A method, system and storage medium for monitoring and identifying fishing of marine economic species

Abstract

The application discloses a kind of marine economic species monitoring identification fishing method, system and storage medium, it is related to fishery fishing technical field, comprising: obtaining target sea area underwater environment monitoring video stream, according to the underwater environment monitoring video stream obtains underwater environment frame image information, and the underwater environment frame image information is preprocessed;Marine organism identification model is constructed, and the underwater environment frame image information after pre-processing is imported into the marine organism identification model to identify the marine economic species in image, and the main growth stage of marine economic species is obtained, the quantity information of marine economic species in target area is estimated, whether it can be fished according to the main growth stage and quantity information is judged, and fishing route and fishing plan are generated according to the judgment result.The present application realizes the intelligent detection and identification of marine economic species in target sea area, ensures the identification accuracy, and is beneficial to the accurate fishing of marine economic species.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fishery capture, and more particularly, to a method and system for monitoring and identifying marine economic species and a storage medium. BACKGROUND

[0002] With the increasing attention of the state to marine resources, using modern marine information technology to establish a modern marine industry system has become one of the important measures of the state. The monitoring and identification of marine economic species have important research significance and value in the fields of aquaculture management, water environment monitoring, and fishery resource research. However, due to the limitations of the special imaging environment underwater, the underwater image has quality degradation phenomena such as low contrast, color distortion, and uneven lighting, which leads to loss of image details and difficulty in automated analysis. In addition, fish has various shapes, sizes, and numerous species with subtle differences, so the identification and capture of specific marine economic species in fishery capture pose a serious challenge.

[0003] In order to achieve accurate capture of target marine economic species in a target sea area, a system needs to be developed to cooperate with it to achieve this. The system obtains underwater environment monitoring video streams of the target sea area, obtains underwater environment frame image information from the underwater environment monitoring video streams, and pre-processes the underwater environment frame image information. A marine organism recognition model is constructed, the pre-processed underwater environment frame image information is imported into the marine organism recognition model to identify marine economic species in the image, and the main growth stage of the marine economic species is obtained. The number of marine economic species in the target area is estimated, and it is determined whether the marine economic species can be captured according to the main growth stage and the number information. A capture route and a capture plan are generated according to the determination result. In the implementation process of the system, how to identify the target marine economic species is an urgent problem to be solved. SUMMARY

[0004] To solve at least one of the above technical problems, the present application provides a method and system for monitoring and identifying marine economic species and a readable storage medium.

[0005] The first aspect of the present application provides a method for monitoring and identifying marine economic species, comprising:

[0006] Obtaining underwater environment monitoring video streams of a target sea area, obtaining underwater environment frame image information from the underwater environment monitoring video streams, and pre-processing the underwater environment frame image information;

[0007] Constructing a marine organism recognition model, importing the pre-processed underwater environment frame image information into the marine organism recognition model, and obtaining the target marine economic species according to the recognition result of the marine organism model;

[0008] Acquire the main growth stage of the target marine economic species, and estimate the quantity information of the target marine economic species in the target area;

[0009] Determine whether fishing can be carried out according to the main growth stage and the quantity information, and generate a fishing route and a fishing plan according to the determination result.

[0010] In the scheme, the marine organism recognition model is constructed, specifically:

[0011] The marine organism recognition model is constructed based on the Mobilenet network and the YOLOv3 network and combined with fishery acoustic technology, different scale feature maps or feature data in underwater environment frame image information are extracted through the Mobilenet network and the fishery acoustic technology;

[0012] The different scale feature maps or feature data are fused to generate comprehensive feature maps or feature data, the comprehensive feature maps or feature data are imported into the YOLOv3 network to predict the prior box, and the predicted value is compared with the true value to calculate the loss function;

[0013] Marine organism images and label data are acquired through big data, and the marine organism recognition model is trained according to the minimized loss function;

[0014] The marine organism recognition model generates a labeled box with marine organism species information in underwater environment frame image information after training.

[0015] In the scheme, the target marine economic species is acquired according to the recognition result of the marine organism recognition model, specifically:

[0016] The environmental characteristics, hydrological characteristics and geographical location characteristics of the target sea area are acquired, and the target sea area characteristic sequence is generated according to the environmental characteristics, hydrological characteristics and geographical location characteristics;

[0017] The marine economic species category information meeting the suitable conditions of the target sea area is acquired through big data retrieval according to the target sea area characteristic sequence, and the target marine economic species list is generated through the marine economic species category information;

[0018] The recognition result of the marine organism recognition model and the target marine economic species list are intersected to acquire the target marine economic species category information contained in the underwater environment frame image information.

[0019] In the scheme, whether fishing can be carried out is determined according to the main growth stage and the quantity information, and a fishing route and a fishing plan are generated according to the determination result, specifically:

[0020] Obtain the time sequence of the underwater environment frame image information within a preset time, estimate the quantity information of the marine economic species in the preset area according to the number of various marine economic species in the underwater environment frame image information at each time point;

[0021] Obtain the morphological parameters of the target marine economic species according to the contour information and the parallax value of the target marine economic species in the underwater environment frame image information, and obtain the main growth stage of the same kind of marine economic species according to the morphological parameters of the same kind of marine economic species;

[0022] According to the target marine economic species category information, preset the corresponding main growth stage and quantity information threshold, and judge whether the main growth stage of the marine economic species is greater than the preset main growth stage threshold;

[0023] If greater, judge whether the quantity information of the marine economic species in the preset area is greater than the preset quantity information threshold;

[0024] If greater, generate a fishing route and a fishing plan according to the position information of the preset area in the target sea area and the weather information of the target sea area.

[0025] In the scheme, the fishing route and the fishing plan are generated according to the position information of the preset area in the target sea area and the weather information of the target sea area, specifically:

[0026] Obtain the position information and the environmental characteristics and hydrological characteristics of the preset area in the target sea area, and construct a feature sequence of the preset area according to the environmental characteristics and the hydrological characteristics;

[0027] Search for other areas with a feature sequence similar to that of the preset area within a preset similarity range in the target sea area, and take the other areas as candidate fishing areas;

[0028] Generate a fishing route according to the position information of the preset area and the position information of the candidate fishing area, and when the catch of the preset area is less than a preset catch threshold, enter the candidate fishing area for fishing;

[0029] At the same time, obtain the weather information of each area in the target sea area, correct the fishing area according to the weather information, and formulate a fishing plan according to the corrected fishing area.

[0030] The scheme further comprises:

[0031] Monitor the movement behavior of the marine organisms in the target area through the underwater environment monitoring video stream of the target sea area;

[0032] Obtain the mark box of the target marine organisms in each underwater environment frame image information within a preset time, and calculate the correlation degree of the mark boxes in two consecutive underwater environment frame image information;

[0033] When the correlation degree is greater than a preset correlation degree threshold, a trajectory segment is generated according to the mark box in the underwater environment frame image information of the two continuous frames, and the trajectory information of the target marine organism in a preset time is obtained according to the trajectory segment;

[0034] The marine organism trajectory recognition model is constructed and initialized and trained, the trajectory feature of the trajectory information of the target marine organism is extracted, and the abnormal motion behavior of the marine organism is obtained based on the biological trajectory recognition model according to the trajectory feature;

[0035] The activity degree of the marine organism is analyzed according to the abnormal motion behavior, different activity degree thresholds are preset according to seasons and time periods, and when the activity degree is greater than the activity degree threshold, target sea area water quality early warning information is generated.

[0036] The second aspect of the application also provides a marine economic species monitoring and identification fishing system, which comprises a memory and a processor, wherein the memory comprises a marine economic species monitoring and identification fishing method program, and the marine economic species monitoring and identification fishing method program is executed by the processor to realize the following steps:

[0037] An underwater environment monitoring video stream of a target sea area is obtained, underwater environment frame image information is obtained according to the underwater environment monitoring video stream, and the underwater environment frame image information is preprocessed;

[0038] A marine organism recognition model is constructed, the preprocessed underwater environment frame image information is imported into the marine organism recognition model, and a target marine economic species is obtained according to the recognition result of the marine organism model;

[0039] The main growth stage of the target marine economic species is obtained, and the number information of the target marine economic species in the target area is estimated;

[0040] It is judged whether the target marine economic species can be fished according to the main growth stage and the number information, and a fishing route and a fishing plan are generated according to the judgment result.

[0041] In the scheme, the marine organism recognition model is constructed, specifically as follows:

[0042] The marine organism recognition model is constructed based on a Mobilenet network and a YOLOv3 network and combined with fishery acoustic technology, different scale feature maps or feature data in the underwater environment frame image information are extracted through the Mobilenet network and the fishery acoustic technology;

[0043] The different scale feature maps or feature data are fused to generate comprehensive feature maps or feature data, the comprehensive feature maps or feature data are imported into the YOLOv3 network to predict the prior box, and the predicted value is compared with the true value to calculate the loss function.

[0044] The marine biological image and the label data are acquired through big data, and the marine biological recognition model is trained according to a minimized loss function.

[0045] The label frame with marine biological species information in the underwater environment frame image information is generated through the trained marine biological recognition model.

[0046] In the scheme, the target marine economic species is acquired according to the recognition result of the marine biological recognition model, and specifically,

[0047] The environmental characteristics, hydrological characteristics and geographical location characteristics of the target sea area are acquired, and the target sea area characteristic sequence is generated according to the environmental characteristics, hydrological characteristics and geographical location characteristics.

[0048] The marine economic species category information meeting the suitable conditions of the target sea area is acquired through big data retrieval according to the target sea area characteristic sequence, and the target marine economic species list is generated through the marine economic species category information.

[0049] The recognition result of the marine biological recognition model is intersected with the target marine economic species list to acquire the target marine economic species category information contained in the underwater environment frame image information.

[0050] In the scheme, whether the marine economic species can be caught is judged according to the main growth stage and quantity information, and the fishing route and fishing plan are generated according to the judgment result, and specifically,

[0051] The underwater environment frame image information time sequence in a preset time is acquired, and the quantity information of the marine economic species in the preset area is estimated according to the number of various marine economic species in the underwater environment frame image information at each time point.

[0052] The shape parameters of the target marine economic species are acquired according to the contour information and parallax value of the target marine economic species in the underwater environment frame image information, and the main growth stage of the same kind of marine economic species is acquired according to the shape parameters of the same kind of marine economic species.

[0053] The main growth stage threshold value corresponding to the target marine economic species category information is preset, and whether the main growth stage of the marine economic species is greater than the preset main growth stage threshold value is judged.

[0054] If it is greater, whether the quantity information of the marine economic species in the preset area is greater than the preset quantity information threshold value is judged.

[0055] If it is greater, the fishing route and fishing plan are generated according to the position information of the preset area in the target sea area and the weather information of the target sea area.

[0056] In the scheme, the fishing route and the fishing plan are generated according to the position information of the preset area in the target sea area and the weather information of the target sea area, and specifically are:

[0057] The position information and the environmental characteristics and hydrological characteristics of the preset area in the target sea area are acquired, and the feature sequence of the preset area is constructed according to the environmental characteristics and the hydrological characteristics;

[0058] Other areas with a feature sequence similarity to the preset area within the target sea area range being in a preset similarity range are searched, and the other areas are taken as candidate fishing areas;

[0059] The fishing route is generated according to the position information of the preset area and the position information of the candidate fishing area, and when the catch of the preset area is less than a preset catch threshold, fishing is carried out in the candidate fishing area;

[0060] Meanwhile, the weather information of each area of the target sea area is acquired, the fishing area is corrected according to the weather information, and the fishing plan is formulated according to the corrected fishing area.

[0061] In the scheme, the following steps are further included:

[0062] The motion behavior of the marine organism in the target area is monitored through the underwater environment monitoring video stream of the target sea area;

[0063] The marking box of the target marine organism in each underwater environment frame image information within a preset time is acquired, and the marking boxes in two continuous underwater environment frame image information are associated for degree calculation;

[0064] When the association degree is greater than a preset association degree threshold, a trajectory segment is generated according to the marking boxes in the two continuous underwater environment frame image information, and the trajectory information of the target marine organism within the preset time is acquired according to the trajectory segment;

[0065] A marine organism trajectory recognition model is constructed and initialized for training, the trajectory feature of the trajectory information of the target marine organism is extracted, and the abnormal motion behavior of the marine organism is acquired based on the biological trajectory recognition model according to the trajectory feature;

[0066] The activity degree of the marine organism is analyzed according to the abnormal motion behavior, different activity degree thresholds are preset according to seasons and time periods, and when the activity degree is greater than the activity degree threshold, target sea area water quality warning information is generated.

[0067] The third aspect of the present application further provides a computer readable storage medium, wherein the computer readable storage medium comprises a marine economic species monitoring and identification fishing method program, and the marine economic species monitoring and identification fishing method program is executed by a processor to realize the steps of the marine economic species monitoring and identification fishing method according to any one of the above.

[0068] The application discloses a marine economic species monitoring and identification fishing method and system and a storage medium, relates to the fishing technology field, and comprises the following steps: acquiring underwater environment monitoring video streams of a target sea area, acquiring underwater environment frame image information according to the underwater environment monitoring video streams, and preprocessing the underwater environment frame image information; constructing a marine biological identification model, inputting the preprocessed underwater environment frame image information into the marine biological identification model to identify marine economic species in the image, acquiring main growth stages of the marine economic species, estimating the quantity information of the marine economic species in the target area, judging whether the marine economic species can be fished according to the main growth stages and the quantity information, and generating a fishing route and a fishing plan according to the judgment result. The application realizes intelligent detection and identification of marine economic species in the target sea area, guarantees identification accuracy, and is beneficial to accurate fishing of the marine economic species. BRIEF DESCRIPTION OF DRAWINGS

[0069] Figure 1 A flowchart of a marine economic species monitoring and identification fishing method is shown.

[0070] Figure 2 A block diagram of a marine economic species monitoring and identification fishing system is shown. DETAILED DESCRIPTION

[0071] In order to more clearly understand the above-mentioned purposes, features and advantages of the application, the application will be further described in detail below in combination with the drawings and specific embodiments. It should be noted that the embodiments of the application and the features in the embodiments can be combined with each other without conflict.

[0072] In the following description, many specific details are set forth in order to provide a thorough understanding of the application, but the application can also be implemented in other ways different from those described herein, therefore, the protection scope of the application is not limited by the specific embodiments disclosed below.

[0073] Figure 1 A flowchart of a marine economic species monitoring and identification fishing method is shown.

[0074] As Figure 1 shown, the first aspect of the application provides a marine economic species monitoring and identification fishing method, which comprises the following steps:

[0075] S102, acquiring underwater environment monitoring video streams of a target sea area, acquiring underwater environment frame image information according to the underwater environment monitoring video streams, and preprocessing the underwater environment frame image information;

[0076] S104, a marine organism recognition model is constructed, the preprocessed underwater environment frame image information is input into the marine organism recognition model, and a target marine economic species is obtained according to a recognition result of the marine organism model;

[0077] S106, main growth stages of the target marine economic species are obtained, and quantity information of the target marine economic species in the target area is estimated;

[0078] S108, whether the target marine economic species can be caught is judged according to the main growth stages and the quantity information, and a fishing route and a fishing plan are generated according to a judgment result.

[0079] It should be noted that the underwater environment monitoring video stream of the target sea area can be obtained by a sensing device distributed in the target sea area, the underwater environment frame image information is enhanced in color and contrast through a dark channel prior algorithm and a threshold image segmentation, the contrast of the underwater environment frame image information is improved, the blur and noise are removed, and the underwater image quality is improved;

[0080] According to the marine organism recognition model, specifically, the marine organism recognition model is constructed based on a Mobilenet network and a YOLOv3 network and combined with fishery acoustic technology, different scale feature maps or feature data in the underwater environment frame image information are extracted through the Mobilenet network and the fishery acoustic technology, the different scale feature maps or feature data are fused to generate comprehensive feature maps or feature data, the comprehensive feature maps or feature data are input into the YOLOv3 network to predict prior boxes, and a predicted value is compared with a true value to calculate a loss function; marine organism images and label data are obtained through big data, the marine organism recognition model is trained according to a minimized loss function, the marine organism recognition model is trained by using a random gradient descent method, an initial learning rate is preset, and the training is stopped until the loss function tends to be stable; the underwater environment frame image information to be detected is input into the trained marine organism recognition model, features of the underwater environment frame image information to be detected are extracted, different scale features are fused through an adaptive feature fusion module to obtain fused features, the fused features are predicted in position and classification, redundant prior boxes are removed through a non-maximum suppression algorithm, and a label box with marine organism species information in the underwater environment frame image information is generated.

[0081] It should be noted that the target marine economic species is obtained according to the identification result of the marine organism identification model, specifically: the environmental characteristics, hydrological characteristics and geographical location characteristics of the target sea area are obtained, and the target sea area characteristic sequence is generated according to the environmental characteristics, hydrological characteristics and geographical location characteristics; the marine economic species category information meeting the survival conditions of the target sea area is obtained through big data retrieval according to the target sea area characteristic sequence, and the target marine economic species list is generated through the marine economic species category information; the identification result of the marine organism identification model is intersected with the target marine economic species list to obtain the target marine economic species category information contained in the underwater environment frame image information.

[0082] It should be noted that whether the fishing can be carried out is judged according to the main growth stage and quantity information, and the fishing route and fishing plan are generated according to the judgment result, specifically: the underwater environment frame image information time sequence in a preset time is obtained, and the quantity information of marine economic species in a preset area is estimated according to the number of various types of marine economic species in the underwater environment frame image information at each time point; the shape parameters of the target marine economic species are obtained according to the contour information and parallax value of the target marine economic species in the underwater environment frame image information, and the main growth stage of the same type of marine economic species is obtained according to the shape parameters of the same type of marine economic species; the main growth stage threshold value of the target marine economic species is preset according to the target marine economic species category information and economic value, and whether the main growth stage of the marine economic species is greater than the preset main growth stage threshold value is judged; if it is greater, whether the quantity information of the marine economic species in the preset area is greater than the preset quantity information threshold value is judged; if it is greater, the fishing route and fishing plan are generated according to the position information of the preset area in the target sea area and the weather information of the target sea area.

[0083] It should be noted that the fishing route and fishing plan are generated according to the position information of the preset area in the target sea area and the weather information of the target sea area, specifically: the position information and environmental characteristics, hydrological characteristics of the preset area in the target sea area are obtained, and the characteristic sequence of the preset area is constructed according to the environmental characteristics and hydrological characteristics; other areas with a similarity degree to the characteristic sequence of the preset area within a preset similarity range are searched in the target sea area, and the other areas are taken as candidate fishing areas; the fishing route is generated according to the position information of the preset area and the position information of the candidate fishing area, and the fishing is carried out in the preset area according to the fishing route, and when the fishing yield of the preset area is less than a preset fishing yield threshold value, the fishing is carried out in the candidate fishing area; the weather information of each area of the target sea area is obtained while the fishing route is determined, whether it is suitable for fishing activities is judged through the weather information, and the fishing area is corrected according to the weather information, and the fishing plan is made according to the corrected fishing area.

[0084] It should be noted that the present application also includes obtaining the abnormal motion behavior of marine organisms in the target area, generating a water quality warning of the target area, specifically

[0085] The motion behavior of marine organisms in the target area is monitored through the underwater environment monitoring video stream of the target sea area, the marking box of the target marine organisms in each underwater environment frame image information within a preset time is obtained, the correlation degree of the marking boxes in two consecutive underwater environment frame image information is calculated, and since the motion continuity of the target marine organisms, the coincidence degree and the center point distance value of the marking boxes in the front and rear two frame image information are used as the judgment standard of the correlation degree, when the correlation degree is greater than a preset correlation degree threshold, a trajectory segment is generated according to the marking boxes in the two consecutive underwater environment frame image information, and the trajectory information of the target marine organisms within the preset time is obtained according to the trajectory segment; a marine organism trajectory recognition model is constructed based on decision tree, recurrent neural network and the like, and is initialized and trained through a Fish4Knowledge data set, the trajectory feature of the trajectory information of the target marine organisms is extracted, the abnormal motion behavior of the marine organisms is obtained based on the trajectory recognition model according to the trajectory feature, and the activity degree of the marine organisms is analyzed according to the abnormal motion behavior, since the marine organisms usually have a stress reaction such as increased swimming speed, floating head and the like when encountering a water quality environment mutation, different activity degree thresholds are preset according to seasons and time periods, and when the activity degree is greater than the activity degree threshold, target sea area water quality warning information is generated.

[0086] Figure 2 A block diagram of a marine economic species monitoring and identification fishing system is shown.

[0087] The second aspect of the present application also provides a marine economic species monitoring and identification fishing system, which comprises a memory and a processor, the memory comprises a marine economic species monitoring and identification fishing method program, and the marine economic species monitoring and identification fishing method program is executed by the processor to realize the following steps:

[0088] An underwater environment monitoring video stream of a target sea area is obtained, underwater environment frame image information is obtained according to the underwater environment monitoring video stream, and the underwater environment frame image information is preprocessed;

[0089] A marine organism recognition model is constructed, the preprocessed underwater environment frame image information is imported into the marine organism recognition model, and a target marine economic species is obtained according to the recognition result of the marine organism model;

[0090] The main growth stage of the target marine economic species is obtained, and the number information of the target marine economic species in the target area is estimated;

[0091] According to the main growth stage and quantity information, it is judged whether fishing can be carried out, and a fishing route and a fishing plan are generated according to the judgment result.

[0092] It should be noted that the underwater environment monitoring video stream of the target sea area can be obtained by the sensing device distributed in the target sea area. The underwater environment frame image information is enhanced in color and contrast through the dark channel prior algorithm and threshold image segmentation, improving the contrast, removing blur and noise, and improving the quality of underwater images.

[0093] According to the construction of the marine organism recognition model, specifically: based on the Mobilenet network and YOLOv3 network and combined with fishery acoustic technology to construct marine organism recognition model, through Mobilenet network and fishery acoustic technology to extract different scale feature maps or feature data in underwater environment frame image information; the different scale feature maps or feature data are generated by feature fusion to generate comprehensive feature maps or feature data, the comprehensive feature maps or feature data are imported into YOLOv3 network for prior box prediction, and the predicted value is compared with the true value to calculate the loss function; through big data to obtain marine organism image and label data, according to the minimum loss function to train the marine organism recognition model, wherein the training of marine organism recognition model adopts the method of stochastic gradient descent, the initial learning rate is preset, and the training is stopped when the loss function tends to be stable; the underwater environment frame image information to be detected is input into the trained marine organism recognition model, the features are extracted, the features of different scales are fused through the adaptive feature fusion module, the fusion features are obtained, the position and classification prediction is carried out on the fusion features, the redundant prior boxes are removed through the non-maximum suppression algorithm, and the label box with marine organism species information in the underwater environment frame image information is generated.

[0094] It should be noted that the target marine economic species is obtained according to the recognition result of the marine organism recognition model, specifically: the environmental characteristics, hydrological characteristics and geographical location characteristics of the target sea area are obtained, and the target sea area characteristic sequence is generated according to the environmental characteristics, hydrological characteristics and geographical location characteristics; according to the target sea area characteristic sequence, the marine economic species category information meeting the suitable conditions of the target sea area is obtained through big data retrieval, and the target marine economic species list is generated through the marine economic species category information; the recognition result of the marine organism recognition model is intersected with the target marine economic species list to obtain the target marine economic species category information contained in the underwater environment frame image information.

[0095] It should be noted that according to the main growth stage and quantity information, it is judged whether fishing can be carried out, and a fishing route and a fishing plan are generated according to the judgment result, specifically: obtaining a time sequence of underwater environment frame image information in a preset time, estimating the quantity information of marine economic species in the preset area according to the number of various marine economic species in the underwater environment frame image information at each time point; obtaining the morphological parameters of the target marine economic species according to the contour information and the parallax value of the target marine economic species in the underwater environment frame image information, and obtaining the main growth stage of the same kind of marine economic species according to the morphological parameters of the same kind of marine economic species; according to the target marine economic species category information and the economic value preset corresponding main growth stage and quantity information threshold, judging whether the main growth stage of the marine economic species is greater than the preset main growth stage threshold; if greater, judging whether the quantity information of the marine economic species in the preset area is greater than the preset quantity information threshold; if greater, generating a fishing route and a fishing plan according to the position information of the preset area in the target sea area and the weather information of the target sea area.

[0096] It should be noted that the fishing route and the fishing plan are generated according to the position information of the preset area in the target sea area and the weather information of the target sea area, specifically: obtaining the position information and the environmental characteristics and hydrological characteristics of the preset area in the target sea area, constructing a feature sequence of the preset area according to the environmental characteristics and the hydrological characteristics; searching for other areas with a feature sequence similarity to the preset area within a preset similarity range in the target sea area, and taking the other areas as candidate fishing areas; generating a fishing route according to the position information of the preset area and the position information of the candidate fishing areas, and preferentially fishing in the preset area according to the fishing route, and when the fishing yield of the preset area is less than a preset fishing yield threshold, entering the candidate fishing area for fishing; obtaining the weather information of each area in the target sea area while the fishing route is being formulated, judging whether it is suitable for fishing activities through the weather information, and correcting the fishing area according to the weather information, and formulating a fishing plan according to the corrected fishing area.

[0097] It should be noted that the present application also includes obtaining the abnormal motion behavior of marine organisms in the target area, generating a water quality warning of the target area, specifically

[0098] The motion behavior of marine organisms in the target area is monitored through the underwater environment monitoring video stream of the target sea area, and the mark box of the target marine organism in each underwater environment frame image information within a preset time is acquired; the mark boxes in two consecutive underwater environment frame image information are associated, and since the motion continuity of the target marine organism, the coincidence degree and the center point distance value of the mark boxes in the front and rear two frame image information are taken as the judgment standard of the association degree, when the association degree is greater than a preset association degree threshold, a track segment is generated according to the mark boxes in the two consecutive underwater environment frame image information, and the track information of the target marine organism within the preset time is acquired according to the track segment; a marine organism track recognition model is constructed based on a decision tree, a recurrent neural network and the like, and is initialized and trained through a Fish4Knowledge data set, the track features of the track information of the target marine organism are extracted, and the abnormal motion behavior of the marine organism is acquired based on the track recognition model according to the track features; and the activity degree of the marine organism is analyzed according to the abnormal motion behavior, since the marine organism usually has a stress reaction such as increased swimming speed, floating head and the like when encountering a water quality environment mutation, different activity degree thresholds are preset according to seasons and time periods, and when the activity degree is greater than the activity degree threshold, target sea area water quality warning information is generated.

[0099] The third aspect of the present application also provides a computer readable storage medium, wherein the computer readable storage medium comprises a marine economic species monitoring and identification fishing method program, and the marine economic species monitoring and identification fishing method program is executed by a processor to realize the steps of the marine economic species monitoring and identification fishing method according to any one of the above.

[0100] The present application discloses a marine economic species monitoring and identification fishing method, system and storage medium, relates to the technical field of fishery fishing, and comprises the following steps: acquiring an underwater environment monitoring video stream of a target sea area, acquiring underwater environment frame image information according to the underwater environment monitoring video stream, and preprocessing the underwater environment frame image information; constructing a marine organism recognition model, inputting the preprocessed underwater environment frame image information into the marine organism recognition model to identify marine economic species in the image, acquiring main growth stages of the marine economic species, estimating the quantity information of the marine economic species in the target area, judging whether the marine economic species can be fished according to the main growth stages and the quantity information, and generating a fishing route and a fishing plan according to the judgment result. The present application realizes intelligent detection and identification of marine economic species in the target sea area, guarantees the identification accuracy, and is beneficial to accurate fishing of marine economic species.

[0101] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other manners. The described device embodiments are merely illustrative. For example, the division of the units is only a logical function division. There can be another division manner for the actual implementation, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed coupling, or direct coupling or communication connection between the components can be indirect coupling or communication connection through some interfaces, devices, or units, and can be electrical, mechanical, or in other forms.

[0102] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units; they can be located in one place, or distributed on multiple network units; and some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0103] In addition, each functional unit in each embodiment of the present application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in the form of hardware or in the form of hardware plus software functional units.

[0104] Those of ordinary skill in the art can understand that all or part of the steps of the above-described method embodiments can be completed by a program instructing related hardware, and the foregoing program can be stored in a computer readable storage medium, and when the program is executed, the steps of the method embodiments are executed; and the foregoing storage medium includes: mobile storage devices, read-only memories (ROMs), random access memories (RAMs), magnetic disks or optical disks, and various media that can store program codes.

[0105] Alternatively, the integrated units of the present application, if implemented in the form of software functional modules and sold or used as independent products, can also be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the embodiments of the present application can be embodied in the form of a software product, and the computer software product is stored in a storage medium, and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in the embodiments of the present application. The foregoing storage medium includes: mobile storage devices, ROMs, RAMs, magnetic disks or optical disks, and various media that can store program codes.

[0106] The above merely provides the specific implementation of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of the changes or replacements within the technical range disclosed by the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method of monitoring, identifying and fishing for a marine economic species, characterized by, The method comprises the following steps: obtaining an underwater environment monitoring video stream of a target sea area, obtaining underwater environment frame image information according to the underwater environment monitoring video stream, and preprocessing the underwater environment frame image information; constructing a marine organism recognition model, importing the preprocessed underwater environment frame image information into the marine organism recognition model, and obtaining a target marine economic species according to the recognition result of the marine organism model; obtaining the main growth stage of the target marine economic species and estimating the quantity information of the target marine economic species in the target area; judging whether the target marine economic species can be caught according to the main growth stage and the quantity information, and generating a fishing route and a fishing plan according to the judgment result; the marine organism recognition model is constructed in the following way: a marine organism recognition model is constructed based on a Mobilenet network and a YOLOv3 network and combined with fishery acoustic technology, different scale feature maps or feature data in the underwater environment frame image information are extracted through the Mobilenet network and the fishery acoustic technology; comprehensive feature maps or feature data are generated through feature fusion of the different scale feature maps or feature data, the comprehensive feature maps or feature data are imported into the YOLOv3 network for prediction of prior boxes, and the predicted value is compared with the true value to calculate a loss function; marine organism images and label data are obtained through big data, and the marine organism recognition model is trained according to the minimization of the loss function; a label box with marine organism species information in the underwater environment frame image information is generated through the trained marine organism recognition model.

2. The method according to claim 1, wherein, the target marine economic species is obtained according to the recognition result of the marine organism recognition model in the following way: environmental characteristics, hydrological characteristics and geographical location characteristics of the target sea area are obtained, and a target sea area feature sequence is generated according to the environmental characteristics, the hydrological characteristics and the geographical location characteristics; marine economic species category information that meets the survival conditions of the target sea area is obtained through big data retrieval according to the target sea area feature sequence, and a target marine economic species list is generated through the marine economic species category information; the recognition result of the marine organism recognition model is intersected with the target marine economic species list to obtain target marine economic species category information contained in the underwater environment frame image information.

3. The method according to claim 1, wherein, whether the target marine economic species can be caught is judged according to the main growth stage and the quantity information, and a fishing route and a fishing plan are generated according to the judgment result in the following way: a time sequence of underwater environment frame image information in a preset time is obtained, and the quantity information of marine economic species in a preset area is estimated according to the number of various types of marine economic species in the underwater environment frame image information at each time point; the morphological parameters of the target marine economic species are obtained according to the contour information and the parallax value of the target marine economic species in the underwater environment frame image information, and the main growth stage of the same type of marine economic species is obtained according to the morphological parameters of the same type of marine economic species; whether the main growth stage of the marine economic species is greater than a preset main growth stage threshold value is judged according to the target marine economic species category information, the preset corresponding main growth stage and the quantity information threshold value. If greater, it is judged whether the quantity information of the preset area marine economic species is greater than a preset quantity information threshold; if greater, a fishing route and a fishing plan are generated according to the position information of the preset area in the target sea area and the weather information of the target sea area.

4. The method of claim 1, wherein, The fishing route and the fishing plan are generated according to the position information of the preset area in the target sea area and the weather information of the target sea area, and specifically: The position information and environmental characteristics and hydrological characteristics of the preset area in the target sea area are acquired, and a feature sequence of the preset area is constructed according to the environmental characteristics and the hydrological characteristics; Other areas with a feature sequence similarity to the feature sequence of the preset area within a preset similarity range are searched in the target sea area, and the other areas are taken as candidate fishing areas; A fishing route is generated according to the position information of the preset area and the position information of the candidate fishing areas, and when the catch of the preset area is less than a preset catch threshold, fishing is performed in the candidate fishing areas; Meanwhile, weather information of each area in the target sea area is acquired, the fishing areas are corrected according to the weather information, and a fishing plan is formulated according to the corrected fishing areas.

5. The method of claim 1, wherein, Further comprising: The motion behavior of the marine organisms in the target area is monitored through the target sea area underwater environment monitoring video stream; A mark frame of the target marine organisms in each underwater environment frame image information within a preset time is acquired, and a correlation degree of the mark frames in two continuous underwater environment frame image information is calculated; When the correlation degree is greater than a preset correlation degree threshold, a trajectory segment is generated according to the mark frames in the two continuous underwater environment frame image information, and trajectory information of the target marine organisms within the preset time is acquired according to the trajectory segment; A marine organism trajectory recognition model is constructed and initialized, trajectory features of the trajectory information of the target marine organisms are extracted, and an abnormal motion behavior of the marine organisms is acquired based on the trajectory features and the biological trajectory recognition model; The activity level of the marine organisms is analyzed according to the abnormal motion behavior, different activity level thresholds are preset according to seasons and time periods, and when the activity level is greater than the activity level threshold, target sea area water quality warning information is generated.

6. A monitoring, identification and fishing system of marine economic species, characterized in that, The system comprises a memory and a processor, and the memory comprises a marine economic species monitoring, identification and fishing method program, and the marine economic species monitoring, identification and fishing method program is executed by the processor to realize the following steps: An underwater environment monitoring video stream of a target sea area is acquired, underwater environment frame image information is acquired according to the underwater environment monitoring video stream, and the underwater environment frame image information is preprocessed; A marine organism recognition model is constructed, the preprocessed underwater environment frame image information is imported into the marine organism recognition model, and a target marine economic species is acquired according to the recognition result of the marine organism model; Main growth stages of the target marine economic species are acquired, and the quantity information of the target marine economic species in a target area is estimated; Whether the target marine economic species can be fished is judged according to the main growth stages and the quantity information, and a fishing route and a fishing plan are generated according to the judgment result; The marine organism recognition model is constructed, and specifically: The marine organism recognition model is constructed based on a Mobilenet network and a YOLOv3 network and in combination with fishery acoustic technology, different scale feature maps or feature data in underwater environment frame image information are extracted through the Mobilenet network and the fishery acoustic technology; The different scale feature maps or feature data are fused to generate comprehensive feature maps or feature data, the comprehensive feature maps or feature data are introduced into the YOLOv3 network to predict prior boxes, and the predicted values are compared with the true values to calculate a loss function; The marine organism image and label data are acquired through big data, and the marine organism recognition model is trained according to the minimized loss function; The marine organism recognition model generates a label box with marine organism species information in the underwater environment frame image information after training.

7. A monitoring, identification and capture system of marine economic species according to claim 6, characterized in that, The target marine economic species is acquired according to the recognition result of the marine organism recognition model, specifically as follows: The environmental characteristics, hydrological characteristics and geographical location characteristics of a target sea area are acquired, and a target sea area characteristic sequence is generated according to the environmental characteristics, hydrological characteristics and geographical location characteristics; Marine economic species category information meeting the survival conditions of the target sea area is acquired through big data retrieval according to the target sea area characteristic sequence, and a target marine economic species list is generated through the marine economic species category information; The recognition result of the marine organism recognition model is intersected with the target marine economic species list to acquire target marine economic species category information contained in the underwater environment frame image information.

8. A monitoring, identification and capture system of marine economic species according to claim 6, characterized in that, Whether the marine economic species can be caught is determined according to the main growth stage and quantity information, and a fishing route and a fishing plan are generated according to the determination result, specifically as follows: A time sequence of underwater environment frame image information within a preset time is acquired, and the quantity information of marine economic species in a preset area is estimated according to the number of various marine economic species in the underwater environment frame image information at each time point; The shape parameters of the target marine economic species are acquired according to the contour information and parallax value of the target marine economic species in the underwater environment frame image information, and the main growth stage of the same type of marine economic species is acquired according to the shape parameters of the same type of marine economic species; Whether the main growth stage of the marine economic species is greater than a preset main growth stage threshold value is determined according to the target marine economic species category information and the preset corresponding main growth stage and quantity information threshold value; If yes, whether the quantity information of the marine economic species in the preset area is greater than a preset quantity information threshold value is determined; if yes, a fishing route and a fishing plan are generated according to the position information of the preset area in the target sea area and the weather information of the target sea area.

9. A computer-readable storage medium, characterized in that: The computer readable storage medium comprises a marine economic species monitoring, identification and fishing method program, and the processor executes the marine economic species monitoring, identification and fishing method program to realize the steps of the marine economic species monitoring, identification and fishing method according to any one of claims 1 to 5.

Images