A deep-sea aquaculture monitoring method and system based on sonar images

Abstract

The present invention discloses a deep-sea aquaculture monitoring method and system based on sonar images. The method acquires target area information to construct a three-dimensional visual monitoring model, monitors the marine environment and divides the aquaculture sub-areas. Sonar technology is used to detect fish schools, identify statistics and analyze the growth and distribution changes of fish schools. Based on the analysis results, a feeding and monitoring plan is formulated, and an equipment deployment plan is generated to realize dynamic deployment of monitoring equipment and improve the utilization rate of monitoring equipment, so as to optimize the automated management of deep-sea aquaculture and the allocation of monitoring equipment resources.

Description

Technical Field

[0001] The present invention relates to the field of sonar image analysis, and more specifically, to a deep-sea aquaculture monitoring method and system based on sonar images. Background Art

[0002] With the continuous development of deep-sea aquaculture technology, the demand for monitoring the aquaculture environment is increasing. As an important means of underwater detection, sonar images can obtain rich underwater terrain and biological information. However, traditional monitoring methods mainly rely on manual sampling and fixed-point observation, which is not only time-consuming and labor-intensive, but also unable to obtain large-scale data monitoring and analysis in real time, making it difficult to meet the needs of deep-sea aquaculture management.

[0003] In addition, in traditional technologies, there is a lack of regional analysis of marine aquaculture and optimized deployment analysis of monitoring equipment, which leads to increased ineffective monitoring costs and low monitoring accuracy. Therefore, there is an urgent need to develop a deep-sea aquaculture monitoring method based on sonar images. Summary of the invention

[0004] The present invention overcomes the defects of the prior art and proposes a deep-sea aquaculture monitoring method and system based on sonar images.

[0005] A first aspect of the present invention provides a deep-sea aquaculture monitoring method based on sonar images, comprising:

[0006] Obtain target area information for deep-sea aquaculture, and build a monitoring area model based on three-dimensional visualization through the target area information;

[0007] Based on the monitoring area model, the marine environment of the target area is monitored, and areas with similar environmental characteristics are marked and merged to form multiple aquaculture sub-areas;

[0008] During a monitoring period, based on sonar imaging technology and sonar positioning technology, sonar detection of farmed fish schools in the target area is carried out to obtain sonar feedback images and fish school positioning information. Farmed fish schools are identified and counted through sonar feedback images to obtain fish school identification information;

[0009] In each aquaculture sub-area, the periodic changes of fish growth and fish distribution are analyzed through fish identification information and fish positioning information to generate growth change characteristics and distribution change characteristics;

[0010] Feeding plan analysis and monitoring plan analysis are carried out based on the growth change characteristics and distribution change characteristics, and monitoring plans and feeding plans are generated. The breeding equipment settings of each breeding sub-area are analyzed through the monitoring plan and feeding plan to generate an equipment deployment plan.

[0011] In this solution, the target area information of deep-sea aquaculture is obtained, and a monitoring area model based on three-dimensional visualization is constructed through the target area information, specifically:

[0012] Obtain information on target areas for deep-sea aquaculture;

[0013] The target area information includes the aquaculture area, the area map outline, and the area ocean depth information;

[0014] A monitoring area model based on three-dimensional visualization is constructed based on the target area information.

[0015] In this scheme, the marine environment of the target area is monitored based on the monitoring area model, and areas with similar environmental characteristics are marked and merged to form multiple aquaculture sub-areas, specifically:

[0016] By monitoring the regional model, the target area is divided into multiple unit areas;

[0017] Acquire ocean environment data of the target area, wherein the ocean environment data includes temperature, dissolved oxygen, salinity, and ocean depth information;

[0018] Through the use of marine environmental data, each unit area is merged based on the preset environmental similarity standards, and adjacent unit areas with similar environmental characteristics are merged to eventually form multiple aquaculture sub-areas.

[0019] In this scheme, within a monitoring period, based on sonar imaging technology and sonar positioning technology, sonar detection of farmed fish schools in the target area is carried out to obtain sonar feedback images and fish school positioning information, and farmed fish schools are identified and counted through sonar feedback images to obtain fish school identification information, specifically:

[0020] During a monitoring period, based on sonar imaging technology, the target area is detected by sonar and sonar feedback images are obtained;

[0021] Convert sonar feedback images into preset image formats and perform image enhancement, image noise reduction, and standardization preprocessing;

[0022] Based on the image recognition module, fish school identification and fish school statistics are performed on the sonar feedback image, and fish school identification information is obtained;

[0023] Based on sonar imaging technology, sonar detection is carried out on the target area, and fish positioning analysis is performed based on sonar feedback to obtain fish positioning information;

[0024] The monitoring time period includes a plurality of monitoring cycles.

[0025] In this solution, in each breeding sub-area, the periodic change analysis of fish growth and fish distribution is performed through fish identification information and fish positioning information to generate growth change characteristics and distribution change characteristics, specifically:

[0026] Taking a breeding sub-area as the analysis unit, the periodic growth size change analysis of the fish schools in a breeding sub-area is carried out through the fish school identification information to obtain the growth characteristic value of each monitoring period;

[0027] Through the fish school identification information and fish school positioning information, the fish school in a breeding sub-area is periodically analyzed for distribution characteristics. Specifically, the distribution characteristics analysis includes: through the fish school identification information, the periodic quantity change analysis of the fish school is performed to form the quantity change characteristic value of each monitoring period. In the three-dimensional coordinates of the monitoring area model, the movement change of the coordinate position is calculated based on multiple monitoring periods through the fish school positioning information to obtain the movement distance value of each monitoring period.

[0028] All quantity change characteristic values ​​and moving distance values ​​are used as distribution change characteristics;

[0029] Based on the periodic change analysis of each breeding sub-area, the corresponding growth change characteristics and distribution change characteristics are obtained.

[0030] In this scheme, the feeding plan analysis and monitoring plan analysis are performed through the growth change characteristics and distribution change characteristics, and the monitoring plan and feeding plan are generated. The breeding equipment setting analysis is performed on each breeding sub-area through the monitoring plan and feeding plan to generate an equipment deployment plan, which is specifically:

[0031] Taking a farming sub-area as the analysis unit, the growth difference is compared with the preset farming plan through the growth change characteristics, and the feeding is planned based on the preset farming goals to form a feeding plan;

[0032] Based on the distribution change characteristics, multiple quantity change characteristic values ​​and multiple moving distance values ​​are obtained;

[0033] Calculate the average of multiple quantity change characteristic values ​​as the first monitoring requirement value, and calculate the average of multiple quantity change characteristic values ​​as the second monitoring requirement value;

[0034] Performing demand analysis and deployment analysis based on environment-related equipment through the first monitoring demand value, the deployment analysis including the number and location of equipment, and generating first equipment deployment information;

[0035] Performing demand analysis and deployment analysis based on monitoring equipment through the second monitoring demand value, the deployment analysis including the number and location of equipment, and generating second equipment deployment information;

[0036] The first device deployment information and the second device deployment information are combined to generate a device deployment solution.

[0037] The second aspect of the present invention further provides a deep-sea aquaculture monitoring system based on sonar images, the system comprising: a memory, a processor, the memory comprising a deep-sea aquaculture monitoring program based on sonar images, the deep-sea aquaculture monitoring program based on sonar images when executed by the processor implements the following steps:

[0038] Obtain target area information for deep-sea aquaculture, and build a monitoring area model based on three-dimensional visualization through the target area information;

[0039] Based on the monitoring area model, the marine environment of the target area is monitored, and areas with similar environmental characteristics are marked and merged to form multiple aquaculture sub-areas;

[0040] During a monitoring period, based on sonar imaging technology and sonar positioning technology, sonar detection of farmed fish schools in the target area is carried out to obtain sonar feedback images and fish school positioning information. Farmed fish schools are identified and counted through sonar feedback images to obtain fish school identification information;

[0041] In each aquaculture sub-area, the periodic changes of fish growth and fish distribution are analyzed through fish identification information and fish positioning information to generate growth change characteristics and distribution change characteristics;

[0042] Feeding plan analysis and monitoring plan analysis are carried out based on the growth change characteristics and distribution change characteristics, and monitoring plans and feeding plans are generated. The breeding equipment settings of each breeding sub-area are analyzed through the monitoring plan and feeding plan to generate an equipment deployment plan.

[0043] The third aspect of the present invention also provides a computer-readable storage medium, which includes a deep-sea aquaculture monitoring program based on sonar images. When the deep-sea aquaculture monitoring program based on sonar images is executed by a processor, the steps of the deep-sea aquaculture monitoring method based on sonar images as described in any one of the above items are implemented.

[0044] The present invention discloses a deep-sea aquaculture monitoring method and system based on sonar images. The method acquires target area information to construct a three-dimensional visual monitoring model, monitors the marine environment and divides the aquaculture sub-areas. Sonar technology is used to detect fish schools, identify statistics and analyze the growth and distribution changes of fish schools. Based on the analysis results, a feeding and monitoring plan is formulated, and an equipment deployment plan is generated to realize dynamic deployment of monitoring equipment and improve the utilization rate of monitoring equipment, so as to optimize the automated management of deep-sea aquaculture and the allocation of monitoring equipment resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1A flow chart of a deep-sea aquaculture monitoring method based on sonar images of the present invention is shown;

[0046] Figure 2 A block diagram of a deep-sea aquaculture monitoring system based on sonar images of the present invention is shown. DETAILED DESCRIPTION

[0047] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0048] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0049] Figure 1 A flow chart of a deep-sea aquaculture monitoring method based on sonar images of the present invention is shown.

[0050] like Figure 1 As shown, the first aspect of the present invention provides a deep-sea aquaculture monitoring method based on sonar images, comprising:

[0051] S102, obtaining target area information of deep-sea aquaculture, and constructing a monitoring area model based on three-dimensional visualization according to the target area information;

[0052] S104, based on the monitoring area model, the marine environment of the target area is monitored, and areas with similar environmental characteristics are marked and merged to form multiple aquaculture sub-areas;

[0053] S106, within a monitoring period, based on sonar imaging technology and sonar positioning technology, sonar detection of farmed fish schools in the target area is performed to obtain sonar feedback images and fish school positioning information, and farmed fish schools are identified and counted through the sonar feedback images to obtain fish school identification information;

[0054] S108, in each breeding sub-area, performing periodic change analysis on fish school growth and fish school distribution through fish school identification information and fish school positioning information, and generating growth change characteristics and distribution change characteristics;

[0055] S110, performing feeding plan analysis and monitoring plan analysis based on growth change characteristics and distribution change characteristics, and generating monitoring plans and feeding plans, performing breeding equipment setting analysis on each breeding sub-area through the monitoring plans and feeding plans, and generating equipment deployment plans.

[0056] It should be noted that the monitoring model is a map model for visually analyzing the location of farmed fish, sensor location, distribution, etc. The sonar imaging technology of the present invention is a multi-beam underwater imaging sonar, which can perform imaging analysis and monitoring of fish.

[0057] According to an embodiment of the present invention, the target area information of deep-sea aquaculture is obtained, and a monitoring area model based on three-dimensional visualization is constructed through the target area information, specifically:

[0058] Obtain information on target areas for deep-sea aquaculture;

[0059] The target area information includes the aquaculture area, the area map outline, and the area ocean depth information;

[0060] A monitoring area model based on three-dimensional visualization is constructed based on the target area information.

[0061] It should be noted that the monitoring area model is a three-dimensional visual map model, through which the distribution of monitoring equipment and fish schools can be visualized, thereby improving the intuitiveness of monitoring. The real-time collected fish school monitoring data can be visualized by sending it to the monitoring area model.

[0062] According to an embodiment of the present invention, the target area is monitored for marine environment based on the monitoring area model, and areas with similar environmental characteristics are marked and merged to form multiple aquaculture sub-areas, specifically:

[0063] By monitoring the regional model, the target area is divided into multiple unit areas;

[0064] Acquire ocean environment data of the target area, wherein the ocean environment data includes temperature, dissolved oxygen, salinity, and ocean depth information;

[0065] Through the use of marine environmental data, each unit area is merged based on the preset environmental similarity standards, and adjacent unit areas with similar environmental characteristics are merged to eventually form multiple aquaculture sub-areas.

[0066] It should be noted that the preset environmental similarity standard is a user-set standard, and the corresponding similar environmental standard is set based on multiple marine environmental information. Generally speaking, areas with similar marine environmental characteristics have similar characteristics in terms of the activity conditions of farmed fish. By dividing areas with different environments into farming sub-areas and conducting research, refined farming analysis can be achieved.

[0067] The target area is the deep-sea aquaculture area. The unit area is the smallest area under study. The environmental characteristics of the unit area are analyzed to merge the unit areas to form a large area with overall environmental similarity, and the large area is monitored and studied.

[0068] According to an embodiment of the present invention, within a monitoring period, based on sonar imaging technology and sonar positioning technology, sonar detection of farmed fish schools in the target area is performed and sonar feedback images and fish school positioning information are obtained, and farmed fish schools are identified and counted through sonar feedback images to obtain fish school identification information, specifically:

[0069] During a monitoring period, based on sonar imaging technology, the target area is detected by sonar and sonar feedback images are obtained;

[0070] Convert sonar feedback images into preset image formats and perform image enhancement, image noise reduction, and standardization preprocessing;

[0071] Based on the image recognition module, fish school identification and fish school statistics are performed on the sonar feedback image, and fish school identification information is obtained;

[0072] Based on sonar imaging technology, sonar detection is carried out on the target area, and fish positioning analysis is performed based on sonar feedback to obtain fish positioning information;

[0073] The monitoring time period includes a plurality of monitoring cycles.

[0074] It should be noted that the image recognition module is a CNN-based image recognition model, which learns and recognizes fish characteristics by training relevant historical sonar imaging data. The fish identification information includes the type of fish, the number of fish, and the size of the fish. The size information of the fish includes the size information of each farmed fish in the recognition image, and periodic growth analysis can be performed later. The monitoring time period includes multiple monitoring cycles, and periodic monitoring and analysis of the fish can be performed later.

[0075] According to an embodiment of the present invention, in each breeding sub-area, fish school identification information and fish school positioning information are used to perform periodic change analysis on fish school growth and fish school distribution, and growth change characteristics and distribution change characteristics are generated, specifically:

[0076] Taking a breeding sub-area as the analysis unit, the periodic growth size change analysis of the fish schools in a breeding sub-area is carried out through the fish school identification information to obtain the growth characteristic value of each monitoring period;

[0077] Through the fish school identification information and fish school positioning information, the fish school in a breeding sub-area is periodically analyzed for distribution characteristics. Specifically, the distribution characteristics analysis includes: through the fish school identification information, the periodic quantity change analysis of the fish school is performed to form the quantity change characteristic value of each monitoring period. In the three-dimensional coordinates of the monitoring area model, the movement change of the coordinate position is calculated based on multiple monitoring periods through the fish school positioning information to obtain the movement distance value of each monitoring period.

[0078] All quantity change characteristic values ​​and moving distance values ​​are used as distribution change characteristics;

[0079] Based on the periodic change analysis of each breeding sub-area, the corresponding growth change characteristics and distribution change characteristics are obtained.

[0080] It should be noted that, among the growth characteristic values ​​of each monitoring cycle, one growth characteristic value is a growth value obtained by comparing and calculating the current monitoring cycle with the previous monitoring cycle. This value is obtained by comparing the size of the fish school in the image. The greater the difference in the size of the fish school between the two monitoring cycles, the greater the value, which effectively reflects the growth rate and status. The movement change calculation of the coordinate position based on multiple monitoring cycles is the distance calculation of the position coordinates between the current monitoring cycle and the previous monitoring cycle. The three-dimensional coordinates are calculated by the spatial coordinate distance to obtain the distance value, which reflects the distribution position change within the two monitoring cycles. The distribution change characteristics include two characteristic change data: quantity and position change.

[0081] According to an embodiment of the present invention, the feeding plan analysis and the monitoring plan analysis are performed through the growth change characteristics and the distribution change characteristics, and a monitoring plan and a feeding plan are generated. The breeding equipment setting analysis is performed on each breeding sub-area through the monitoring plan and the feeding plan to generate an equipment deployment plan, specifically:

[0082] Taking a farming sub-area as the analysis unit, the growth difference is compared with the preset farming plan through the growth change characteristics, and the feeding is planned based on the preset farming goals to form a feeding plan;

[0083] Based on the distribution change characteristics, multiple quantity change characteristic values ​​and multiple moving distance values ​​are obtained;

[0084] Calculate the average of multiple quantity change characteristic values ​​as the first monitoring requirement value, and calculate the average of multiple quantity change characteristic values ​​as the second monitoring requirement value;

[0085] Performing demand analysis and deployment analysis based on environment-related equipment through the first monitoring demand value, the deployment analysis including the number and location of equipment, and generating first equipment deployment information;

[0086] Performing demand analysis and deployment analysis based on monitoring equipment through the second monitoring demand value, the deployment analysis including the number and location of equipment, and generating second equipment deployment information;

[0087] The first device deployment information and the second device deployment information are combined to generate a device deployment solution.

[0088] It should be noted that the preset breeding plan includes reference information of fish size in different time periods, fish growth change plan information, preset breeding target information (target quantity, size), etc. The initial feeding plan includes feeding amount and feeding cycle.

[0089] It is worth mentioning that in different marine aquaculture areas, due to the influence of environmental changes, pollution, and feeding conditions, there will be great differences in aquaculture conditions, which may affect the final aquaculture benefits. Therefore, it is necessary to achieve precise aquaculture area monitoring.

[0090] In the calculation of the mean of multiple quantity change characteristic values ​​as the first monitoring demand value and the calculation of the mean of multiple quantity change characteristic values ​​as the second monitoring demand value, the first monitoring demand value is a change index reflecting the number of fish, which is obtained based on the periodic monitoring analysis of the sub-region, and the value also reflects the monitoring demand. In the present invention, this value is used as an analysis index for the sub-region monitoring equipment, such as underwater video monitoring devices, sonar monitoring devices, etc. The larger the value, the greater the monitoring demand, the more the number of corresponding equipment, and the denser the distribution of monitoring equipment; the second monitoring demand value is a change index reflecting the position distribution of fish, which is obtained based on the periodic monitoring analysis of the sub-region, and the value also reflects the monitoring demand. In the present invention, this value is used as an analysis index for the sub-region environmental equipment, such as temperature sensors, dissolved oxygen sensors, pH sensors, water depth sensors, etc. The larger the value, the greater the monitoring demand, the more the number of corresponding environmental monitoring equipment, and the denser the distribution of equipment positions. Each breeding sub-region corresponds to a feeding plan and equipment deployment plan.

[0091] Through the present invention, it is possible to effectively realize precise marine aquaculture monitoring, analyze the growth and distribution changes of farmed fish in different aquaculture areas, and formulate corresponding monitoring equipment deployment plans according to the conditions of different aquaculture areas, realize dynamic deployment of monitoring equipment, improve the utilization rate of monitoring equipment, reduce unnecessary monitoring equipment settings, and achieve cost reduction and efficiency improvement for marine aquaculture.

[0092] According to an embodiment of the present invention, the step of converting the sonar feedback image into a preset image format and performing image enhancement, image noise reduction, and standardization preprocessing further includes:

[0093] In a current breeding sub-area, the sonar image with a recognition rate lower than a preset value is extracted through fish school recognition information and sonar feedback images to obtain first image data;

[0094] Performing fish area analysis and feature extraction on the first image data through an image recognition module to form fish feature data;

[0095] Mark the breeding sub-areas adjacent to the current breeding sub-area to obtain the adjacent areas;

[0096] Obtain sonar feedback images of adjacent areas and extract corresponding fish feature data, and integrate the feature data to form a comparative feature data set;

[0097] The fish feature data and the comparison feature data set are respectively converted into feature vectors, and based on the standard Euclidean distance, the features similar to the comparison feature data set and the fish feature data are analyzed, and the features with similarity greater than a preset threshold are extracted to form enhanced feature data;

[0098] Based on the DCA feature fusion algorithm, the fish feature data and the enhanced feature data are fused, the fused feature data are used to identify and count the secondary farmed fish schools, and the fish school identification information is updated in real time.

[0099] It should be noted that the adjacent breeding sub-areas include at least one.

[0100] It is worth mentioning that in the actual sonar image recognition process, due to the complexity of the marine environment, the sonar image often has small feature quantity, blurred features, low recognition, etc., which affects the recognition data. Therefore, corresponding image enhancement means are needed. In the present invention, when performing image feature analysis on the current area, the monitoring sonar image features of the adjacent area are used to perform feature enhancement analysis. Since the fish schools in the adjacent area are closest to the current area in terms of region, the corresponding fish school image features have a certain similarity. Through the preset similarity calculation method, the features with higher similarity are extracted and image enhancement is performed to improve the recognition rate of farmed fish schools in sonar images.

[0101] Figure 2 A block diagram of a deep-sea aquaculture monitoring system based on sonar images of the present invention is shown.

[0102] The second aspect of the present invention further provides a deep-sea aquaculture monitoring system 2 based on sonar images, the system comprising: a memory 21, a processor 22, the memory comprising a deep-sea aquaculture monitoring program based on sonar images, the deep-sea aquaculture monitoring program based on sonar images when executed by the processor implements the following steps:

[0103] Obtain target area information for deep-sea aquaculture, and build a monitoring area model based on three-dimensional visualization through the target area information;

[0104] Based on the monitoring area model, the marine environment of the target area is monitored, and areas with similar environmental characteristics are marked and merged to form multiple aquaculture sub-areas;

[0105] During a monitoring period, based on sonar imaging technology and sonar positioning technology, sonar detection of farmed fish schools in the target area is carried out to obtain sonar feedback images and fish school positioning information. Farmed fish schools are identified and counted through sonar feedback images to obtain fish school identification information;

[0106] In each aquaculture sub-area, the periodic changes of fish growth and fish distribution are analyzed through fish identification information and fish positioning information to generate growth change characteristics and distribution change characteristics;

[0107] Feeding plan analysis and monitoring plan analysis are carried out based on the growth change characteristics and distribution change characteristics, and monitoring plans and feeding plans are generated. The breeding equipment settings of each breeding sub-area are analyzed through the monitoring plan and feeding plan to generate an equipment deployment plan.

[0108] It should be noted that the monitoring model is a map model for visually analyzing the location of farmed fish, sensor location, distribution, etc. The sonar imaging technology of the present invention is a multi-beam underwater imaging sonar, which can perform imaging analysis and monitoring of fish.

[0109] According to an embodiment of the present invention, the target area information of deep-sea aquaculture is obtained, and a monitoring area model based on three-dimensional visualization is constructed through the target area information, specifically:

[0110] Obtain information on target areas for deep-sea aquaculture;

[0111] The target area information includes the aquaculture area, the area map outline, and the area ocean depth information;

[0112] A monitoring area model based on three-dimensional visualization is constructed based on the target area information.

[0113] It should be noted that the monitoring area model is a three-dimensional visual map model, through which the distribution of monitoring equipment and fish schools can be visualized, thereby improving the intuitiveness of monitoring. The real-time collected fish school monitoring data can be visualized by sending it to the monitoring area model.

[0114] According to an embodiment of the present invention, the target area is monitored for marine environment based on the monitoring area model, and areas with similar environmental characteristics are marked and merged to form multiple aquaculture sub-areas, specifically:

[0115] By monitoring the regional model, the target area is divided into multiple unit areas;

[0116] Acquire ocean environment data of the target area, wherein the ocean environment data includes temperature, dissolved oxygen, salinity, and ocean depth information;

[0117] Through the use of marine environmental data, each unit area is merged based on the preset environmental similarity standards, and adjacent unit areas with similar environmental characteristics are merged to eventually form multiple aquaculture sub-areas.

[0118] It should be noted that the preset environmental similarity standard is a user-set standard, and the corresponding similar environmental standard is set based on multiple marine environmental information. Generally speaking, areas with similar marine environmental characteristics have similar characteristics in terms of the activity conditions of farmed fish. By dividing areas with different environments into farming sub-areas and conducting research, refined farming analysis can be achieved.

[0119] The target area is the deep-sea aquaculture area. The unit area is the smallest area under study. The environmental characteristics of the unit area are analyzed to merge the unit areas to form a large area with overall environmental similarity, and the large area is monitored and studied.

[0120] According to an embodiment of the present invention, within a monitoring period, based on sonar imaging technology and sonar positioning technology, sonar detection of farmed fish schools in the target area is performed and sonar feedback images and fish school positioning information are obtained, and farmed fish schools are identified and counted through sonar feedback images to obtain fish school identification information, specifically:

[0121] During a monitoring period, based on sonar imaging technology, the target area is detected by sonar and sonar feedback images are obtained;

[0122] Convert sonar feedback images into preset image formats and perform image enhancement, image noise reduction, and standardization preprocessing;

[0123] Based on the image recognition module, fish school identification and fish school statistics are performed on the sonar feedback image, and fish school identification information is obtained;

[0124] Based on sonar imaging technology, sonar detection is carried out on the target area, and fish positioning analysis is performed based on sonar feedback to obtain fish positioning information;

[0125] The monitoring time period includes a plurality of monitoring cycles.

[0126] It should be noted that the image recognition module is a CNN-based image recognition model, which learns and recognizes fish characteristics by training relevant historical sonar imaging data. The fish identification information includes the type of fish, the number of fish, and the size of the fish. The size information of the fish includes the size information of each farmed fish in the recognition image, and periodic growth analysis can be performed later. The monitoring time period includes multiple monitoring cycles, and periodic monitoring and analysis of the fish can be performed later.

[0127] According to an embodiment of the present invention, in each breeding sub-area, fish school identification information and fish school positioning information are used to perform periodic change analysis on fish school growth and fish school distribution, and growth change characteristics and distribution change characteristics are generated, specifically:

[0128] Taking a breeding sub-area as the analysis unit, the periodic growth size change analysis of the fish schools in a breeding sub-area is carried out through the fish school identification information to obtain the growth characteristic value of each monitoring period;

[0129] Through the fish school identification information and fish school positioning information, the fish school in a breeding sub-area is periodically analyzed for distribution characteristics. Specifically, the distribution characteristics analysis includes: through the fish school identification information, the periodic quantity change analysis of the fish school is performed to form the quantity change characteristic value of each monitoring period. In the three-dimensional coordinates of the monitoring area model, the movement change of the coordinate position is calculated based on multiple monitoring periods through the fish school positioning information to obtain the movement distance value of each monitoring period.

[0130] All quantity change characteristic values ​​and moving distance values ​​are used as distribution change characteristics;

[0131] Based on the periodic change analysis of each breeding sub-area, the corresponding growth change characteristics and distribution change characteristics are obtained.

[0132] It should be noted that, among the growth characteristic values ​​of each monitoring cycle, one growth characteristic value is a growth value obtained by comparing and calculating the current monitoring cycle with the previous monitoring cycle. This value is obtained by comparing the size of the fish school in the image. The greater the difference in the size of the fish school between the two monitoring cycles, the greater the value, which effectively reflects the growth rate and status. The movement change calculation of the coordinate position based on multiple monitoring cycles is the distance calculation of the position coordinates between the current monitoring cycle and the previous monitoring cycle. The three-dimensional coordinates are calculated by the spatial coordinate distance to obtain the distance value, which reflects the distribution position change within the two monitoring cycles. The distribution change characteristics include two characteristic change data: quantity and position change.

[0133] According to an embodiment of the present invention, the feeding plan analysis and the monitoring plan analysis are performed through the growth change characteristics and the distribution change characteristics, and a monitoring plan and a feeding plan are generated. The breeding equipment setting analysis is performed on each breeding sub-area through the monitoring plan and the feeding plan to generate an equipment deployment plan, specifically:

[0134] Taking a farming sub-area as the analysis unit, the growth difference is compared with the preset farming plan through the growth change characteristics, and the feeding is planned based on the preset farming goals to form a feeding plan;

[0135] Based on the distribution change characteristics, multiple quantity change characteristic values ​​and multiple moving distance values ​​are obtained;

[0136] Calculate the average of multiple quantity change characteristic values ​​as the first monitoring requirement value, and calculate the average of multiple quantity change characteristic values ​​as the second monitoring requirement value;

[0137] Performing demand analysis and deployment analysis based on environment-related equipment through the first monitoring demand value, the deployment analysis including the number and location of equipment, and generating first equipment deployment information;

[0138] Performing demand analysis and deployment analysis based on monitoring equipment through the second monitoring demand value, the deployment analysis including the number and location of equipment, and generating second equipment deployment information;

[0139] The first device deployment information and the second device deployment information are combined to generate a device deployment solution.

[0140] It should be noted that the preset breeding plan includes reference information of fish size in different time periods, fish growth change plan information, preset breeding target information (target quantity, size), etc. The initial feeding plan includes feeding amount and feeding cycle.

[0141] It is worth mentioning that in different marine aquaculture areas, due to the influence of environmental changes, pollution, and feeding conditions, there will be great differences in aquaculture conditions, which may affect the final aquaculture benefits. Therefore, it is necessary to achieve precise aquaculture area monitoring.

[0142] In the calculation of the mean of multiple quantity change characteristic values ​​as the first monitoring demand value and the calculation of the mean of multiple quantity change characteristic values ​​as the second monitoring demand value, the first monitoring demand value is a change index reflecting the number of fish, which is obtained based on the periodic monitoring analysis of the sub-region, and the value also reflects the monitoring demand. In the present invention, this value is used as an analysis index for the sub-region monitoring equipment, such as underwater video monitoring devices, sonar monitoring devices, etc. The larger the value, the greater the monitoring demand, the more the number of corresponding equipment, and the denser the distribution of monitoring equipment; the second monitoring demand value is a change index reflecting the position distribution of fish, which is obtained based on the periodic monitoring analysis of the sub-region, and the value also reflects the monitoring demand. In the present invention, this value is used as an analysis index for the sub-region environmental equipment, such as temperature sensors, dissolved oxygen sensors, pH sensors, water depth sensors, etc. The larger the value, the greater the monitoring demand, the more the number of corresponding environmental monitoring equipment, and the denser the distribution of equipment positions. Each breeding sub-region corresponds to a feeding plan and equipment deployment plan.

[0143] Through the present invention, it is possible to effectively realize precise marine aquaculture monitoring, analyze the growth and distribution changes of farmed fish in different aquaculture areas, and formulate corresponding monitoring equipment deployment plans according to the conditions of different aquaculture areas, realize dynamic deployment of monitoring equipment, improve the utilization rate of monitoring equipment, reduce unnecessary monitoring equipment settings, and achieve cost reduction and efficiency improvement for marine aquaculture.

[0144] According to an embodiment of the present invention, the step of converting the sonar feedback image into a preset image format and performing image enhancement, image noise reduction, and standardization preprocessing further includes:

[0145] In a current breeding sub-area, the sonar image with a recognition rate lower than a preset value is extracted through fish school recognition information and sonar feedback images to obtain first image data;

[0146] Performing fish area analysis and feature extraction on the first image data through an image recognition module to form fish feature data;

[0147] Mark the breeding sub-areas adjacent to the current breeding sub-area to obtain the adjacent areas;

[0148] Obtain sonar feedback images of adjacent areas and extract corresponding fish feature data, and integrate the feature data to form a comparative feature data set;

[0149] The fish feature data and the comparison feature data set are respectively converted into feature vectors, and based on the standard Euclidean distance, the features similar to the comparison feature data set and the fish feature data are analyzed, and the features with similarity greater than a preset threshold are extracted to form enhanced feature data;

[0150] Based on the DCA feature fusion algorithm, the fish feature data and the enhanced feature data are fused, the fused feature data are used to identify and count the secondary farmed fish schools, and the fish school identification information is updated in real time.

[0151] It should be noted that the adjacent breeding sub-areas include at least one.

[0152] It is worth mentioning that in the actual sonar image recognition process, due to the complexity of the marine environment, the sonar image often has small feature quantity, blurred features, low recognition, etc., which affects the recognition data. Therefore, corresponding image enhancement means are needed. In the present invention, when performing image feature analysis on the current area, the monitoring sonar image features of the adjacent area are used to perform feature enhancement analysis. Since the fish schools in the adjacent area are closest to the current area in terms of region, the corresponding fish school image features have a certain similarity. Through the preset similarity calculation method, the features with higher similarity are extracted and image enhancement is performed to improve the recognition rate of farmed fish schools in sonar images.

[0153] The third aspect of the present invention also provides a computer-readable storage medium, which includes a deep-sea aquaculture monitoring program based on sonar images. When the deep-sea aquaculture monitoring program based on sonar images is executed by a processor, the steps of the deep-sea aquaculture monitoring method based on sonar images as described in any one of the above items are implemented.

[0154] The present invention discloses a deep-sea aquaculture monitoring method and system based on sonar images. The method acquires target area information to construct a three-dimensional visual monitoring model, monitors the marine environment and divides the aquaculture sub-areas. Sonar technology is used to detect fish schools, identify statistics and analyze the growth and distribution changes of fish schools. Based on the analysis results, a feeding and monitoring plan is formulated, and an equipment deployment plan is generated to realize dynamic deployment of monitoring equipment and improve the utilization rate of monitoring equipment, so as to optimize the automated management of deep-sea aquaculture and the allocation of monitoring equipment resources.

[0155] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0156] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0157] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0158] A person of ordinary skill in the art can understand that: all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: a mobile storage device, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0159] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0160] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A deep-sea aquaculture monitoring method based on sonar images, characterized in that: include: Obtain target area information for deep-sea aquaculture, and build a monitoring area model based on three-dimensional visualization through the target area information; Based on the monitoring area model, the marine environment of the target area is monitored, and areas with similar environmental characteristics are marked and merged to form multiple aquaculture sub-areas; During a monitoring period, based on sonar imaging technology and sonar positioning technology, sonar detection of farmed fish schools in the target area is carried out to obtain sonar feedback images and fish school positioning information. Farmed fish schools are identified and counted through sonar feedback images to obtain fish school identification information; In each aquaculture sub-area, the periodic changes of fish growth and fish distribution are analyzed through fish identification information and fish positioning information to generate growth change characteristics and distribution change characteristics; Feeding plan analysis and monitoring plan analysis are carried out based on growth change characteristics and distribution change characteristics, and monitoring plans and feeding plans are generated. Through monitoring plans and feeding plans, aquaculture equipment setting analysis is carried out for each aquaculture sub-area to generate equipment deployment plans; Wherein, in each breeding sub-area, the periodic change analysis of fish growth and fish distribution is performed through fish identification information and fish positioning information to generate growth change characteristics and distribution change characteristics, specifically: Taking a breeding sub-area as the analysis unit, the periodic growth size change analysis of the fish schools in a breeding sub-area is carried out through the fish school identification information to obtain the growth characteristic value of each monitoring period; Through the fish school identification information and fish school positioning information, the fish school in a breeding sub-area is periodically analyzed for distribution characteristics. Specifically, the distribution characteristics analysis includes: through the fish school identification information, the periodic quantity change analysis of the fish school is performed to form the quantity change characteristic value of each monitoring period. In the three-dimensional coordinates of the monitoring area model, the movement change of the coordinate position is calculated based on multiple monitoring periods through the fish school positioning information to obtain the movement distance value of each monitoring period. All quantity change characteristic values ​​and moving distance values ​​are used as distribution change characteristics; Based on the periodic change analysis of each breeding sub-area, the corresponding growth change characteristics and distribution change characteristics are obtained; Among them, the feeding plan analysis and monitoring plan analysis are performed through the growth change characteristics and distribution change characteristics, and the monitoring plan and feeding plan are generated. The breeding equipment setting analysis is performed on each breeding sub-area through the monitoring plan and feeding plan to generate an equipment deployment plan, specifically: Taking a farming sub-area as the analysis unit, the growth difference is compared with the preset farming plan through the growth change characteristics, and the feeding is planned based on the preset farming goals to form a feeding plan; Based on the distribution change characteristics, multiple quantity change characteristic values ​​and multiple moving distance values ​​are obtained; Calculate the average of multiple quantity change characteristic values ​​as the first monitoring requirement value, and calculate the average of multiple quantity change characteristic values ​​as the second monitoring requirement value; Performing demand analysis and deployment analysis based on environment-related equipment through the first monitoring demand value, the deployment analysis including the number and location of equipment, and generating first equipment deployment information; Performing demand analysis and deployment analysis based on monitoring equipment through the second monitoring demand value, the deployment analysis including the number and location of equipment, and generating second equipment deployment information; The first device deployment information and the second device deployment information are combined to generate a device deployment solution.

2. A deep-sea aquaculture monitoring method based on sonar images according to claim 1, characterized in that: The target area information of deep-sea aquaculture is obtained, and a monitoring area model based on three-dimensional visualization is constructed according to the target area information, specifically: Obtain information on target areas for deep-sea aquaculture; The target area information includes the aquaculture area, the area map outline, and the area ocean depth information; A monitoring area model based on three-dimensional visualization is constructed based on the target area information.

3. A deep-sea aquaculture monitoring method based on sonar images according to claim 2, characterized in that: Based on the monitoring area model, the target area is monitored for marine environment, and areas with similar environmental characteristics are marked and merged to form multiple aquaculture sub-areas, specifically: By monitoring the regional model, the target area is divided into multiple unit areas; Acquire ocean environment data of the target area, wherein the ocean environment data includes temperature, dissolved oxygen, salinity, and ocean depth information; Through marine environmental data, based on preset environmental similarity standards, adjacent unit areas with similar environmental characteristics are merged to eventually form multiple aquaculture sub-areas.

4. A deep-sea aquaculture monitoring method based on sonar images according to claim 3, characterized in that: In the monitoring period, based on sonar imaging technology and sonar positioning technology, sonar detection of farmed fish schools in the target area is performed to obtain sonar feedback images and fish school positioning information, and farmed fish schools are identified and counted through sonar feedback images to obtain fish school identification information, specifically: During a monitoring period, based on sonar imaging technology, the target area is detected by sonar and sonar feedback images are obtained; Convert sonar feedback images into preset image formats and perform image enhancement, image noise reduction, and standardization preprocessing; Based on the image recognition module, fish school identification and fish school statistics are performed on the sonar feedback image, and fish school identification information is obtained; Based on sonar imaging technology, sonar detection is carried out on the target area, and fish positioning analysis is performed based on sonar feedback to obtain fish positioning information; The monitoring time period includes a plurality of monitoring cycles.

5. A deep-sea aquaculture monitoring system based on sonar images, characterized in that: The system includes: a memory and a processor, wherein the memory includes a deep-sea aquaculture monitoring program based on sonar images, and when the deep-sea aquaculture monitoring program based on sonar images is executed by the processor, the following steps are implemented: Obtain target area information for deep-sea aquaculture, and build a monitoring area model based on three-dimensional visualization through the target area information; Based on the monitoring area model, the marine environment of the target area is monitored, and areas with similar environmental characteristics are marked and merged to form multiple aquaculture sub-areas; During a monitoring period, based on sonar imaging technology and sonar positioning technology, sonar detection of farmed fish schools in the target area is carried out to obtain sonar feedback images and fish school positioning information. Farmed fish schools are identified and counted through sonar feedback images to obtain fish school identification information; In each aquaculture sub-area, the periodic changes of fish growth and fish distribution are analyzed through fish identification information and fish positioning information to generate growth change characteristics and distribution change characteristics; Feeding plan analysis and monitoring plan analysis are carried out based on growth change characteristics and distribution change characteristics, and monitoring plans and feeding plans are generated. Through monitoring plans and feeding plans, aquaculture equipment setting analysis is carried out for each aquaculture sub-area to generate equipment deployment plans; Wherein, in each breeding sub-area, the periodic change analysis of fish growth and fish distribution is performed through fish identification information and fish positioning information to generate growth change characteristics and distribution change characteristics, specifically: Taking a breeding sub-area as the analysis unit, the periodic growth size change analysis of the fish schools in a breeding sub-area is carried out through the fish school identification information to obtain the growth characteristic value of each monitoring period; Through the fish school identification information and fish school positioning information, the fish school in a breeding sub-area is periodically analyzed for distribution characteristics. Specifically, the distribution characteristics analysis includes: through the fish school identification information, the periodic quantity change analysis of the fish school is performed to form the quantity change characteristic value of each monitoring period. In the three-dimensional coordinates of the monitoring area model, the movement change of the coordinate position is calculated based on multiple monitoring periods through the fish school positioning information to obtain the movement distance value of each monitoring period. All quantity change characteristic values ​​and moving distance values ​​are used as distribution change characteristics; Based on the periodic change analysis of each breeding sub-area, the corresponding growth change characteristics and distribution change characteristics are obtained; Among them, the feeding plan analysis and monitoring plan analysis are performed through the growth change characteristics and distribution change characteristics, and the monitoring plan and feeding plan are generated. The breeding equipment setting analysis is performed on each breeding sub-area through the monitoring plan and feeding plan to generate an equipment deployment plan, specifically: Taking a farming sub-area as the analysis unit, the growth difference is compared with the preset farming plan through the growth change characteristics, and the feeding is planned based on the preset farming goals to form a feeding plan; Based on the distribution change characteristics, multiple quantity change characteristic values ​​and multiple moving distance values ​​are obtained; Calculate the average of multiple quantity change characteristic values ​​as the first monitoring requirement value, and calculate the average of multiple quantity change characteristic values ​​as the second monitoring requirement value; Performing demand analysis and deployment analysis based on environment-related equipment through the first monitoring demand value, the deployment analysis including the number and location of equipment, and generating first equipment deployment information; Performing demand analysis and deployment analysis based on monitoring equipment through the second monitoring demand value, the deployment analysis including the number and location of equipment, and generating second equipment deployment information; The first device deployment information and the second device deployment information are combined to generate a device deployment solution.

6. A deep-sea aquaculture monitoring system based on sonar images according to claim 5, characterized in that: The target area information of deep-sea aquaculture is obtained, and a monitoring area model based on three-dimensional visualization is constructed according to the target area information, specifically: Obtain information on target areas for deep-sea aquaculture; The target area information includes the aquaculture area, the area map outline, and the area ocean depth information; A monitoring area model based on three-dimensional visualization is constructed based on the target area information.

7. The deep-sea aquaculture monitoring system based on sonar images according to claim 5 is characterized in that: Based on the monitoring area model, the target area is monitored for marine environment, and areas with similar environmental characteristics are marked and merged to form multiple aquaculture sub-areas, specifically: By monitoring the regional model, the target area is divided into multiple unit areas; Acquire ocean environment data of the target area, wherein the ocean environment data includes temperature, dissolved oxygen, salinity, and ocean depth information; Through marine environmental data, based on preset environmental similarity standards, adjacent unit areas with similar environmental characteristics are merged to eventually form multiple aquaculture sub-areas.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a deep-sea aquaculture monitoring program based on sonar images. When the deep-sea aquaculture monitoring program based on sonar images is executed by a processor, the steps of the deep-sea aquaculture monitoring method based on sonar images as described in any one of claims 1 to 4 are implemented.

Images