阅读说明：本技术 一种河蟹养殖投饵量的精准确定方法 (Method for accurately determining feeding amount for river crab culture ) 是由 赵德安 贺帆 孙月平 秦云 刘晓洋 于 2019-08-29 设计创作，主要内容包括：本发明公开了一种河蟹养殖投饵量的精准确定方法。包括构建河蟹养殖池塘全域各网格水位水质水温参数数据库和河蟹养殖池塘特定点水位水质水温参数数据库,通过神经网络建立河蟹养殖池塘全域各网格水位水质水温参数与特定点水位水质水温参数之间的映射关系,估计当前池塘全域各网格水位水质水温参数；由水质水温参数对河蟹摄食能力的影响,确定反映河蟹摄食能力的各网格摄食系数；由自动导航船搭载水下摄像系统和导航系统,形成池塘全域各网格河蟹密度分布图；在河蟹处于平均密度分布区域,设置观测点,确定平均投饵密度,结合全域各网格河蟹密度分布及河蟹摄食系数,可得到全域各网格投饵量密度分布,根据各网格的面积,确定全域投放饵料总量。(The invention discloses a method for accurately determining the bait feeding amount for river crab culture. Constructing a water quality and temperature parameter database of each grid water level and water temperature parameter database of the universe of the river crab aquaculture pond and a water quality and water temperature parameter database of a specific point water level of the river crab aquaculture pond, establishing a mapping relation between the water quality and water temperature parameter of each grid water level and water temperature parameter of the universe of the river crab aquaculture pond and the water quality and water temperature parameter of the specific point water level through a neural network, and estimating each grid water level and water quality and water temperature parameter of the univers; determining each grid ingestion coefficient reflecting the ingestion ability of the river crabs according to the influence of the water quality and water temperature parameters on the ingestion ability of the river crabs; carrying an underwater camera system and a navigation system by an automatic navigation ship to form density distribution maps of river crabs in all grids of the pond universe; and setting an observation point in the region where the river crabs are in the average density distribution, determining the average bait casting density, combining the density distribution of the river crabs in each grid of the universe and the ingestion coefficient of the river crabs to obtain the bait casting quantity density distribution of each grid of the universe, and determining the total bait casting quantity of the universe according to the area of each grid.)

1. An automatic and accurate determination method for the feeding amount of river crab culture is characterized by comprising the following steps:

(a) carrying water level, water quality and water temperature sensors by using an automatic navigation ship to realize the acquisition of water level parameters of the water quality and the water temperature of the water level of the global bottom layer and the middle layer of the river crab culture pond for multiple times, and establishing a global two-dimensional gridding water level, water quality and water temperature parameter database of the river crab culture pond by combining longitude and latitude information of a GPS/Beidou navigation system of the automatic navigation ship;

(b) arranging water quality, water temperature and water level sensors at the bottom layer of 1 optional point of each of a waterweed area and a waterweed-free area of the river crab culture pond within a distance of more than 30m from the side of the pond, collecting the dissolved oxygen content, the PH value, the water temperature and the water level parameters of the river crab culture pond, and establishing a water quality and water temperature parameter historical database of the water level of a specific point of the river crab culture pond by combining longitude and latitude information and time information of the specific point;

(c) and estimating the global two-dimensional gridding water level, water quality and water temperature parameters of the current river crab culture pond according to the current water quality and water temperature parameters of the specific point water level and the global two-dimensional gridding water level, water quality and water temperature parameters by combining the water quality and water temperature parameter database of the specific point water level of the river crab culture pond and the global two-dimensional gridding water level, water quality and water temperature parameter database. Determining each two-dimensional grid ingestion coefficient C reflecting the ingestion capability of the river crabs according to the influence of the water quality and water temperature parameters on the ingestion capability of the river crabs_ij:

C_ij＝C_tij*C_dij

Wherein C is_tijWater temperature ingestion coefficient of i row and j column in two-dimensional grid, C_dijThe dissolved oxygen feeding coefficients of i rows and j columns in the two-dimensional grid are obtained;

(d) the method comprises the following steps of carrying an underwater camera system by using an automatic navigation ship to obtain underwater images of natural growth conditions of river crabs in the whole pond, identifying the river crabs on line by combining a digital image processing technology, accurately obtaining activity conditions of the river crabs, fusing GPS positioning information in the operation process of the automatic navigation ship, and forming a two-dimensional gridding river crab density distribution diagram Q of a pond bait feeding area:

Q＝{q₁₁，q₁₂，q₁₃，…q_ij…q_mn}

wherein m and n are respectively the maximum values of the grid quantity in the row direction and the column direction in the two-dimensional grid;

(e) the density of the river crabs is in the average density distribution area of the river crabs in the pond:

(f) and obtaining the bait density distribution omega required by the pond universe by combining the average bait feeding density omega, and the density distribution Q and the ingestion coefficient C of each grid river crab in the pond universe:

(g) combining the area S of each grid_ijThe total amount of bait T required by the pond universe can be calculated:

2. the method for automatically and accurately determining the feeding amount of river crab culture according to claim 1, wherein the method comprises the following steps: the step a specifically comprises the following steps:

step1, detecting the global water quality parameters of the river crab culture pond at least 2 times per month in the early stage of river crab culture and at least 1 time per month in the later stage of culture, wherein the parameters at least comprise dissolved oxygen content, water quality PH value, water temperature and water level change;

step2, in order to overcome the influence of dynamic change of water quality and temperature parameters of the river crab culture pond, collecting the water quality and temperature parameters each time after the water falls into the mountains in the sun, so that photosynthesis of water plants in the pond is stopped, the content of dissolved oxygen in the water body of the pond is in a monotonous reduction process, and river crabs usually eat at night;

step3, dividing the whole pond into m × n grids, preferably taking the area of each grid equal, taking the side length of the grid to be 7-10m, carrying out water level, water quality and water temperature parameter detection on the central position of each grid, carrying out detection on the water level, the water quality and the water temperature for 1 time according to the original path of the automatic navigation ship, and recording corresponding parameters and detection time;

step4, recording longitude and latitude information of the automatic navigation ship in the water quality and water temperature acquisition process, and establishing a global gridding water level, water quality and water temperature parameter database of the river crab culture pond.

3. The method for automatically and accurately determining the feeding amount of river crab culture according to claim 1, wherein the method comprises the following steps: the estimation of the water quality and temperature parameters in the step c and the influence on the feeding capacity of the river crabs specifically comprise the following steps:

3.1) according to the change trend of the two detection results of the same point in the global two-dimensional gridding water level water quality and water temperature parameter database, matching the change trend of the water quality parameter of the corresponding time period of the two specific point water level water quality and water temperature parameter historical databases of the river crab culture pond, determining whether each gridding water quality parameter detection point belongs to a water grass area or a water grass-free area, so as to be bound with the corresponding specific point water level water quality and water temperature parameter historical database, and estimating the global two-dimensional gridding water level water quality and water temperature parameter of the current river crab culture pond according to the current two specific point water level water quality and water temperature parameters;

3.2) the river crab is suitable for growing in slightly alkaline fresh water, the pH value is 7.5-8.5, the pH value is used as a monitoring index, and a method for conditioning water quality is adopted when the pH value exceeds the standard. The higher the dissolved oxygen in water is, the larger the food intake of the river crabs is, and the optimal dissolved oxygen is required to be more than 5 mg/L;

3.3) relationship between the feeding intensity of the river crab and the water temperature is as follows: when the water temperature is below 10 ℃, the ingestion intensity is obviously reduced; when the water temperature is as low as 5 ℃, the river crabs basically do not eat; when the water temperature reaches more than 10 ℃, particularly 20-30 ℃, the river crabs can eat food vigorously.

4. The method for automatically and accurately determining the feeding amount of river crab culture according to claim 1, wherein the method comprises the following steps: the step d specifically comprises:

4.1) the underwater camera system consists of 1-3 underwater cameras with lighting systems, and the distance between the cameras is 1 m. The underwater camera system adjusts the distance between the underwater camera and the pond bottom to be 50cm according to the water level value of each grid, and underwater images of the natural growth condition of river crabs in the whole area of the pond are obtained;

4.2) the digital image processing technology relates to image denoising filtering and training multilayer convolutional neural networks, wherein the convolutional neural networks are alternately composed of 10 convolutional layers and 6 maximum pooling layers, and online recognition of the underwater river crabs is realized;

4.3) the GPS positioning information of the integrated automatic navigation ship is specifically as follows: in the process of online identifying the river crabs, a synchronous system clock positioned by a GPS is adopted. And on the basis of time, combining the online river crab identification information and the GPS positioning information, dividing the number of the river crabs identified in each grid by the number of the processed pictures to obtain the relative river crab density of the corresponding grid, and integrating the information of all the grids to form a pond bait casting area river crab relative density distribution diagram Q.

5. The method for automatically and accurately determining the feeding amount of river crab culture according to claim 1, wherein the method comprises the following steps: and e, setting at least 3 observation points, wherein the distance between every two observation points is more than or equal to 30m, and the monitoring of the foraging condition of the river crabs specifically comprises the following steps: bait slightly higher than the average feeding density is thrown, and the foraging condition of the river crabs is observed:

if the bait is completely consumed, the bait throwing density is low, and the bait throwing density needs to be properly improved;

if more baits remain, the bait throwing density is higher, and the bait throwing density needs to be properly reduced;

if the bait remains less, the bait feeding density is just proper and is used as the basis omega of the average feeding density of the bait₁；

Average bait feeding density omega obtained by the method₁Divided by the water temperature intake coefficient c at the observation point_tAnd dissolved oxygen feeding coefficient c_dObtaining the average feeding density omega of the standard bait₀；

Average feeding density omega of standard baits obtained from all observation points₀And averaging to obtain the average bait feeding density omega.

Technical Field

The invention relates to an accurate determination method of the feeding amount of river crab culture, in particular to the influence of water temperature and water quality parameters on the feeding capacity, the density distribution of each grid river crab in the whole breeding pond, the average feeding density and the feeding coefficient of the river crab, the density distribution of the feeding amount of each grid in the whole pond is determined, and the feeding amount of the river crab culture is accurately determined by combining the area of each grid.

Background

The river crab contains rich protein and trace elements, has high nutritive value and delicious taste, and the requirement of people on the river crab is gradually increased, so that the river crab breeding industry is rapidly developed. The river crab cultivation in China is mainly distributed in Jiangsu provinces, Hubei provinces, Anhui provinces and other provinces, but the Jiangsu province is most obvious, and the river crab yield of China occupies more than half of the domestic market. Although the river crab breeding industry is developed greatly, a problem to be solved urgently still exists, namely how to accurately determine the feeding amount of the river crab breeding bait. At present, bait feeding amount is determined according to previous breeding experience in the river crab breeding process, influence of water temperature and water quality parameters on feeding capacity of the river crabs is not fully considered, density distribution of the river crabs in a river crab breeding pond cannot be determined quantitatively, and whether the feeding amount of each time meets the actual growth requirement of the river crabs cannot be considered. In addition, most of the feeding is carried out by manpower, so that the feeding randomness is high, and the feeding is not uniform. The patent with the application number of 201610710797.3 discloses an autonomous navigation river crab breeding bait casting device and an uniform bait casting method, which uses an autonomous navigation operating boat as a carrier to perform uniform bait casting, and solves the problem of uneven bait casting. However, the method of the invention of patent application No. 201610710797.3 only solves the problem of uniform feeding, but cannot scientifically determine whether the bait feeding amount meets the actual feeding requirement of the river crabs.

Disclosure of Invention

The invention aims to provide an automatic and accurate determination method of the bait casting amount for river crab culture, which is used for determining the bait casting amount required by each casting, accurately determining the bait casting amount density of each grid of the whole area of a river crab culture pond and scientifically realizing the bait casting according to the requirement.

The technical scheme of the invention comprises the following steps:

(a) carrying water level, water quality and water temperature sensors by using an automatic navigation ship to realize the acquisition of water quality and water temperature parameters of the global bottom layer (about 20cm away from the bottom of the pond), the middle layer (about 50cm away from the bottom of the pond) and the water level value of the crab pond for multiple times, and establishing a global two-dimensional gridding water level, water quality and water temperature parameter database of the river crab culture pond by combining the longitude and latitude information of a GPS/Beidou navigation system of the automatic navigation ship;

(b) arranging water quality, water temperature and water level sensors at the bottom layer of 1 optional point (about 20cm away from the bottom of the pond) of each of a water grass area and a water grass-free area within the range of more than 30m away from the side of the pond in the river crab culture pond, collecting parameters of the river crab culture pond such as dissolved oxygen content, a pH value, water temperature and water level, and establishing a water quality and water temperature parameter historical database of the water level of a specific point of the river crab culture pond by combining longitude and latitude information and time information of the specific point;

(c) and estimating the global two-dimensional gridding water level, water quality and water temperature parameters of the current river crab culture pond according to the current water quality and water temperature parameters of the specific point water level and the global two-dimensional gridding water level, water quality and water temperature parameters by combining the water quality and water temperature parameter database of the specific point water level of the river crab culture pond and the global two-dimensional gridding water level, water quality and water temperature parameter database. Determining each two-dimensional grid ingestion coefficient C reflecting the ingestion capability of the river crabs according to the influence of the water quality and water temperature parameters on the ingestion capability of the river crabs_ij:

C_ij＝C_tij*C_dij

Wherein C is_tijWater temperature feeding coefficients for i rows and j columns in the two-dimensional grid are determined from table 1, C_dijThe dissolved oxygen feeding coefficients of i rows and j columns in the two-dimensional grid are determined from table 2;

(d) the method comprises the following steps of carrying an underwater camera system by using an automatic navigation ship to obtain underwater images of natural growth conditions of river crabs in the whole pond, identifying the river crabs on line by combining a digital image processing technology, accurately obtaining activity conditions of the river crabs, fusing GPS positioning information in the operation process of the automatic navigation ship, and forming a two-dimensional gridding river crab density distribution diagram Q of a pond bait feeding area:

Q＝{q₁₁，q₁₂，q₁₃，…q_ij…q_mn}

wherein m and n are respectively the maximum values of the grid quantity in the row direction and the column direction in the two-dimensional grid;

(e) the density of the river crabs is in the average density distribution area of the river crabs in the pond:

setting at least 3 observation points (the distance between each point is more than or equal to 30m), wherein each observation point consists of a water temperature and water quality sensor and a feeding platform, monitoring the foraging condition of the river crabs and determining the average bait feeding density omega;

(f) and obtaining the bait density distribution omega required by the pond universe by combining the average bait feeding density omega, and the density distribution Q and the ingestion coefficient C of each grid river crab in the pond universe:

(g) combining the area S of each grid_ijThe total amount of bait T required by the pond universe can be calculated:

further, an automatic and accurate determination method for the feeding amount of river crab culture is characterized by comprising the following steps: the step a specifically comprises the following steps:

step1, detecting the global water quality parameters of the river crab culture pond at least 2 times per month in the early stage of river crab culture and at least 1 time per month in the later stage of culture, wherein the parameters at least comprise dissolved oxygen content, water quality PH value, water temperature and water level change;

step2, in order to overcome the influence of dynamic change of water quality and temperature parameters of the river crab culture pond, collecting the water quality and temperature parameters each time after the water falls into the mountains in the sun, so that photosynthesis of water plants in the pond is stopped, the content of dissolved oxygen in the water body of the pond is in a monotonous reduction process, and river crabs usually eat at night;

step3, dividing the whole pond into m × n grids, preferably taking the area of each grid equal, taking the side length of the grid to be 7-10m, carrying out water temperature, water quality and water level parameter detection on the central position of each grid, collecting water quality, water temperature and water level parameters, carrying out detection 1 time according to the original path of the automatic navigation ship, and recording corresponding parameters and detection time;

step4, recording longitude and latitude information of the automatic navigation ship in the water quality and water temperature acquisition process, and establishing a global gridding water level, water quality and water temperature parameter database of the river crab culture pond;

further, an automatic and accurate determination method for the feeding amount of river crab culture is characterized by comprising the following steps: the estimation of the water quality and temperature parameters in the step c and the influence on the feeding capacity of the river crabs specifically comprise the following steps:

(1) according to the change trend of the two detection results of the same point in the global two-dimensional gridded water level, water quality and water temperature parameter database, matching the change trend of the water quality and water temperature parameter of the two specific point water levels in the river crab culture pond in the corresponding time period, determining whether each gridded water quality parameter detection point belongs to a water grass area or a water grass-free area so as to be bound with the corresponding specific point water level, water quality and water temperature parameter historical database, and estimating the global two-dimensional gridded water level, water quality and water temperature parameter of the current river crab culture pond according to the current two specific point water level, water quality and water temperature parameters;

(2) the river crab is suitable for growing in slightly alkaline fresh water, the pH value is 7.5-8.5, the pH value is used as a monitoring index, and a method for conditioning water quality is adopted when the pH value exceeds the standard. The higher the dissolved oxygen in water is, the larger the food intake of the river crabs is, and the optimal dissolved oxygen is required to be more than 5 mg/L;

(3) the relationship between the ingestion intensity of the river crabs and the water temperature is as follows: when the water temperature is below 10 ℃, the ingestion intensity is obviously reduced; when the water temperature is as low as 5 ℃, the river crabs basically do not eat; when the water temperature reaches more than 10 ℃, particularly 20-30 ℃, the river crabs ingest exuberantly;

further, an automatic and accurate determination method for the feeding amount of river crab culture is characterized by comprising the following steps: the underwater imaging system carried in the step d combines a digital image processing technology to identify river crabs on line, and the underwater imaging system and the GPS positioning information integrated in the operation process of the automatic navigation ship specifically comprise:

(1) the underwater camera system is composed of 1-3 underwater cameras with lighting systems, and the distance between the cameras is 1 m. The underwater camera system adjusts the distance between the underwater camera and the pond bottom to be 50cm according to the water level value of each grid, and underwater images of the natural growth condition of river crabs in the whole area of the pond are obtained;

(2) the digital image processing technology relates to the technology of image denoising filtering and training multilayer convolutional neural network, the convolutional neural network in the invention is composed of 10 convolutional layers and 6 maximum pooling layers alternately (the method is a known technology, a convolutional neural network library can be directly called in MATLAB, and the number of the convolutional layers 10 and the pooling layers is set to be 6, so that the optimal recognition effect can be achieved) so as to realize the online recognition of the underwater river crabs;

(3) the GPS positioning information of the integrated automatic navigation ship is specifically as follows: in the process of online identifying the river crabs, a synchronous system clock positioned by a GPS is adopted. On the basis of time, combining online river crab identification information and GPS positioning information, dividing the number of river crabs identified in each grid by the number of processed pictures to obtain the relative density of river crabs in the corresponding grid, and integrating the information of all grids to form a pond bait casting area river crab relative density distribution diagram Q;

further, the method is simple. An automatic and accurate determination method for the bait feeding amount of river crab culture is characterized by comprising the following steps: the step e of setting at least 3 observation points (the distance between each observation point is more than or equal to 30m) to monitor the foraging condition of the river crabs specifically comprises the following steps: bait slightly higher than the average feeding density is thrown, and the foraging condition of the river crabs is observed:

(1) if the bait is completely consumed, the bait throwing density is low, and the bait throwing density needs to be properly improved;

(2) if more baits remain, the bait throwing density is higher, and the bait throwing density needs to be properly reduced;

(3) if the bait remains less, the bait feeding density is just proper and is used as the basis omega of the average feeding density of the bait₁；

(4) Average bait feeding density omega obtained by the method₁Divided by the water temperature intake coefficient c at the observation point_tAnd dissolvingOxygen intake coefficient c_dObtaining the average feeding density omega of the standard bait₀；

(5) Average feeding density omega of standard baits obtained from all observation points₀And averaging to obtain the average bait feeding density omega.

Compared with the prior art, the invention has the following beneficial effects: the method comprises the steps of establishing the correlation between the water level, water quality and water temperature parameters of all network water levels of the river crab culture pond universe and the water quality and water temperature parameters of specific point water levels through a neural network, and estimating the water level, water quality and water temperature parameters of all grid water levels of the pond universe according to the water quality and water temperature parameters of the current specific point water levels; determining an ingestion coefficient reflecting the ingestion capability of the river crabs according to the influence of the water quality and the water temperature parameters on the ingestion capability of the river crabs; carrying an underwater camera system by an automatic navigation ship, acquiring the distribution of each network river crab in the universe of the pond, and forming a distribution density map of each grid river crab by combining positioning information of a GPS/Beidou navigation system; setting observation points in the region where the river crabs are in the average density distribution, accurately obtaining the feeding capacity information of the river crabs, and determining the average feeding density; and determining the bait feeding amount required by each feeding by combining the density distribution of each grid river crab in the universe, the river crab feeding coefficient, the grid area and the average bait feeding density, accurately determining the bait feeding amount density of each grid in the universe of the river crab culture pond, and scientifically realizing bait feeding according to requirements.

Drawings

FIG. 1 is a view of an automatic navigation work vessel platform

FIG. 2 is a schematic view of the automatic navigation boat for feeding bait to and fro

FIG. 3 shows the installation positions of the fixed-point water level, water quality and water temperature sensors at the bottom of the river crab culture pond and the feeding density observation points

Detailed Description

The following further describes embodiments of the present invention with reference to the schematic drawings.

The invention provides a method for accurately determining the bait feeding amount for river crab culture. Is used for solving the problem of accurately determining the bait feeding amount in the river crab breeding process. The method is particularly characterized in that a pond global two-dimensional gridding water level, water temperature and water quality parameter database and a specific point water level, water temperature and water quality parameter historical database are combined, the correlation between each global grid water level, water quality and water temperature parameter and a specific point water level, water temperature and water temperature parameter of the river crab aquaculture pond is established through a neural network, and the global two-dimensional gridding water level, water temperature and water quality parameter of the aquaculture pond is estimated according to the current specific point water level, water quality and water temperature parameter of the pond. And establishing a river crab density distribution map by adopting the underwater camera and the longitude and latitude information of the GPS. The installation and the embodiment of the device are explained in detail below.

1. Installation of equipment

Fig. 1 is a simplified diagram of an automatic navigation ship platform, which comprises a control cabinet 1, a water level, water temperature and water quality sensor 2, underwater camera equipment (3, 4 and 5) and underwater auxiliary lighting systems (6 and 7). A control cabinet is arranged at the stern of a navigation ship, a navigation ship control panel and GPS navigation equipment (including a mobile station and a base station) are contained in the control cabinet, and a water level, water temperature and water quality sensor is arranged at the position of the bottom of the control cabinet submerged in water. The paddle wheels are distributed on two sides of the middle part of the navigation ship and have the same horizontal position. 3 underwater camera devices are arranged at the front connecting rod of the navigation ship, the distance between the devices is 1m, and 2 underwater auxiliary lighting systems are arranged between the camera devices.

Fig. 2 is a diagram of an automatic navigation path of an automatic navigation ship in a river crab culture pond, wherein the abscissa in a coordinate system represents latitude information, and the ordinate represents longitude information. M1(lat1, lon1), M2(lat2, lon2), M3(lat3, lon3) and M4(lat4, lon4) respectively represent longitude and latitude information of four end points of the river crab culture pond.

And the position of the flag in fig. 3 is a reference diagram of the setting position of the water quality and temperature sensor at a specific point of the bottom layer of the river crab culture pond. The circle area is an assumed density distribution area of 3 river crabs, an average bait feeding density observation point is arranged at the area, and the observation point is composed of a water temperature and water quality sensor and a feeding platform.

2 establishing a global water level, water quality and water temperature parameter database of the river crab culture pond

And an automatic navigation ship is adopted to carry water level, water quality and water temperature sensors to realize the acquisition of water quality and water temperature parameters of the global bottom layer (about 20cm away from the bottom of the pond), the central layer (about 50cm away from the bottom of the pond) and the water level value of the crab pond. And detecting global two-dimensional gridding water quality parameters of the river crab culture pond at least 2 times per month in the early stage of river crab culture and at least 1 time per month in the later stage of culture, wherein the parameters at least comprise dissolved oxygen content, water quality PH value, water temperature and water level change. In order to overcome the influence of dynamic change of water quality and temperature parameters of the river crab culture pond, the water quality and temperature parameters are collected after the sun falls on the mountains every time, so that the photosynthesis of water plants in the pond is stopped, the dissolved oxygen content in the water body of the pond is in a monotonous reduction process, and the river crabs usually eat the water at night. The pond universe is divided into m-n grids, the area of each grid is preferably equal, the side length of each grid is 7-10m generally, the central position of each grid is used for detecting water temperature, water quality and water level parameters, the water quality, water temperature and water level parameters are collected according to the path of an automatic navigation ship shown in a graph 2, the detection is carried out for 1 time, the average value of 2 detection parameter results is obtained, the longitude and latitude and time information of the automatic navigation ship in the water quality and water temperature collection process are recorded, and a universe grid water level, water quality and water temperature parameter database of the river crab culture pond is established.

3 establishing a water quality and temperature parameter database of a specific point water level of the river crab culture pond

Water quality, water temperature and water level sensors are arranged at the optional 1-point bottom layer (about 20cm away from the bottom of the pond) of each of a float grass area and a non-float grass area within the range that the distance between the river crab culture pond and the side of the pond is more than 30m, and are shown by the small flag marks in figure 3. And parameters such as the dissolved oxygen content, the PH value, the water temperature, the water level and the like of the river crab culture pond are collected at the specific points, and a historical database of water quality and water temperature parameters of the water level at the specific points of the river crab culture pond is established by combining longitude and latitude and time information of the specific points.

4 acquisition of feeding coefficient C

And the relation between the parameters in the same time period can be obtained by combining the historical database of the water quality and water temperature parameters of the specific point water level of the river crab culture pond and the database of the water quality and water temperature parameters of the global two-dimensional gridding water level of the river crab culture pond, and then the water quality and water temperature parameters of the global two-dimensional gridding water level of the river crab culture pond are estimated according to the current water quality and water temperature parameters of the specific point water. The estimation of the water quality and water temperature parameters and the influence on the feeding capacity of the river crabs are specifically as follows:

(1) according to the change trend of the two detection results of the same point in the global two-dimensional gridded water level, water quality and water temperature parameter database, matching the change trend of the water quality and water temperature parameter of the two specific point water levels in the river crab culture pond in the corresponding time period, determining whether each gridded water quality parameter detection point belongs to a water grass area or a water grass-free area so as to be bound with the corresponding specific point water level, water quality and water temperature parameter historical database, and estimating the global two-dimensional gridded water level, water quality and water temperature parameter of the current river crab culture pond according to the current two specific point water level, water quality and water temperature parameters;

(2) the river crab is suitable for growing in slightly alkaline fresh water, the pH value is 7.5-8.5, the pH value is used as a monitoring index, and a method for conditioning water quality is adopted when the pH value exceeds the standard. The higher the dissolved oxygen in water is, the larger the food intake of the river crabs is, and the optimal dissolved oxygen is required to be more than 5 mg/L;

(3) the relationship between the ingestion intensity of the river crabs and the water temperature is as follows: when the water temperature is below 10 ℃, the ingestion intensity is obviously reduced; when the water temperature is as low as 5 ℃, the river crabs basically do not eat; when the water temperature reaches more than 10 ℃, particularly 20-30 ℃, the river crabs ingest exuberantly;

according to the water quality and water temperature parameters of the river crab culture universe, corresponding ingestion coefficients C are given to all grid areas_ijNamely:

C_ij＝C_tij*C_dij

wherein C is_tijWater temperature feeding coefficients for i rows and j columns in the two-dimensional grid are determined from table 1, C_dijThe dissolved oxygen intake coefficients for rows i and columns j in the two-dimensional grid are determined from table 2.

5 obtaining river crab density distribution Q in pond feeding area

And carrying an underwater camera system by adopting an automatic navigation ship, and adjusting the distance between an underwater camera and the bottom of the pond to be 50cm according to the water level value of each grid of the water level to obtain an underwater image of the natural growth condition of the river crabs in the whole pond. The method adopts digital image processing technologies such as image denoising filtering and multilayer convolutional neural network training, wherein the convolutional neural network is composed of 10 convolutional layers and 6 maximum pooling layers in an alternating mode, and online recognition of the underwater river crabs is achieved. In the process of online identifying the river crabs, a synchronous system clock positioned by a GPS is adopted. By taking time as a reference, combining online recognition information of the river crabs with GPS longitude and latitude information in the running process of the automatic navigation ship, dividing the number of the river crabs recognized in each grid by the number of the processed pictures to obtain the relative density of the river crabs of the corresponding grid, and integrating the information of all the grids, a two-dimensional gridding river crab relative density distribution diagram Q in a pond bait casting area can be formed:

Q＝{q₁₁，q₁₂，q₁₃，…q_ij…q_mn}

wherein m and n are respectively the maximum value of the grid number in the row direction and the column direction in the two-dimensional grid.

6 set bait feeding density observation points

The density of the river crabs is in the average density distribution area of the river crabs in the pond:

(taking the circle of the figure 3 as an example, the distance between each point is more than or equal to 30m), an observation point is arranged, and the observation point consists of a water temperature and water quality sensor and a feeding platform and is used for monitoring the foraging condition of the river crabs. Specifically, bait slightly higher than average feeding density is put in, and the foraging condition of the river crabs is observed:

(1) if the bait is completely consumed, the bait throwing density is low, and the bait throwing density needs to be properly improved;

(2) if more baits remain, the bait throwing density is higher, and the bait throwing density needs to be properly reduced;

(3) if the bait remains less, the bait feeding density is just proper and is used as the basis omega of the average feeding density of the bait₁；

(4) Average bait feeding density omega obtained by the method₁Divided by the water temperature intake coefficient c at the observation point_tAnd dissolved oxygen feeding coefficient c_dObtaining the average feeding density omega of the standard bait₀；

(5) Average feeding density omega of standard baits obtained from all observation points₀And averaging to obtain the average bait feeding density omega.

7 calculating the total amount of bait

Combining the current average bait feeding density omega, and the two-dimensional gridding river crab density distribution Q and the ingestion coefficient C of the pond universe, the bait density distribution omega required by the two-dimensional gridding of the pond universe can be obtained

Combining the area S of each grid_ijThe total amount of bait T required by the pond universe can be calculated:

table 1 shows the corresponding relationship between water temperature and feeding coefficient of river crabs

Table 2 shows the corresponding relationship between the dissolved oxygen and the feeding coefficient of river crabs

Dissolved oxygen content mg/L	Dissolved oxygen feeding coefficient
		≤2	0
2.5	0.3
		3	0.46
3.5	0.53
		4	0.6
4.5	0.75
		5	0.9
5.5	0.95
		≥6	1

In conclusion, the invention provides the method for accurately determining the feeding amount of the river crab culture. Constructing a water quality and temperature parameter database of each grid water level and water temperature parameter database of the universe of the river crab aquaculture pond and a water quality and water temperature parameter database of a specific point water level of the river crab aquaculture pond, establishing a mapping relation between the water quality and water temperature parameter of each grid water level and water temperature parameter of the universe of the river crab aquaculture pond and the water quality and water temperature parameter of the specific point water level through a neural network, and estimating each grid water level and water quality and water temperature parameter of the univers; determining each grid ingestion coefficient reflecting the ingestion ability of the river crabs according to the influence of the water quality and water temperature parameters on the ingestion ability of the river crabs; carrying an underwater camera system and a navigation system by an automatic navigation ship to form density distribution maps of river crabs in all grids of the pond universe; and setting an observation point in the region where the river crabs are in the average density distribution, determining the average bait casting density, combining the density distribution of the river crabs in each grid of the universe and the ingestion coefficient of the river crabs to obtain the bait casting quantity density distribution of each grid of the universe, and determining the total bait casting quantity of the universe according to the area of each grid.

In the description herein, references to the description of the term "one embodiment," "some embodiments," "an illustrative embodiment," "an example," "a specific example," or "some examples" or the like mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

While embodiments of the invention have been shown and described, it will be understood by those of ordinary skill in the art that: various changes, modifications, substitutions and alterations can be made to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.