Development Co., Ltd.

[0052] Based on the image recognition technology, the statistical method of fishery resources in the marine ranch is as follows: Through multi-beam bathymetric survey to construct an underwater three-dimensional terrain model, calculate the amount of reef throwing according to the underwater three-dimensional terrain model and the multi-beam bathymetric survey depth chart; then select the edge of the demonstration area and artificial reefs, and use image recognition technology to count the fishery resources in the unit reef body of the marine ranch, and complete the statistical work of fishery resources in the marine ranch based on image recognition technology.

[0053] It includes the following steps:

[0054] (1) Construct an underwater three-dimensional terrain model;

[0055] (2) Draw a multi-beam bathymetric survey depth chart;

[0056] (3) Calculate the amount of artificial reef throwing;

[0057] (4) Set up a demonstration area and collect videos of fishery resources in the demonstration area of the marine ranch, and calculate the improvement rate of fishery resources in the marine ranch through the frequency of fishery resources appearing in the video collection.

[0058] Specifically, in the implementation, the construction method of the underwater three-dimensional terrain model in step (1) is as follows:

[0059] After the multi-beam original data is edited in line mode and SUBSET sub-region editing, and the interference signals are removed, the TPU value of each survey line is used to generate an underwater three-dimensional terrain model. As shown in Figure 1 、 Figure 2 、 Figure 3 It can be seen from the underwater three-dimensional terrain model that a large number of artificial reefs have been thrown in the survey area, and the artificial reefs thrown are distributed irregularly on the seabed surface, mainly concentrated in the southwestern area of the survey area.

[0060] The drawing of the multi-beam bathymetric survey depth chart in step (2)

[0061] According to the requirements of the "Code for Hydrographic Survey of Water Transport Engineering" JTS131-2012, the bathymetric data is based on the theoretical lowest tide level, the mapping uses the CGCS2000 coordinate system, and the drawing is compiled using "AUTOCAD2004". The title of the drawing is "Underwater Multi-beam Bathymetric Survey Depth Chart of the Artificial Reef Area of the Demonstration Project of Jiayi Marine Ranch in the Eastern Sea Area of the Miaodao Archipelago" (1:1000). The local bathymetric chart near the artificial fish reef body is as shown in Figure 4 shown.

[0062] The method for calculating the amount of artificial reef throwing in step (3):

[0063] The artificial fish reef deployment volume consists of two parts; one part is the actual measured volume of fish reefs in the upper part of the seabed surveyed by the multibeam sounding system; the other part is the estimated volume of sunken fish reefs based on the ground settlement and distribution area caused by the deployment of fish reefs; the sum of the two parts is the total deployed fish reef volume.

[0064] Set up demonstration areas and collect videos of the fishery resources in the marine ranch within the demonstration areas. Calculate the improvement rate of the fishery resources in the marine ranch through the frequency of the appearance of the fishery resources in the video collection. The fishery resources in the marine ranch include the nekton and the amount of demersal eggs in the marine ranch.

[0065] (1) During implementation, the calculation method of nekton

[0066] A total of 8 survey stations were set up within the demonstration areas and 4 control stations outside the demonstration areas. The distribution map and coordinates of the survey stations and control stations are shown in Figure 5 and Table 1.

[0067] Table 1 Station coordinates

[0068]

[0069]

[0070] Video collections of the distribution areas of larvae in the artificial fish reef area were carried out in the autumn of 2020 and 2021 respectively. A total of 6 videos were randomly selected for each of the years 2020 and 2021. For each video, 1 minute of video material was randomly selected, and a video screenshot was taken every 10 seconds. Statistical data on the appearance frequency of the target fish Sebastes schlegelii (an important economic fish) were counted based on the video screenshots, and thus the improvement ratio of fishery resources was calculated. The screenshots of the video materials in 2020 are as follows Figure 1 , and the statistical results are shown in Table 2 and Table 3. Figures 6 to 11

[0071] Table 2 Frequency of larvae of economic fish counted using image recognition technology in 2020

[0072]

[0073]

[0074] Table 3 Frequency of larvae of economic fish counted using image recognition technology in 2021

[0075]

[0076] Based on the above, using image recognition technology, the appearance frequency of larvae of important economic fish was 14.00 individuals / minute in 2020 and 20.67 individuals / minute in 2021.

[0077] Compared with the autumn of 2020, the fishery resources of larvae of important economic fish species increased by 47.62% in the autumn of 2021.

[0078] (2) In the implementation, the calculation method of demersal fish eggs

[0079] Select 5 sites where the edge of the demonstration area intersects with the diagonal line, such as Figure 18 , and select 1 artificial fish reef with intact individuals respectively for fixed-point observation. The underwater camera method is used to collect the number of egg masses per reef body in the video, and the egg masses attached to oysters are taken back to the laboratory for rinsing and counting. Video collection and sample collection are carried out in the autumn of 2020 and 2021 respectively. The number of fish egg masses per reef body and the number of fish eggs per reef body are counted, so as to estimate the improvement ratio of fish egg biological resources in the demonstration area in two years.

[0080] The screenshot of video collection in 2020 is shown in Figure 19 , the egg masses collected back to the laboratory are shown in Figure 20 , the rinsing, weighing and counting are shown in Figure 21 , and the data statistics are shown in Tables 4 to 5.

[0081] Table 4 Number of fish egg masses collected in the video in 2020

[0082]

[0083] Table 5 Statistics of the number of fish eggs collected underwater in 2020

[0084] The screenshot of video collection in 2021 is shown in Figure 22 , the egg masses collected back to the laboratory are shown in Figure 23 , the rinsing, weighing and counting are shown in Figure 24 , and the data statistics are shown in Tables 6 to 7.

[0085] Table 6 Number of fish egg masses collected in the video in 2021

[0086]

[0087] Table 7 Statistics of the number of fish eggs collected underwater in 2021

[0088] Item Reef Body 1 Reef Body 2 Reef Body 3 Reef Body 4 Reef Body 5 Average Number of Fish Eggs (pcs) 479 1355 1837 995 995 1132.2

[0089] According to the statistics of the collected data, in 2020, a total of 25.7 g of wet weight of fish eggs were collected, with a total of 4,287 eggs. The density of fish egg masses per reef body was 3.2 masses / reef body, and the number of fish eggs was 857.4 eggs / reef body. In 2021, a total of 31.4 g of wet weight of fish eggs were collected, with a total of 5,661 eggs. The density of fish egg masses per reef body was 4.6 masses / reef body, and the number of fish eggs was 1,132.2 eggs / reef body. Compared with 2020, the density of fish egg masses of important economic fish in the demonstration area in 2021 increased by 43.8%; calculated according to the total mass of the wet weight of fish eggs, the fish egg biomass in 2021 increased by 22.2% compared with 2020; calculated according to the number of fish eggs, the fish egg biomass in 2021 increased by 32.1% compared with 2020.

[0090] Based on the above, bathymetric topographic surveys are carried out by multi-beam to construct an underwater three-dimensional topographic model, and the amount of fish reefs to be placed is calculated according to the underwater three-dimensional topographic model and the bathymetric map of multi-beam topographic surveys; then the edge of the demonstration area and artificial fish reefs are selected, and image recognition technology is used to count the fishery resources per reef body, completing the statistics of the fishery resources in the marine ranch based on image recognition technology; finally, data support is provided for the development of the marine ranch.

Claims

1. A method for calculating the fishery resource volume in a marine ranch based on image recognition technology, characterized in that, bathymetric terrain measurement is carried out by multi-beam to construct an underwater three-dimensional terrain model, and the amount of artificial reefs placed is calculated according to the underwater three-dimensional terrain model and the multi-beam bathymetric terrain measurement depth map; then the edge of the demonstration area and artificial reefs are selected, and image recognition technology is used to count the fishery resource volume of the marine ranch per reef body, completing the statistics of the fishery resource volume in the marine ranch based on image recognition technology; including the following steps: (1) Construct an underwater three-dimensional terrain model; (2) Draw a multi-beam bathymetric terrain measurement depth map; (3) Calculate the amount of artificial reefs placed; (4) Set up a demonstration area and collect videos of the fishery resources in the marine ranch in the demonstration area. The improvement rate of the fishery resources in the marine ranch is calculated by the frequency of occurrence of nekton in the fishery resources in the marine ranch and the quantity of demersal egg resources collected through the video; The calculation method of the improvement ratio of demersal egg resources is: (1) Select 5 sites where the edge of the demonstration area intersects the diagonal line, and select 1 artificial reef with a complete individual respectively for fixed-point observation; (2) Adopt the underwater camera method, use the video to collect the number of egg masses per reef body, and at the same time bring the egg masses attached to shellfish into the laboratory for rinsing and counting; (3) Count the number of fish egg masses per reef body and the number of fish eggs per reef body, so as to estimate the improvement ratio of fish egg biological resources in the demonstration area in two years.

2. The method for calculating the fishery resource volume in a marine ranch based on image recognition technology according to claim 1, characterized in that, step (3) The amount of artificial reefs placed includes two parts; one part is the measured reef volume on the upper part of the seabed investigated by the multi-beam sounding system; the other part is the estimated settlement reef volume based on the ground settlement and distribution area caused by the placement of reefs; the sum of the two parts is the total amount of reefs placed.

3. The method for calculating the fishery resource volume in a marine ranch based on image recognition technology according to claim 1, characterized in that, In step (2), the multi-beam bathymetric terrain measurement depth map is specifically prepared according to the requirements of the "Code for Hydrographic Survey of Water Transport Engineering" JTS131-2012. The water depth data is based on the theoretical lowest tide level, the mapping uses the CGCS2000 coordinate system, and AUTOCAD 2004 is used for compilation and drawing.

4. The method for calculating the fishery resource volume in a marine ranch based on image recognition technology according to claim 1, characterized in that, The calculation method of the improvement rate of nekton in the marine ranch is: S1: Set up survey stations in the demonstration area and control stations outside the demonstration area; S2: Take videos and screenshots of the fishery resources in the artificial reef areas of the survey stations in the demonstration area and the control stations outside the demonstration area; S3: Count the frequency of occurrence of the fishery resources in the marine ranch according to the video screenshots; S4: Calculate the frequency of occurrence of the fishery resources in the marine ranch in different years according to the frequency of occurrence in step S3, and calculate the improvement rate of the fishery resources in the marine ranch.

Images