CN112683814A - Method and system for evaluating aquatic feed based on big data of aquaculture water quality - Google Patents
Method and system for evaluating aquatic feed based on big data of aquaculture water quality Download PDFInfo
- Publication number
- CN112683814A CN112683814A CN202110293244.3A CN202110293244A CN112683814A CN 112683814 A CN112683814 A CN 112683814A CN 202110293244 A CN202110293244 A CN 202110293244A CN 112683814 A CN112683814 A CN 112683814A
- Authority
- CN
- China
- Prior art keywords
- closed area
- spectral image
- gradient
- experimental
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
- Farming Of Fish And Shellfish (AREA)
Abstract
The invention discloses a method and a system for evaluating aquatic feed based on big data of aquaculture water quality, which can objectively, quickly, conveniently, continuously and accurately detect the pollution of feed to the aquaculture water quality by quickly comparing the edge lines of an experimental spectral image sequence and a reference spectral image sequence of feed to be detected and standard feed diffused in an experimental pond, calculating the corresponding gradient value and quickly analyzing the peak value of big data. The influence of uncontrollable factors such as the ingestion speed of aquatic animals on the feed and the degradation speed of pollution sources in the feed on feed evaluation in practical application is avoided, so that the quality of the aquatic feed is dynamically judged and measured.
Description
Technical Field
The disclosure belongs to the technical field of big data, computer image processing technology and aquatic feed detection, and particularly relates to a method and a system for evaluating aquatic feed based on aquaculture water quality big data.
Background
Aquaculture water quality is a key to influencing animal growth and health and determining the economic benefits of aquaculture. The main indexes for evaluating the culture water quality comprise: pH value (pH), dissolved oxygen content, ammonia nitrogen content, nitrite content, total phosphorus content, total nitrogen content, heavy metal content and the like. In the prior art, two methods, namely chemical detection and instrument detection, are generally adopted, and the chemical detection method adopts a commercial water quality detection kit (such as a water quality detection kit series) to judge the water quality index content through the final water color change by the chemical reaction between a chemical reagent in the kit and a specific component in a water body. The defects are that the detection range is limited, the detection value is not accurate, the sampling amount is not representative, and time and labor are wasted. The instrument detection method is characterized in that a sensor is adopted to detect water quality, time and labor are saved relatively, but the instrument is often calibrated, the general accuracy is not high, the error is large, and the repeatability is poor.
The environmental protection and quality of the feed can be objectively evaluated through water quality indexes, and the two methods are generally adopted at present. The feed is put into a water body to feed aquatic animals, a water body sample at a specific time point or time period is collected, and the influence of the feed on the water quality is judged according to the detected water quality index. Due to the limitations of the above detection methods, such as low precision, small sample size, time and labor consumption, the evaluation result of the feed is often too extensive and has a large defect. And the water quality index of the fishpond is dynamically changed and is often influenced by weather conditions, air temperature, air pressure and animal activities in a water body, and the water quality cannot be scientifically predicted according to a detection result within a certain time or a short time, so that the feed evaluation is difficult.
The existing feed inspection mainly aims to improve the biological value of protein, so that an ideal protein mode is adopted, the balance condition of various amino acids in the protein is improved, the biological utilization value of the protein can be improved, the level of crude protein in the feed is effectively reduced, the utilization rate of nitrogen in the feed is improved, and the excretion of nitrogen in excrement is reduced. Thus, a large amount of natural protein feed resources can be saved, and the nitrogen pollution degree of the intensive aquaculture to the environment can be reduced.
Because the demand of aquatic animals for nutrients such as protein and the like is considered in a single aspect, the pollution of nitrogen and phosphorus emission of the feed can be caused by adding a large amount of fish meal or animal protein raw materials in improper proportion in the compound feed. How to detect out the pollutant in the fodder, for example nitrogen, phosphorus content and standard fodder compare whether exceed standard, direct chemical detection is wasted time and energy, and can't be fast, sustainability, accurately detect the pollution of fodder to quality of water. In practical application, factors such as feed intake speed of aquatic animals to feed and degradation speed of pollution sources in the feed are uncontrollable, and the quality of the aquatic feed cannot be measured according to direct chemical indexes.
Disclosure of Invention
The invention aims to provide a method for evaluating aquatic feed based on big data of aquaculture water quality, which solves one or more technical problems in the prior art and provides at least one beneficial choice or creation condition.
And (3) testing environment: 2 aquaculture ponds (5-20 mu culture ponds) with the same size are arranged as an experimental pond and a reference pond, wherein the experimental pond and the reference pond are used for stocking fish fries with the same density (taking tilapia as an example, the standard is 8, and the density is 2000 plus 3000 tails per mu), and the water depth and the water quality of the experimental pond and the reference pond are the same; equal amounts of aquatic feed to be detected (newly configured fish feed to be detected) and standard aquatic feed (commercial tilapia feed No. 0) are taken and fed at feeding points of an experimental pond and a reference pond. The daily feeding amount of the feed is 3% -5% of the weight of the fry, and the feed is fed once in the morning and at night.
To achieve the above objects, according to one aspect of the present disclosure, there is provided a method for evaluating aquaculture feed based on aquaculture water quality big data, the method comprising the steps of:
after the aquatic feed to be detected and the standard aquatic feed are fed at the feeding points of the experimental tank and the reference tank respectively, the following steps are carried out:
s100, synchronously acquiring spectral images of the water surfaces of the experimental pool and the reference pool by spectral image acquisition equipment arranged above feeding points of the experimental pool and the reference pool, respectively recording the spectral images as experimental spectral images and reference spectral images, taking the time for synchronously acquiring the spectral images for the first time as T0, and acquiring the experimental spectral images and the reference spectral images at intervals of time TG to obtain experimental spectral image sequences and reference spectral image sequences; wherein the initial value of the time interval TG is set to [5, 60] minutes;
s200, respectively filtering the experimental spectral image sequence and the reference spectral image in the reference spectral image sequence, then carrying out binarization, carrying out edge detection by an edge detection operator,
thereby obtaining a first closed area formed by the edge line of the outermost layer in the experimental spectral image through edge detection, obtaining a second closed area formed by the edge line of the outermost layer in the reference spectral image through edge detection,
if a third closed area formed by a third edge line appears in the first closed area or the second closed area for the first time, jumping to the step S300; if a third closed area formed by a third edge line does not exist in the first closed area or the second closed area, jumping to the step S100; the third edge line is formed by edge lines obtained by performing edge detection in any closed area of the first closed area or the second closed area;
s300, keeping the time when the third closed region appears for the first time as T1, and modifying and setting a time interval TG as TG = (T1-T0)/n, namely the acquisition time length (T1-T0) from the start of acquiring the spectral images to the time when the third closed region appears to be divided by n, namely the difference value of T1 and T0, wherein n is the number of experimental spectral images in the experimental spectral image sequence or the number of reference spectral images in the reference spectral image sequence; (the third closed area at the moment of T1 correspondingly shows that the pollutant diffusion phenomenon begins to occur in one of the aquatic feed to be detected and the standard aquatic feed, and the time interval of the time for acquiring the spectral images is modified to enable the number of the acquired images to be less, so that the overall detection speed of the system is accelerated);
s400, if the third closed area appears in the second closed area first, the aquatic feed to be detected is judged to be qualified, and the step is finished; if the third closed area first appears in the first closed area, the step S500 is switched to the next test; (the physical meaning of the method is that the standard aquatic feed pollutants in the reference pool are diffused more quickly, and the aquatic feed to be detected does not pollute the water body);
s500, if a fourth closed area formed by a fourth edge line appears in the second closed area, recording the appearance time of the fourth closed area as T2; respectively calculating diffusion gradients of a first closed area and a third closed area, and a second closed area and a fourth closed area to obtain a first gradient and a second gradient, wherein the fourth closed area is a closed area formed by edge lines newly generated in the second closed area obtained by performing edge detection in the newly acquired reference spectral image;
s600, circularly executing the steps S100-S500, and calculating to obtain each first gradient and each second gradient of the experimental spectral image and the reference spectral image which are shot at the synchronous time in the experimental spectral image sequence and the reference spectral image sequence; when the changes of the first gradient and the second gradient are changed from increasing to decreasing, or the first gradient and the second gradient are the same in a last continuous time period (T2-T1), selecting the maximum value of the first gradient corresponding to each experimental spectral image in the experimental spectral image sequence as a first peak value (maximum first gradient value), selecting the maximum value of the second gradient corresponding to each reference spectral image in the reference spectral image sequence as a second peak value (maximum second gradient value), and going to step S700 (indicating that the pollutant is changed from a diffusion trend to a contraction trend or has been diffused without degradation phenomenon); wherein (T2-T1) is the time length from the first occurrence of the third closed region to the occurrence of the fourth closed region in the reference spectral image sequence, i.e. the difference between T2 and T1;
s700, when the first peak value is smaller than the second peak value, the aquatic feed to be detected is judged to be qualified, otherwise, the aquatic feed to be detected is judged to be unqualified.
Further, in S100, the feeding points are: any point in the experimental pool and the reference pool which is 0.5 to 2 meters away from the shore on the water surface.
Further, in S100, the spectral image collecting device includes any one of an imaging hyperspectral spectrometer, a fiber optic spectrometer, and a multispectral camera.
Further, in S100, the experimental spectrum image and the reference spectrum image are images obtained by inverting any one of physical quantities of a dissolved oxygen content, an ammonia nitrogen content, a nitrite content, a total phosphorus content, a total nitrogen content, and a heavy metal content of the hyperspectral remote sensing image or the spectrum image through any one of a water quality inversion model, a single-band remote sensing model, a band combination remote sensing model, or a least square support vector machine model disclosed in CN 109557030A.
Further, in S500 and S600, the method of calculating the diffusion gradients of the first closed region and the third closed region is:
calculating a diffusion gradient by a calculation formula of G1= | B1-A1| ÷ H1, wherein a G1 value is a concentration diffusion gradient value, or diffusion gradient value, of the corresponding water body in an interval between the edge line of the first closed area and the edge line of the third closed area in the first closed area;
the calculation method of B1 is as follows:
calculating the average value of Euclidean distances from each point on the edge line of the third closed area to the feeding point as the average value of the diffusion radius of the third closed area, and forming a circular corresponding circular arc line in the experimental spectrum image by taking the feeding point as the center of a circle and the average value of the diffusion radius of the third closed area as the radius to obtain a circular arc line X1; (namely, making an arc on the experimental spectrum image by using the circle center and the radius, wherein the arc part on the experimental spectrum image is an arc line);
obtaining an arithmetic mean value of the concentration of the pollutants in the water domain at the corresponding position of each pixel point on the circular arc line X1 as B1 or calculating an arithmetic mean value of the gray value of the corresponding pixel point on the circular arc line X1 in the experimental spectral image as B1 by inverting the corresponding position of the circular arc of the experimental spectral image through the inversion model;
the calculation method of A1 is as follows: (same as B1 calculation method)
Calculating the average value of Euclidean distances from a first closed area edge line to a feeding point as the average value of the diffusion radius of the first closed area, and forming a circular corresponding circular arc line in an experimental spectrum image by taking the feeding point as the center of a circle and the average value of the diffusion radius of the first closed area as the radius to obtain a circular arc line Y1;
obtaining an arithmetic mean value of the concentration of the pollutants in the water domain at the corresponding position of each pixel point on the circular arc line Y1 as A1 by inverting the corresponding position of the circular arc of the experimental spectrum image through the inversion model, or calculating an arithmetic mean value of the gray value of the corresponding pixel point on the circular arc line Y1 in the experimental spectrum image as A1;
where H1 is the distance between the arc line X1 and the arc line Y1 (the absolute value of the difference between the two closest points on X1 and Y1).
Wherein, the inversion model is: any physical quantity of dissolved oxygen, ammonia nitrogen content, nitrite content, total phosphorus content, total nitrogen content and heavy metal content is determined by any one of a water quality inversion model, a single-waveband remote sensing model, a waveband combination remote sensing model or a least square support vector machine model in the publication number CN 109557030A.
Further, in S500 and S600, the method of calculating the diffusion gradients of the second and fourth closed regions is:
calculating a diffusion gradient by a calculation formula of G2= | B2-A2| ÷ H2, wherein the G2 value is a concentration diffusion gradient value, or a diffusion gradient value, of the corresponding water body in an interval between the edge line of the second closed area and the edge line of the fourth closed area in the second closed area;
the calculation method of B2 is as follows:
calculating the average value of Euclidean distances from each point on the edge line of a fourth closed area to a feeding point as the average value of diffusion radius of the fourth closed area, and forming a circular corresponding circular arc line in a reference spectrum image by taking the feeding point as the center of a circle and the average value of the diffusion radius of the fourth closed area as the radius to obtain a circular arc line X2; (namely, making an arc on the reference spectrum image by using the circle center and the radius, wherein the arc part on the reference spectrum image is an arc line);
inverting the corresponding position of the circular arc of the reference spectrum image through the inversion model to obtain an arithmetic mean value of the concentration of the pollutants in the water domain of the corresponding position of each pixel point on the circular arc line X2 as B2, or calculating an arithmetic mean value of the gray value of the corresponding pixel point on the circular arc line X2 and the reference spectrum image as B2;
the calculation method of A2 is as follows: (same as B2 calculation method)
Calculating the average value of Euclidean distances from a second closed area edge line to a feeding point as the average value of second closed area diffusion radius, and forming a circular corresponding circular arc line in a reference spectrum image by taking the feeding point as the center of a circle and the average value of the second closed area diffusion radius as the radius to obtain a circular arc line Y2;
inverting the corresponding position of the circular arc of the reference spectrum image through the inversion model to obtain an arithmetic mean value of the concentration of the pollutants in the water domain of the corresponding position of each pixel point on the circular arc line Y2 as A2, or calculating an arithmetic mean value of the gray value of the corresponding pixel point on the circular arc line Y2 in the reference spectrum image as A2;
where H2 is the distance between the arc line X2 and the arc line Y2 (the absolute value of the difference between the two closest points on X2 and Y2).
Wherein, the inversion model is: any physical quantity of dissolved oxygen, ammonia nitrogen content, nitrite content, total phosphorus content, total nitrogen content and heavy metal content is determined by any one of a water quality inversion model, a single-waveband remote sensing model, a waveband combination remote sensing model or a least square support vector machine model in the publication number CN 209557030A.
Further, in S600, the method for determining that the changes of the first gradient and the second gradient are both changed from increasing to decreasing is: in the experimental spectral image sequence and the reference spectral image sequence, the shooting time of the latest acquired experimental spectral image and reference spectral image is T3, and the first gradient and the second gradient of the experimental spectral image and reference spectral image acquired at T3 are respectively V1 and V2; let the first and second gradients of the experimental spectral image and the reference spectral image collected at T3-TG be V3 and V4, respectively; if state 1 of the first and second gradients transitions to state 2, the change in both the first and second gradients transitions from increasing to decreasing, wherein,
the state 1 is: v1 is more than or equal to V3 or V2 is more than or equal to V4;
the state 2 is: v3 > V1 and V4 > V2.
Further, in S600, the method of determining whether the first gradient and the second gradient value are the same in the last consecutive (T2-T1) period is:
s601, taking the shooting time of the newly collected experimental spectral image and the reference spectral image as T3; calculating a sliding quantity m = (T2-T1)/TG, and taking n as the number of experimental spectral images in the experimental spectral image sequence or the number of reference spectral images in the reference spectral image sequence; setting a variable i with an initial value of 0 and setting a variable j with an initial value of 0;
s602, if i is less than m, comparing whether the first gradient value of the n-i-j test spectrum image in the test spectrum image sequence is equal to the second gradient value of the n-i-j reference spectrum image in the reference spectrum image sequence, if so, increasing the value of j by 1, and going to the step S603; if not, setting the value of j to 0 and increasing the value of i by 1 and going to step S603;
s603, if i is larger than or equal to m, judging that the first gradient and the second gradient in the latest continuous (T2-T1) time period are different, and finishing the step;
s604, if j is less than m, go to step S602; if j is larger than or equal to m, judging that the first gradient and the second gradient in the latest continuous time period (T2-T1) are the same, and ending the step.
The invention also provides a system for evaluating aquatic feed based on big data of aquaculture water quality, which comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the computer program to run in the units of the following system:
the spectral image synchronous acquisition unit is used for starting to synchronously acquire spectral images of the water surfaces of the experimental pool and the reference pool through spectral image acquisition equipment arranged above feeding points of the experimental pool and the reference pool, the spectral images are respectively marked as experimental spectral images and reference spectral images, the time for starting to synchronously acquire the spectral images for the first time is T0, and the experimental spectral images and the reference spectral images are acquired at intervals of time TG to obtain an experimental spectral image sequence and a reference spectral image sequence;
the spectral image edge detection unit is used for respectively filtering the experimental spectral image sequence and the reference spectral image in the reference spectral image sequence, then carrying out binarization and carrying out edge detection by an edge detection operator, so that a first closed area formed by outermost edge lines in the experimental spectral image sequence is obtained through edge detection, a second closed area formed by outermost edge lines in the reference spectral image is obtained through edge detection, and if a third closed area formed by third edge lines appears in the first closed area or the second closed area for the first time, the acquisition time interval correction unit is skipped; if a third closed area formed by a third edge line does not exist in the first closed area or the second closed area, jumping to a spectral image synchronous acquisition unit;
an acquisition time interval correction unit, configured to note that the time when the third closed region first appears is T1, and modify the set time interval TG to TG = (T1-T0)/n, that is, the acquisition time duration from the start of acquiring the spectral image to the time when the third closed region appears (T1-T0) is divided by n, that is, the difference between T1 and T0;
the aquatic feed initial detection unit is used for judging that the aquatic feed to be detected is qualified and finishing the step if the third closed area first appears in the second closed area; if the third closed area first appears in the first closed area, turning to a gradient trend calculation unit;
a gradient tendency calculation unit for recording the time of occurrence of a fourth closed region composed of a fourth edge line as T2 if the fourth closed region occurs inside the second closed region; respectively calculating diffusion gradients of the first closed area, the third closed area, the second closed area and the fourth closed area to obtain a first gradient and a second gradient;
the peak value selection unit is used for circularly executing the spectral image synchronous acquisition unit to the gradient trend calculation unit, and calculating to obtain each first gradient and each second gradient of the experimental spectral image and the reference spectral image which are shot at the synchronous time in the experimental spectral image sequence and the reference spectral image sequence; when the changes of the first gradient and the second gradient are changed from increasing to decreasing, or the first gradient and the second gradient are the same in the latest continuous time period (T2-T1), selecting the maximum value of the first gradient corresponding to each experimental spectral image in the experimental spectral image sequence as a first peak value, selecting the maximum value of the second gradient corresponding to each reference spectral image in the reference spectral image sequence as a second peak value, and transferring to an aquatic feed qualification judging unit;
and the aquatic feed qualification judging unit is used for judging that the aquatic feed to be detected is qualified when the first peak value is smaller than the second peak value, and otherwise, judging that the aquatic feed to be detected is unqualified.
The beneficial effect of this disclosure does: the invention provides a method and a system for evaluating aquatic feed based on big data of aquaculture water quality, which can objectively, quickly, conveniently, continuously and accurately detect the pollution of the feed to the water quality of a water body by quickly comparing diffusion images of the feed to be detected and standard feed in an experimental pond. The influence of uncontrollable factors such as the ingestion speed of aquatic animals on the feed and the degradation speed of pollution sources in the feed on feed evaluation in practical application is avoided, so that the quality of the aquatic feed is dynamically judged and measured.
Drawings
The foregoing and other features of the present disclosure will become more apparent from the detailed description of the embodiments shown in conjunction with the drawings in which like reference characters designate the same or similar elements throughout the several views, and it is apparent that the drawings in the following description are merely some examples of the present disclosure and that other drawings may be derived therefrom by those skilled in the art without the benefit of any inventive faculty, and in which:
FIG. 1 is a flow chart of a method for evaluating aquaculture feed based on aquaculture water quality big data;
FIG. 2 is a diagram showing a system for evaluating aquatic feed based on big data of aquaculture water quality.
Detailed Description
The conception, specific structure and technical effects of the present disclosure will be clearly and completely described below in conjunction with the embodiments and the accompanying drawings to fully understand the objects, aspects and effects of the present disclosure. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.
In the description of the present invention, a plurality of means is one or more, a plurality of means is two or more, and greater than, less than, more than, etc. are understood as excluding the present numbers, and greater than, less than, more than, etc. are understood as including the present numbers, and outer and inner are understood as relative inside-outside relationships. If the first and second are described for the purpose of distinguishing technical features, they are not to be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated or implicitly indicating the precedence of the technical features indicated.
Referring to fig. 1, a flow chart of a method for evaluating aquaculture feed based on aquaculture water quality big data according to the present invention is shown, and a method for evaluating aquaculture feed based on aquaculture water quality big data according to an embodiment of the present invention is described below with reference to fig. 1.
The invention provides a method for evaluating aquatic feed based on big data of aquaculture water quality, which specifically comprises the following steps:
and (3) testing environment: 2 aquaculture ponds (5-20 mu culture ponds) with the same size are arranged as an experimental pond and a reference pond, wherein the experimental pond and the reference pond are used for stocking fish fries with the same density (taking tilapia as an example, the standard is 8, and the density is 2000 plus 3000 tails per mu), and the water depth and the water quality of the experimental pond and the reference pond are the same; and (3) taking the same amount of aquatic feed to be detected and standard aquatic feed (commercial tilapia feed No. 0) and feeding the same at feeding points of the experimental pond and the reference pond. The daily feeding amount of the feed is 3% -5% of the weight of the fry, and the feed is fed once in the morning and at night. (physical principle: because the diffusion rate of each pollutant in water is different, on the continuous spectrum image, the interference of water flow and the like is eliminated, and various pollutants are shown as concentric circles taking a feeding point as the center of a circle on the graph according to the speed of diffusion in water); the concentric circles will later on change from a tendency to diffuse to a tendency to contract due to fish feeding and degradation of contaminants).
S100, synchronously acquiring spectral images of the water surfaces of the experimental pool and the reference pool by spectral image acquisition equipment arranged above feeding points of the experimental pool and the reference pool, respectively recording the spectral images as experimental spectral images and reference spectral images, taking the time for synchronously acquiring the spectral images for the first time as T0, and acquiring the experimental spectral images and the reference spectral images at intervals of time TG to obtain experimental spectral image sequences and reference spectral image sequences; wherein the initial value of the time interval TG is set to [5, 60] minutes;
s200, respectively filtering the experimental spectral image sequence and the reference spectral image in the reference spectral image sequence, then carrying out binarization, carrying out edge detection by a Sobel or Robert edge detection operator,
if the edge detection and the edge line mentioned in the text are not specially stated, the edge line is extracted from the image by using a Sobel or Robert edge detection operator;
thereby obtaining a first closed area formed by the edge line of the outermost layer in the experimental spectral image through edge detection, obtaining a second closed area formed by the edge line of the outermost layer in the reference spectral image through edge detection,
if a third closed area formed by a third edge line appears in the first closed area or the second closed area for the first time, jumping to the step S300; if a third closed area formed by a third edge line does not exist in the first closed area or the second closed area, jumping to the step S100; the third edge line is formed by edge lines obtained by performing edge detection in any closed area of the first closed area or the second closed area;
s300, recording the time when the third closed region appears for the first time as T1, and modifying and setting a time interval TG as TG = (T1-T0)/n, namely dividing the acquisition time (T1-T0) from the start of acquiring the spectral images to the appearance of the third closed region by n, wherein n is the number of experimental spectral images in the experimental spectral image sequence or the number of reference spectral images in the reference spectral image sequence; (the third closed area at the moment of T1 correspondingly shows that the pollutant diffusion phenomenon begins to occur in one of the aquatic feed to be detected and the standard aquatic feed, and the time interval of the time for acquiring the spectral images is modified to enable the number of the acquired images to be less, so that the overall detection speed of the system is accelerated);
(in an ideal spectral image the spectrum appears as concentric circles, e.g., where phosphorus diffuses faster than nitrogen in water, the intersection of the inner and outer circles is phosphorus and nitrogen, and the outer circle, which is not the intersection, is phosphorus only);
s400, if the third closed area appears in the second closed area first, the aquatic feed to be detected is judged to be qualified; if the third closed area first appears in the first closed area, the step S500 is switched to carry out the next test; (indicating that the standard aquatic feed in the reference pool diffuses more quickly and the aquatic feed to be detected does not pollute the water body);
s500, if a fourth closed area formed by a fourth edge line appears in the second closed area, recording the appearance time of the fourth closed area as T2; respectively calculating diffusion gradients of a first closed area and a third closed area, and a second closed area and a fourth closed area to obtain a first gradient and a second gradient, wherein the fourth closed area is a closed area formed by edge lines newly generated in the second closed area obtained by performing edge detection in the newly acquired reference spectral image;
s600, circularly executing the steps S100 to S500, and calculating each first gradient and each second gradient of the experimental spectral image and the reference spectral image which are shot at the synchronous time in the experimental spectral image sequence and the reference spectral image sequence; when the changes of the first gradient and the second gradient are changed from increasing to decreasing, or the first gradient and the second gradient are the same in a last continuous time period (T2-T1), selecting the maximum value of the first gradient corresponding to each experimental spectral image in the experimental spectral image sequence as a first peak value, selecting the maximum value of the second gradient corresponding to each reference spectral image in the reference spectral image sequence as a second peak value, and going to step S700 (indicating that the pollutant is changed from a diffusion trend to a contraction trend or has diffused without degradation); wherein (T2-T1) is the time period from the first occurrence of the third occlusion region to the occurrence of the fourth occlusion region in the sequence of reference spectral images;
s700, when the first peak value is smaller than the second peak value, the aquatic feed to be detected is judged to be qualified, otherwise, the aquatic feed to be detected is judged to be unqualified.
Further, in S100, the feeding points are: any point in the experimental pool and the reference pool which is 0.5 to 2 meters away from the shore on the water surface.
Further, in S100, the spectral image collecting device includes any one of an imaging hyperspectral spectrometer, a fiber optic spectrometer, and a multispectral camera.
Further, in S100, the experimental spectrum image and the reference spectrum image are images obtained by inverting any one of physical quantities of a dissolved oxygen content, an ammonia nitrogen content, a nitrite content, a total phosphorus content, a total nitrogen content, and a heavy metal content of the hyperspectral remote sensing image or the spectrum image through any one of a water quality inversion model, a single-band remote sensing model, a band combination remote sensing model, or a least square support vector machine model disclosed in CN 109557030A.
Further, in S500 and S600, the method of calculating the diffusion gradients of the first closed region and the third closed region is:
calculating a diffusion gradient by a calculation formula of G1= | B1-A1| ÷ H1, wherein a G1 value is a concentration diffusion gradient value, or diffusion gradient value, of the corresponding water body in an interval between the edge line of the first closed area and the edge line of the third closed area in the first closed area;
the calculation method of B1 is as follows:
calculating the average value of Euclidean distances from each point on the edge line of the third closed area to the feeding point as the average value of the diffusion radius of the third closed area, and forming a circular corresponding circular arc line in the experimental spectrum image by taking the feeding point as the center of a circle and the average value of the diffusion radius of the third closed area as the radius to obtain a circular arc line X1;
obtaining an arithmetic mean value of the concentration of the pollutants in the water domain at the corresponding position of each pixel point on the circular arc line X1 as B1 or calculating an arithmetic mean value of the gray value of the corresponding pixel point on the circular arc line X1 in the experimental spectral image as B1 by inverting the corresponding position of the circular arc of the experimental spectral image through the inversion model;
the calculation method of A1 is as follows: (same as B1 calculation method)
Calculating the average value of Euclidean distances from a first closed area edge line to a feeding point as the average value of the diffusion radius of the first closed area, and forming a circular corresponding circular arc line in an experimental spectrum image by taking the feeding point as the center of a circle and the average value of the diffusion radius of the first closed area as the radius to obtain a circular arc line Y1;
obtaining an arithmetic mean value of the concentration of the pollutants in the water domain at the corresponding position of each pixel point on the circular arc line Y1 as A1 by inverting the corresponding position of the circular arc of the experimental spectrum image through the inversion model, or calculating an arithmetic mean value of the gray value of the corresponding pixel point on the circular arc line Y1 in the experimental spectrum image as A1;
where H1 is the distance (absolute value of the difference) between the circular arc line X1 and the circular arc line Y1.
Wherein, the inversion model is: any physical quantity of dissolved oxygen, ammonia nitrogen content, nitrite content, total phosphorus content, total nitrogen content and heavy metal content is determined by any one of a water quality inversion model, a single-waveband remote sensing model, a waveband combination remote sensing model or a least square support vector machine model in the publication number CN 109557030A.
Further, in S500 and S600, the method of calculating the diffusion gradients of the second and fourth closed regions is:
calculating a diffusion gradient by a calculation formula of G2= | B2-A2| ÷ H2, wherein the G2 value is a concentration diffusion gradient value, or a diffusion gradient value, of the corresponding water body in an interval between the edge line of the second closed area and the edge line of the fourth closed area in the second closed area;
the calculation method of B2 is as follows:
calculating the average value of Euclidean distances from each point on the edge line of a fourth closed area to a feeding point as the average value of diffusion radius of the fourth closed area, and forming a circular corresponding circular arc line in a reference spectrum image by taking the feeding point as the center of a circle and the average value of the diffusion radius of the fourth closed area as the radius to obtain a circular arc line X2;
inverting the corresponding position of the circular arc of the reference spectrum image through the inversion model to obtain an arithmetic mean value of the concentration of the pollutants in the water domain of the corresponding position of each pixel point on the circular arc line X2 as B2, or calculating an arithmetic mean value of the gray value of the corresponding pixel point on the circular arc line X2 and the reference spectrum image as B2;
the calculation method of A2 is as follows: (same as B2 calculation method)
Calculating the average value of Euclidean distances from a second closed area edge line to a feeding point as the average value of second closed area diffusion radius, and forming a circular corresponding circular arc line in a reference spectrum image by taking the feeding point as the center of a circle and the average value of the second closed area diffusion radius as the radius to obtain a circular arc line Y2;
inverting the corresponding position of the circular arc of the reference spectrum image through the inversion model to obtain an arithmetic mean value of the concentration of the pollutants in the water domain of the corresponding position of each pixel point on the circular arc line Y2 as A2, or calculating an arithmetic mean value of the gray value of the corresponding pixel point on the circular arc line Y2 in the reference spectrum image as A2;
where H2 is the distance (absolute value of the difference) between the circular arc line X2 and the circular arc line Y2.
Wherein, the inversion model is: any physical quantity of dissolved oxygen, ammonia nitrogen content, nitrite content, total phosphorus content, total nitrogen content and heavy metal content is determined by any one of a water quality inversion model, a single-waveband remote sensing model, a waveband combination remote sensing model or a least square support vector machine model in the publication number CN 209557030A.
Further, in S600, the method for determining that the changes of the first gradient and the second gradient are both changed from increasing to decreasing is: in the experimental spectral image sequence and the reference spectral image sequence, the shooting time of the latest acquired experimental spectral image and reference spectral image is T3, and the first gradient and the second gradient of the experimental spectral image and reference spectral image acquired at T3 are respectively V1 and V2; let the first and second gradients of the experimental spectral image and the reference spectral image collected at T3-TG be V3 and V4, respectively; if state 1 of the first and second gradients transitions to state 2, the change in both the first and second gradients transitions from increasing to decreasing, wherein,
the state 1 is: v1 is more than or equal to V3 or V2 is more than or equal to V4;
the state 2 is: v3 > V1 and V4 > V2.
Further, in S600, the method of determining whether the first gradient and the second gradient value are the same in the last consecutive (T2-T1) period is:
s601, taking the shooting time of the newly collected experimental spectral image and the reference spectral image (namely the last experimental spectral image and the reference spectral image) as T3; calculating a sliding quantity m = (T2-T1)/TG, and taking n as the number of experimental spectral images in the experimental spectral image sequence or the number of reference spectral images in the reference spectral image sequence; setting a variable i with an initial value of 0 and setting a variable j with an initial value of 0;
s602, if i is less than m, comparing whether the first gradient value of the n-i-j test spectrum image in the test spectrum image sequence is equal to the second gradient value of the n-i-j reference spectrum image in the reference spectrum image sequence, if so, increasing the value of j by 1, and going to the step S603; if not, setting the value of j to 0 and increasing the value of i by 1 and going to step S603;
s603, if i is larger than or equal to m, judging that the first gradient and the second gradient in the latest continuous (T2-T1) time period are different, and finishing the step;
s604, if j is less than m, go to step S602; if j is larger than or equal to m, judging that the first gradient and the second gradient in the latest continuous time period (T2-T1) are the same, and ending the step.
The system for evaluating aquatic feeds based on the big data of the aquaculture water provided by the embodiment of the disclosure is shown in fig. 2 as a structure diagram of the system for evaluating aquatic feeds based on the big data of the aquaculture water, and the system for evaluating aquatic feeds based on the big data of the aquaculture water of the embodiment comprises: a processor, a memory and a computer program stored in the memory and executable on the processor, the processor when executing the computer program implementing the steps in an embodiment of the system for evaluating aquatic feed based on aquaculture water quality big data as described above.
The system comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the computer program to run in the units of the following system:
the spectral image synchronous acquisition unit is used for starting to synchronously acquire spectral images of the water surfaces of the experimental pool and the reference pool through spectral image acquisition equipment arranged above feeding points of the experimental pool and the reference pool, the spectral images are respectively marked as experimental spectral images and reference spectral images, the time for starting to synchronously acquire the spectral images for the first time is T0, and the experimental spectral images and the reference spectral images are acquired at intervals of time TG to obtain an experimental spectral image sequence and a reference spectral image sequence;
the spectral image edge detection unit is used for respectively filtering the experimental spectral image sequence and the reference spectral image in the reference spectral image sequence, then carrying out binarization and carrying out edge detection by an edge detection operator, so that a first closed area formed by outermost edge lines in the experimental spectral image sequence is obtained through edge detection, a second closed area formed by outermost edge lines in the reference spectral image is obtained through edge detection, and if a third closed area formed by third edge lines appears in the first closed area or the second closed area for the first time, the acquisition time interval correction unit is skipped; if a third closed area formed by a third edge line does not exist in the first closed area or the second closed area, jumping to a spectral image synchronous acquisition unit;
an acquisition time interval correction unit, configured to note that the time when the third closed region first appears is T1, and modify the set time interval TG to TG = (T1-T0)/n, that is, the acquisition time duration from the start of acquiring the spectral image to the time when the third closed region appears (T1-T0) is divided by n, that is, the difference between T1 and T0;
the aquatic feed initial detection unit is used for judging that the aquatic feed to be detected is qualified and finishing the step if the third closed area first appears in the second closed area; if the third closed area first appears in the first closed area, turning to a gradient trend calculation unit;
a gradient tendency calculation unit for recording the time of occurrence of a fourth closed region composed of a fourth edge line as T2 if the fourth closed region occurs inside the second closed region; respectively calculating diffusion gradients of the first closed area, the third closed area, the second closed area and the fourth closed area to obtain a first gradient and a second gradient;
the peak value selection unit is used for circularly executing the spectral image synchronous acquisition unit to the gradient trend calculation unit, and calculating to obtain each first gradient and each second gradient of the experimental spectral image and the reference spectral image which are shot at the synchronous time in the experimental spectral image sequence and the reference spectral image sequence; when the changes of the first gradient and the second gradient are changed from increasing to decreasing, or the first gradient and the second gradient are the same in the latest continuous time period (T2-T1), selecting the maximum value of the first gradient corresponding to each experimental spectral image in the experimental spectral image sequence as a first peak value, selecting the maximum value of the second gradient corresponding to each reference spectral image in the reference spectral image sequence as a second peak value, and transferring to an aquatic feed qualification judging unit;
and the aquatic feed qualification judging unit is used for judging that the aquatic feed to be detected is qualified when the first peak value is smaller than the second peak value, and otherwise, judging that the aquatic feed to be detected is unqualified.
The system for evaluating the aquatic feed based on the big data of the aquaculture water quality can be operated in computing equipment such as a desktop computer, a notebook computer, a palm computer and a cloud server. The system for evaluating the aquatic feed based on the big data of the aquaculture water quality can be operated by comprising, but not limited to, a processor and a memory. It will be understood by those skilled in the art that the example is merely illustrative of a system for evaluating aquaculture feed based on aquaculture water quality big data and does not constitute a limitation of a system for evaluating aquaculture feed based on aquaculture water quality big data, and may include more or less components than the system, or some components in combination, or different components, for example, the system for evaluating aquaculture feed based on aquaculture water quality big data may further include input and output devices, network access devices, buses, etc.
The Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component, or the like. The general processor can be a microprocessor or the processor can be any conventional processor and the like, the processor is a control center of the system operation system for evaluating the aquatic feeds based on the culture water quality big data, and various interfaces and lines are utilized to connect all parts of the whole system operable system for evaluating the aquatic feeds based on the culture water quality big data.
The memory can be used for storing the computer program and/or the module, and the processor realizes various functions of the system for evaluating the aquatic feed based on the big data of the aquaculture water quality by operating or executing the computer program and/or the module stored in the memory and calling the data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data (such as audio data, a phonebook, etc.) created according to the use of the cellular phone, and the like. In addition, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.
Although the description of the present disclosure has been rather exhaustive and particularly described with respect to several illustrated embodiments, it is not intended to be limited to any such details or embodiments or any particular embodiments, so as to effectively encompass the intended scope of the present disclosure. Furthermore, the foregoing describes the disclosure in terms of embodiments foreseen by the inventor for which an enabling description was available, notwithstanding that insubstantial modifications of the disclosure, not presently foreseen, may nonetheless represent equivalent modifications thereto.
Claims (7)
1. A method for evaluating aquatic feed based on big data of aquaculture water quality is characterized by comprising the following steps:
s100, synchronously acquiring spectral images of the water surfaces of the experimental pool and the reference pool by spectral image acquisition equipment arranged above feeding points of the experimental pool and the reference pool, respectively recording the spectral images as experimental spectral images and reference spectral images, taking the time for synchronously acquiring the spectral images for the first time as T0, and acquiring the experimental spectral images and the reference spectral images at intervals of time TG to obtain experimental spectral image sequences and reference spectral image sequences;
s200, respectively filtering the experimental spectral image sequence and the reference spectral image in the reference spectral image sequence, then carrying out binarization, carrying out edge detection by an edge detection operator,
thereby obtaining a first closed area formed by the edge line of the outermost layer in the experimental spectral image through edge detection, obtaining a second closed area formed by the edge line of the outermost layer in the reference spectral image through edge detection,
if a third closed area formed by a third edge line appears in the first closed area or the second closed area for the first time, jumping to the step S300; if a third closed area formed by a third edge line does not exist in the first closed area or the second closed area, jumping to the step S100; the third edge line is formed by edge lines obtained by performing edge detection in any closed area of the first closed area or the second closed area;
s300, recording the time when the third closed region appears for the first time as T1, and modifying the set time interval TG to TG = (T1-T0)/n, namely dividing the acquisition time length (T1-T0) from the start of acquiring the spectral image to the appearance of the third closed region by n; wherein n is the number of experimental spectral images in the experimental spectral image sequence or the number of reference spectral images in the reference spectral image sequence;
s400, if the third closed area appears in the second closed area first, the aquatic feed to be detected is judged to be qualified, and the step is finished; if the third closed area first appears in the first closed area, the step S500 is switched to the next test;
s500, if a fourth closed area formed by a fourth edge line appears in the second closed area, recording the appearance time of the fourth closed area as T2; respectively calculating diffusion gradients of the first closed area, the third closed area, the second closed area and the fourth closed area to obtain a first gradient and a second gradient; the fourth closed region is a closed region formed by newly generated edge lines in the second closed region obtained by performing edge detection in the newly acquired reference spectral image;
s600, circularly executing the steps S100-S500, and calculating to obtain each first gradient and each second gradient of the experimental spectral image and the reference spectral image which are shot at the synchronous time in the experimental spectral image sequence and the reference spectral image sequence; when the changes of the first gradient and the second gradient are changed from increasing to decreasing, or the first gradient and the second gradient are the same in a latest continuous time period (T2-T1), selecting the maximum value of the first gradient corresponding to each experimental spectral image in the experimental spectral image sequence as a first peak value, selecting the maximum value of the second gradient corresponding to each reference spectral image in the reference spectral image sequence as a second peak value, and going to step S700;
s700, when the first peak value is smaller than the second peak value, the aquatic feed to be detected is judged to be qualified, otherwise, the aquatic feed to be detected is judged to be unqualified.
2. The method for evaluating the aquatic feed based on the big data of the aquaculture water quality as recited in claim 1, wherein in S100, the spectral image acquisition device comprises any one of an imaging high-speed spectrometer, a fiber-optic spectrometer and a multispectral camera.
3. The method for evaluating aquatic feed based on big data of aquaculture water of claim 1, wherein in S500 and S600, the method for calculating the diffusion gradient of the first closed area and the third closed area comprises:
calculating a diffusion gradient by a calculation formula of G1= | B1-A1| ÷ H1, wherein a G1 value is a concentration diffusion gradient value, or diffusion gradient value, of the corresponding water body in an interval between the edge line of the first closed area and the edge line of the third closed area in the first closed area;
the calculation method of B1 is as follows:
calculating the average value of Euclidean distances from each point on the edge line of the third closed area to the feeding point as the average value of the diffusion radius of the third closed area, and forming a circular corresponding circular arc line in the experimental spectrum image by taking the feeding point as the center of a circle and the average value of the diffusion radius of the third closed area as the radius to obtain a circular arc line X1;
obtaining an arithmetic mean value of the concentration of the pollutants in the water domain at the corresponding position of each pixel point on the circular arc line X1 as B1 or calculating an arithmetic mean value of the gray value of the corresponding pixel point on the circular arc line X1 in the experimental spectral image as B1 by inverting the corresponding position of the circular arc of the experimental spectral image through the inversion model;
the calculation method of A1 is as follows:
calculating the average value of Euclidean distances from a first closed area edge line to a feeding point as the average value of the diffusion radius of the first closed area, and forming a circular corresponding circular arc line in an experimental spectrum image by taking the feeding point as the center of a circle and the average value of the diffusion radius of the first closed area as the radius to obtain a circular arc line Y1;
obtaining an arithmetic mean value of the concentration of the pollutants in the water domain at the corresponding position of each pixel point on the circular arc line Y1 as A1 by inverting the corresponding position of the circular arc of the experimental spectrum image through the inversion model, or calculating an arithmetic mean value of the gray value of the corresponding pixel point on the circular arc line Y1 in the experimental spectrum image as A1;
wherein H1 is the distance between arc line X1 and arc line Y1.
4. The method for evaluating aquatic feed based on big data of aquaculture water quality as claimed in claim 1, wherein in S500 and S600, the method for calculating the diffusion gradient of the second closed area and the fourth closed area comprises:
calculating a diffusion gradient by a calculation formula of G2= | B2-A2| ÷ H2, wherein the G2 value is a concentration diffusion gradient value, or a diffusion gradient value, of the corresponding water body in an interval between the edge line of the second closed area and the edge line of the fourth closed area in the second closed area;
the calculation method of B2 is as follows:
calculating the average value of Euclidean distances from each point on the edge line of a fourth closed area to a feeding point as the average value of diffusion radius of the fourth closed area, and forming a circular corresponding circular arc line in a reference spectrum image by taking the feeding point as the center of a circle and the average value of the diffusion radius of the fourth closed area as the radius to obtain a circular arc line X2;
inverting the corresponding position of the circular arc of the reference spectrum image through the inversion model to obtain an arithmetic mean value of the concentration of the pollutants in the water domain of the corresponding position of each pixel point on the circular arc line X2 as B2, or calculating an arithmetic mean value of the gray value of the corresponding pixel point on the circular arc line X2 and the reference spectrum image as B2;
the calculation method of A2 is as follows:
calculating the average value of Euclidean distances from a second closed area edge line to a feeding point as the average value of second closed area diffusion radius, and forming a circular corresponding circular arc line in a reference spectrum image by taking the feeding point as the center of a circle and the average value of the second closed area diffusion radius as the radius to obtain a circular arc line Y2;
inverting the corresponding position of the circular arc of the reference spectrum image through the inversion model to obtain an arithmetic mean value of the concentration of the pollutants in the water domain of the corresponding position of each pixel point on the circular arc line Y2 as A2, or calculating an arithmetic mean value of the gray value of the corresponding pixel point on the circular arc line Y2 in the reference spectrum image as A2;
wherein H2 is the distance between arc line X2 and arc line Y2.
5. The method for evaluating aquatic feed based on big data of aquaculture water quality as claimed in claim 1, wherein in S600, the method for judging whether the changes of the first gradient and the second gradient are both gradually increased or decreased is as follows: in the experimental spectral image sequence and the reference spectral image sequence, the shooting time of the latest acquired experimental spectral image and reference spectral image is T3, and the first gradient and the second gradient of the experimental spectral image and reference spectral image acquired at T3 are respectively V1 and V2; let the first and second gradients of the experimental spectral image and the reference spectral image collected at T3-TG be V3 and V4, respectively; if state 1 of the first and second gradients transitions to state 2, the change in both the first and second gradients transitions from increasing to decreasing, wherein,
the state 1 is: v1 is more than or equal to V3 or V2 is more than or equal to V4;
the state 2 is: v3 > V1 and V4 > V2.
6. The method for evaluating aquatic feed based on big data of aquaculture water quality as claimed in claim 1, wherein in S600, the method for determining whether the first gradient and the second gradient are the same in the last continuous time period (T2-T1) is:
s601, taking the shooting time of the newly collected experimental spectral image and the reference spectral image as T3; calculating a sliding quantity m = (T2-T1)/TG, and taking n as the number of experimental spectral images in the experimental spectral image sequence or the number of reference spectral images in the reference spectral image sequence; setting a variable i with an initial value of 0 and setting a variable j with an initial value of 0;
s602, if i is less than m, comparing whether the first gradient value of the n-i-j test spectrum image in the test spectrum image sequence is equal to the second gradient value of the n-i-j reference spectrum image in the reference spectrum image sequence, if so, increasing the value of j by 1, and going to the step S603; if not, setting the value of j to 0 and increasing the value of i by 1 and going to step S603;
s603, if i is larger than or equal to m, judging that the first gradient and the second gradient in the latest continuous (T2-T1) time period are different, and finishing the step;
s604, if j is less than m, go to step S602; if j is larger than or equal to m, judging that the first gradient and the second gradient in the latest continuous time period (T2-T1) are the same, and ending the step.
7. A system for evaluating aquaculture feed based on aquaculture water quality big data, the system comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the computer program to run in the units of the following system:
the spectral image synchronous acquisition unit is used for starting to synchronously acquire spectral images of the water surfaces of the experimental pool and the reference pool through spectral image acquisition equipment arranged above feeding points of the experimental pool and the reference pool, the spectral images are respectively marked as experimental spectral images and reference spectral images, the time for starting to synchronously acquire the spectral images for the first time is T0, and the experimental spectral images and the reference spectral images are acquired at intervals of time TG to obtain an experimental spectral image sequence and a reference spectral image sequence;
the spectral image edge detection unit is used for respectively filtering the experimental spectral image sequence and the reference spectral image in the reference spectral image sequence, then carrying out binarization and carrying out edge detection by an edge detection operator, so that a first closed area formed by outermost edge lines in the experimental spectral image sequence is obtained through edge detection, a second closed area formed by outermost edge lines in the reference spectral image is obtained through edge detection, and if a third closed area formed by third edge lines appears in the first closed area or the second closed area for the first time, the acquisition time interval correction unit is skipped; if a third closed area formed by a third edge line does not exist in the first closed area or the second closed area, jumping to a spectral image synchronous acquisition unit;
an acquisition time interval correction unit, configured to note that the time when the third closed region first appears is T1, and modify the set time interval TG to TG = (T1-T0)/n, that is, the acquisition time duration from the start of acquiring the spectral image to the time when the third closed region appears (T1-T0) is divided by n, that is, the difference between T1 and T0; wherein n is the number of experimental spectral images in the experimental spectral image sequence or the number of reference spectral images in the reference spectral image sequence;
the aquatic feed initial detection unit is used for judging that the aquatic feed to be detected is qualified and finishing the step if the third closed area first appears in the second closed area; if the third closed area first appears in the first closed area, turning to a gradient trend calculation unit;
a gradient tendency calculation unit for recording the time of occurrence of a fourth closed region composed of a fourth edge line as T2 if the fourth closed region occurs inside the second closed region; respectively calculating diffusion gradients of the first closed area, the third closed area, the second closed area and the fourth closed area to obtain a first gradient and a second gradient;
the peak value selection unit is used for circularly executing the spectral image synchronous acquisition unit to the gradient trend calculation unit, and calculating to obtain each first gradient and each second gradient of the experimental spectral image and the reference spectral image which are shot at the synchronous time in the experimental spectral image sequence and the reference spectral image sequence; when the changes of the first gradient and the second gradient are changed from increasing to decreasing, or the first gradient and the second gradient are the same in the latest continuous time period (T2-T1), selecting the maximum value of the first gradient corresponding to each experimental spectral image in the experimental spectral image sequence as a first peak value, selecting the maximum value of the second gradient corresponding to each reference spectral image in the reference spectral image sequence as a second peak value, and transferring to an aquatic feed qualification judging unit;
and the aquatic feed qualification judging unit is used for judging that the aquatic feed to be detected is qualified when the first peak value is smaller than the second peak value, and otherwise, judging that the aquatic feed to be detected is unqualified.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110293244.3A CN112683814B (en) | 2021-03-19 | 2021-03-19 | Method and system for evaluating aquatic feed based on big data of aquaculture water quality |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110293244.3A CN112683814B (en) | 2021-03-19 | 2021-03-19 | Method and system for evaluating aquatic feed based on big data of aquaculture water quality |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112683814A true CN112683814A (en) | 2021-04-20 |
| CN112683814B CN112683814B (en) | 2021-06-11 |
Family
ID=75455654
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110293244.3A Active CN112683814B (en) | 2021-03-19 | 2021-03-19 | Method and system for evaluating aquatic feed based on big data of aquaculture water quality |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112683814B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114659995A (en) * | 2022-05-24 | 2022-06-24 | 广东省农业科学院动物科学研究所 | Fish phagostimulant effect evaluation method and system based on microscopic hyperspectrum |
| CN115452763A (en) * | 2022-10-27 | 2022-12-09 | 佛山市南海区杰大饲料有限公司 | Functional fish feed detection method and system based on near infrared spectrum |
| CN116432909A (en) * | 2023-06-13 | 2023-07-14 | 广东省农业科学院动物科学研究所 | Test method for evaluating feeding attraction effect of aquatic products |
| CN116930182A (en) * | 2023-07-17 | 2023-10-24 | 江苏海洋大学 | Dynamic collection, test and analysis method for feed particle diffusion distribution in culture net cage |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09262576A (en) * | 1996-03-27 | 1997-10-07 | Canon Inc | Diffusion preventing method for soil pollution |
| RU2176701C2 (en) * | 1999-11-17 | 2001-12-10 | Казанский государственный технологический университет | Method of oil gathering from under ice cover of basin with running water |
| CN103282769A (en) * | 2010-10-18 | 2013-09-04 | Mikasa商事株式会社 | Water quality monitoring apparatus |
| US20160120410A1 (en) * | 2014-11-03 | 2016-05-05 | Samsung Electronics Co., Ltd. | Spectrometer including vertical stack structure and non-invasive biometric sensor including the spectrometer |
| CN106290276A (en) * | 2016-08-01 | 2017-01-04 | 江苏大学 | A kind of quality evaluation method of enclosed waters aquiculture fresh water |
| CN109270237A (en) * | 2018-11-27 | 2019-01-25 | 广州市怡纯净水科技有限公司 | A kind of Monitoring And Analysis of The Quality system based on big data |
| CN111639456A (en) * | 2020-06-17 | 2020-09-08 | 太原理工大学 | Full space-time water area pollutant emergency situation sensing system based on unmanned aerial vehicle |
| CN111829824A (en) * | 2020-06-05 | 2020-10-27 | 重庆普绿斯环保科技发展有限公司 | Automatic sampling device and water quality monitoring method |
| CN111896481A (en) * | 2020-06-18 | 2020-11-06 | 中国市政工程华北设计研究总院有限公司 | Automatic modeling and water quality parameter automatic identification system for river and lake water environment and operation method |
-
2021
- 2021-03-19 CN CN202110293244.3A patent/CN112683814B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09262576A (en) * | 1996-03-27 | 1997-10-07 | Canon Inc | Diffusion preventing method for soil pollution |
| RU2176701C2 (en) * | 1999-11-17 | 2001-12-10 | Казанский государственный технологический университет | Method of oil gathering from under ice cover of basin with running water |
| CN103282769A (en) * | 2010-10-18 | 2013-09-04 | Mikasa商事株式会社 | Water quality monitoring apparatus |
| US20160120410A1 (en) * | 2014-11-03 | 2016-05-05 | Samsung Electronics Co., Ltd. | Spectrometer including vertical stack structure and non-invasive biometric sensor including the spectrometer |
| CN106290276A (en) * | 2016-08-01 | 2017-01-04 | 江苏大学 | A kind of quality evaluation method of enclosed waters aquiculture fresh water |
| CN109270237A (en) * | 2018-11-27 | 2019-01-25 | 广州市怡纯净水科技有限公司 | A kind of Monitoring And Analysis of The Quality system based on big data |
| CN111829824A (en) * | 2020-06-05 | 2020-10-27 | 重庆普绿斯环保科技发展有限公司 | Automatic sampling device and water quality monitoring method |
| CN111639456A (en) * | 2020-06-17 | 2020-09-08 | 太原理工大学 | Full space-time water area pollutant emergency situation sensing system based on unmanned aerial vehicle |
| CN111896481A (en) * | 2020-06-18 | 2020-11-06 | 中国市政工程华北设计研究总院有限公司 | Automatic modeling and water quality parameter automatic identification system for river and lake water environment and operation method |
Non-Patent Citations (1)
| Title |
|---|
| 李辉 等: "近红外光谱分析技术在饲料质量安全评价中的应用研究进展", 《农业工程学报》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114659995A (en) * | 2022-05-24 | 2022-06-24 | 广东省农业科学院动物科学研究所 | Fish phagostimulant effect evaluation method and system based on microscopic hyperspectrum |
| CN115452763A (en) * | 2022-10-27 | 2022-12-09 | 佛山市南海区杰大饲料有限公司 | Functional fish feed detection method and system based on near infrared spectrum |
| CN116432909A (en) * | 2023-06-13 | 2023-07-14 | 广东省农业科学院动物科学研究所 | Test method for evaluating feeding attraction effect of aquatic products |
| CN116432909B (en) * | 2023-06-13 | 2023-10-20 | 广东省农业科学院动物科学研究所 | Test method for evaluating feeding attraction effect of aquatic products |
| CN116930182A (en) * | 2023-07-17 | 2023-10-24 | 江苏海洋大学 | Dynamic collection, test and analysis method for feed particle diffusion distribution in culture net cage |
| CN116930182B (en) * | 2023-07-17 | 2024-04-05 | 江苏海洋大学 | Dynamic collection, test and analysis method for feed particle diffusion distribution in culture net cage |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112683814B (en) | 2021-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112683814B (en) | Method and system for evaluating aquatic feed based on big data of aquaculture water quality | |
| Fry et al. | Sulfur stable isotope indicators of residency in estuarine fish | |
| Chapal et al. | Resolving noise–control conflict by gene duplication | |
| CN111047073B (en) | Aquaculture water quality prediction method and system based on neural network | |
| Monkman et al. | Accurate estimation of fish length in single camera photogrammetry with a fiducial marker | |
| Zimmer et al. | Uniform carbon fluxes in shallow lakes in alternative stable states | |
| Scheuffele et al. | Thermal performance curves for aerobic scope in a tropical fish (Lates calcarifer): flexible in amplitude but not breadth | |
| CN105574516B (en) | The ornamental pine apple chlorophyll detection method returned based on logistic in visible images | |
| Bronte et al. | Morphological variation of siscowet lake trout in Lake Superior | |
| Peters et al. | Lipids as a proxy for larval starvation and feeding condition in small pelagic fish: a field approach on match-mismatch effects on Baltic sprat | |
| Branscombe et al. | Seasonal shellfishing across the east Adriatic Mesolithic-Neolithic transition: oxygen isotope analysis of Phorcus turbinatus from Vela Spila (Croatia) | |
| CN113204990B (en) | Machine learning method and device based on intelligent fishing rod | |
| CN107071802B (en) | Mobile phone APP positioning abnormal data detection method and device | |
| Curth et al. | Ichnology of an extant belly-dragging lizard—analogies to early reptile locomotion? | |
| Sarà | Aquaculture effects on some physical and chemical properties of the water column: a meta-analysis | |
| Cameron et al. | At the Central European–Balkan transition: forest land snail faunas of the Banat contrasted with those of the Carpathian chain | |
| Duan et al. | Three dimensional analysis of ontogenetic variation in fossil globorotaliiform planktic foraminiferal tests and its implications for ecology, life processes and functional morphology | |
| CN113179981B (en) | Apartment crab auxiliary breeding method, system and device based on deep learning | |
| Domínguez-Petit et al. | Evaluating the use of the autodiametric method for estimating fecundity of Reinhardtius hippoglossoides, a species with an unusual oocyte development strategy | |
| Brieske et al. | Method dependency of maximum oxygen uptake rate and its repeatability in the Gulf killifish, Fundulus grandis | |
| Gal et al. | The effect of summer monsoon on pelagic and littoral food webs in a large regulated reservoir (Lake Paldang, Korea): A stable isotope approach | |
| CN117215218B (en) | Control method for dynamic pollution discharge of fish circulating water culture pond | |
| Mee et al. | Distribution of Phoxinus eos, Phoxinus neogaeus, and their asexually-reproducing hybrids (Pisces: Cyprinidae) in Algonquin Provincial Park, Ontario | |
| Meerbeek | Retention of visible implant alphanumeric tags in adult Walleye | |
| CN116740767B (en) | Fish feed throwing method based on machine vision |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |