Device and method for measuring residual feed intake of megalobrama amblycephala based on image processing

Abstract

The invention discloses a device and a method for measuring residual feed intake of megalobrama amblycephala based on image processing, which are characterized in that juvenile megalobrama amblycephala is independently fed by an isolated cage culture method, and high-precision batch analysis of feed counting is realized by combining Image J software for the first time. The technical problems that the food consumption character of megalobrama amblycephala is difficult to measure and the water quality is difficult to regulate and control through independent feeding are solved. According to the method, the fish body cannot be damaged in the measurement process, the water body replacement efficiency is improved, the stress reaction of the fish caused by water body replacement is effectively reduced, in addition, the method is convenient to manufacture, low in cost and low in site requirement, individual feed intake data of megalobrama amblycephala can be accurately and rapidly obtained, phenotype data of remaining feed intake characters of megalobrama amblycephala can be obtained, and the method is suitable for popularization and application. And a foundation is laid for subsequent genetic evaluation and excellent new product breeding of the economic character.

Description

Technical Field

[0001] The invention belongs to the technical field of economic trait selection and breeding of aquatic animals, and particularly relates to a device and method for measuring residual feed intake of Megalobrama amblycephala based on image processing. Background Art

[0002] Fish, as one of the main sources of protein for humans, plays a crucial role in the human food supply. my country, a major aquaculture producer, accounts for over 60% of the world's total aquaculture output. However, with the continuous rise in feed raw material prices in recent years, their share of fish farming costs has reached as high as 60-70%. Breeding new fish species with high feed efficiency has become a hot topic among fish breeders. Improving fish feed efficiency not only reduces farming costs but also effectively promotes the green and healthy development of the aquaculture industry. Residual feed intake (RFI), a key indicator of feed efficiency, refers to the difference between the actual feed intake of an animal and the expected feed intake required to maintain and meet growth and development during a given feeding period. However, the difficulty in measuring feed intake in individual fish makes it difficult to select for optimal feed efficiency traits during breeding, seriously hindering the genetic improvement of this important economic trait. Currently, methods for measuring feed intake in individual fish include X-ray irradiation, video recording, and individual container rearing. These methods all suffer from issues such as high workload, high cost, low efficiency, and damage to the fish. Summary of the Invention

[0003] The present invention provides a device and method for calculating the residual feed intake of Megalobrama amblycephala based on image processing, which solves the problem that the feed efficiency trait cannot be genetically improved due to the difficulty in obtaining the feed intake phenotype of fish.

[0004] To solve the above problems, the present invention provides the following technical solutions:

[0005] An embodiment of the present invention provides a device for measuring the residual feed intake of a large-headed bream based on image processing, comprising a circular culture barrel (1), a plurality of water pipes (2), a plurality of net cages (3) and an aeration device (6), wherein the plurality of water pipes (2) are fixed perpendicularly to each other on the edge of the culture barrel (1), the plurality of net cages (3) are fixed on the water pipes (2) in a 4-6-6-4 symmetrical arrangement, and the aeration device (6) is arranged in the culture barrel (1), the aeration device (6) is composed of an aeration head connected to an aerator through a rubber hose, and each net cage (3) comprises a galvanized iron mesh (4) and an 80-mesh nylon mesh (5) attached to the outside of the galvanized iron mesh (4).

[0006] In a preferred embodiment of the present invention, the mesh (4-1) of the galvanized iron mesh (4) has a hole diameter of 1.6 cm, and the diameter of a single iron wire of the galvanized iron mesh (4) is 0.7 mm.

[0007] In a preferred embodiment of the present invention, the length of the cage (3) is more than 4 times the total length of the experimental fish, the width of the cage (3) is more than 2.5 times the total length of the experimental fish, and the height of the cage (3) is more than 40 cm.

[0008] In a preferred embodiment of the present invention, the water depth is four-fifths of the height of the cage (3); and the bottom of the cage (3) is more than 20 cm higher than the bottom of the breeding barrel (1).

[0009] An embodiment of the present invention provides a method for calculating the residual feed intake of amblycephala bream based on image processing, comprising the following steps:

[0010] Step 1: Place one young amblycephala bream in each cage, number the cages, and temporarily raise them for two weeks to adapt to the water environment; before measuring the weight, stop feeding for one day and record the weight (BW) of each young amblycephala bream, which is recorded as BW 初 ;

[0011] Step 2: Place a feeder and fix it on the edge of the net cage, feed the fish with floating pellets daily. The particle size varies according to the size of the juvenile amblycephala, calculate the average weight of each pellet, and feed the fish at a daily rate of 4% of their body weight.

[0012] Step 3: The experimental period was 75 days, and the fish were fed at a fixed time and place every day. After the feeding, the remaining bait was collected, placed on a white dry paper, photographed, and the number of feed pellets consumed by each fish was counted using Image J;

[0013] Step 4, according to the ImageJ image processing method: Image grayscale processing: convert the original color image to an 8-bit grayscale image, Image→Type→8-bit, reduce the computational complexity, and unify the pixel range; dynamic threshold segmentation: through adaptive threshold adjustment, Image→Adjust→Threshold, check Don't set range and apply it to separate the target particles from the background, retain the complete grayscale information to enhance the segmentation robustness; adhesion particle segmentation: perform the watershed algorithm on the binary image after threshold segmentation, Process→Binary→Watershed, accurately separate the conjoined particles by simulating the principle of water erosion; particle analysis and counting: set the particle size range, Analyze→Analyze Particles, set Size to 10-Infinity (adjusted according to the actual particle size), and filter out small noise particles; check Display results, Clear results and Add to Manager to generate a particle outline map, Show: Outline, and output the statistical results as shown in the figure. Figure 5 ;

[0014] Step 5: Calculate the daily feed intake (DFI) of each fish according to the formula: feed intake = (total number of feed pellets fed - number of remaining feed pellets) × average weight of each feed pellet;

[0015] Step 6: At the end of the experimental period, the fish were fasted for one day and the body weight (BW) of each fish was measured and recorded, which was recorded as BW 末 , calculate the average daily gain ADG during the experimental period, the calculation formula is:

[0016] ADG=(BW 末 -BW 初 ) / test days;

[0017] Step 7: Collect fish phenotypic data and calculate the residual feed intake (RFI) according to the formula:

[0018] RFI=DFI-β0-β1×MW 0.8 -β2×ADG;

[0019] DFI is daily feed intake; β0 is the regression intercept; β1 is the partial regression coefficient of DFI on metabolic body weight; MW is the mean body weight, calculated as MW = (BW 初 +BW 末 ) / 2,MW 0.8 is metabolic weight (0.8 is the metabolic weight coefficient developed by Lupatsch et al.), β2 is the partial regression coefficient for ADG

[0020] Compared with the prior art, the embodiments of the present invention provide an image processing-based device and method for measuring the residual feed intake of amblycephala bream, which has the following beneficial effects: The present invention provides, for the first time, a method and device for measuring the residual feed intake of amblycephala bream, individually raising amblycephala juveniles through an isolated cage culture method, and for the first time proposes combining Image J software to achieve high-precision batch analysis of feed counting, overcoming the technical problems of the difficulty in measuring the feed intake trait of amblycephala bream and the difficulty in regulating water quality when raising amblycephala bream individually. This method does not cause damage to the fish during the measurement process, improves the efficiency of water replacement, and effectively reduces the stress response of fish caused by water replacement. In addition, the method is easy to manufacture, low-cost, and has low site requirements. It can accurately and quickly obtain individual feed intake data of amblycephala bream, and then obtain phenotypic data of its residual feed intake trait, laying the foundation for subsequent genetic evaluation of this economic trait and the selection and breeding of excellent new varieties. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0022] Figure 1 Schematic diagram of the structure of the device for measuring the residual feed intake (RFI) of individual amblycephala bream provided in the embodiment of the present application.

[0023] Figure 2 Schematic diagram of the cage structure of the device for measuring the residual feed intake (RFI) of individual amblycephala bream provided in an embodiment of the present application.

[0024] Figure 3 This is a statistical chart of the individual residual feed intake (RFI) of 175 amblycephala bream provided in the examples of this application.

[0025] Figure 4 Schematic diagram of the individual residual feed intake (RFI) of 175 amblycephalic bream analyzed using the SPSS software provided in the examples of this application.

[0026] Figure 5 This is a statistical chart of the amount of feed remaining counted using the Image J software provided in this application example. DETAILED DESCRIPTION

[0027] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application. The "upper", "lower", "front", "rear", "left", "right", etc. used in the installation position or direction of the structure or parts of this embodiment are based on the orientation of the given drawings. They are only for the convenience of expression to distinguish the relative positions of the various parts or directions, and do not represent the orientation of the device or functional parts of this embodiment when in use.

[0028] like Figure 1 and Figure 2 As shown, an embodiment of the present invention provides a device for measuring the residual feed intake of amblycephalic bream based on image processing. The device comprises a circular culture tank 1, multiple water pipes 2, multiple net cages 3, and an aeration device 6. The multiple water pipes 2 are fixed perpendicularly to the edge of the culture tank 1. The multiple net cages 3 and the aeration device 6 are arranged within the culture tank 1. Each net cage 3 comprises a galvanized iron mesh 4 and an 80-mesh nylon mesh 5 attached to the outer surface of the galvanized iron mesh 4. In this embodiment, the multiple net cages 3 are preferably fixed to the water pipe 2 in a symmetrical 4-6-6-4 arrangement. The aeration device 6 consists of an aeration head connected to an aerator via a rubber hose.

[0029] The mesh 4-1 of the galvanized iron mesh 4 has a diameter of 1.6 cm, and the diameter of each wire of the galvanized iron mesh 4 is 0.7 mm. The length of the cage 3 is at least 4 times the total length of the experimental fish, the width of the cage 3 is at least 2.5 times the total length of the experimental fish, and the height of the cage 3 is at least 40 cm. The water depth is 4 / 5 of the height of the cage 3; the bottom of the cage 3 is at least 20 cm higher than the bottom of the culture tank 1.

[0030] The cage 3 is manufactured as follows: 175 rectangular cages (25 cm long × 16 cm wide × 40 cm high) are made of galvanized iron mesh 4 with a 1.6 cm aperture, and are surrounded by 80-mesh nylon mesh 5. The advantages of choosing galvanized iron mesh for the cages are that it is non-deformable, easy to fold, and corrosion-resistant. This maximizes the swimming range of the amblycephalic bream, prevents water pollution, and extends the service life of the cages. Furthermore, it is easy to control water quality, easy to manufacture, and low in cost. The cage size, aperture, and water depth are determined based on the size, living habits, and suitable swimming space of the amblycephalic bream juveniles, maximizing the prevention of fish from escaping due to overly large cage apertures. This also ensures that fish excrement sinks for easy removal. The nylon mesh aperture is selected based on the particle size of the floating feed being fed, to prevent the feed particles from floating everywhere, making them difficult to collect and thus preventing accurate measurement of individual feed intake.

[0031] Nine circular aquaculture barrels 1 (120 cm diameter x 90 cm height) were selected, with 20 small net cages 3 placed in each barrel. The net cages 3 were secured in the barrels 1. The barrels 1 were filled with water to a depth of four-fifths of the height of the net cages 3. After the water was filled, aeration devices 6 were placed. The advantages of using aquaculture barrels 1 are ease of storage, low cost, and easy control of water quality.

[0032] An embodiment of the present invention provides a method for calculating the residual feed intake of amblycephala bream based on image processing, comprising the following steps:

[0033] Step 1: Place one young amblycephala bream in each cage, number the cages, and temporarily raise them for two weeks to adapt to the water environment; before measuring the weight, stop feeding for one day and record the weight (BW) of each young amblycephala bream, which is recorded as BW 初 ;

[0034] Step 2: Place a feeder and fix it on the edge of the net cage, feed the fish with floating pellets daily. The particle size varies according to the size of the juvenile amblycephala, calculate the average weight of each pellet, and feed the fish at a daily rate of 4% of their body weight.

[0035] Step 3: The experimental period was 75 days, and the fish were fed at a fixed time and place every day. After the feeding, the remaining bait was collected, placed on a white dry paper, photographed, and the number of feed pellets consumed by each fish was counted using Image J;

[0036] Step 4, according to ImageJ image processing method:

[0037] Image grayscale processing: Convert the original color image to an 8-bit grayscale image, Image→Type→8-bit, reduce computational complexity, and unify the pixel range; Dynamic threshold segmentation: Through adaptive threshold adjustment, Image→Adjust→Threshold, check Don't set range and apply it to separate the target particles from the background, retaining complete grayscale information to enhance segmentation robustness; Adhesion particle segmentation: Execute the watershed algorithm on the binary image after threshold segmentation, Process→Binary→Watershed, and accurately separate conjoined particles by simulating the principle of water erosion; Particle analysis and counting: Set the particle size range, Analyze→Analyze Particles, set Size to 10-Infinity (adjusted according to the actual particle size) to filter out small noise particles; Check Display results, Clear results, and Add to Manager to generate a particle outline map, Show: Outline, and output statistical results as shown below: Figure 5 ;

[0038] Step 5: Calculate the daily feed intake (DFI) of each fish according to the formula: feed intake = (total number of feed pellets fed - number of remaining feed pellets) × average weight of each feed pellet;

[0039] Step 6: At the end of the experimental period, the fish were fasted for one day and the body weight (BW) of each fish was measured and recorded, which was recorded as BW 末 , calculate the average daily gain (ADG) during the experimental period, the calculation formula is:

[0040] ADG=(BW 末 -BW 初 ) / test days;

[0041] Step 7: Collect fish phenotypic data and calculate the residual feed intake (RFI) according to the formula:

[0042] RFI=DFI-β0-β1×MW 0.8 -β2×ADG;

[0043] DFI is daily feed intake; β0 is the regression intercept; β1 is the partial regression coefficient of DFI on metabolic body weight; MW is the mean body weight, calculated as MW = (BW 初 +BW 末 ) / 2,MW 0.8is metabolic body weight (0.8 is the metabolic body weight coefficient developed by Lupatsch et al.), and β2 is the partial regression coefficient for ADG.

[0044] Example 1 A method for calculating the residual feed intake of Megalobrama amblycephala based on image processing comprises the following specific steps:

[0045] Step 1: Place one juvenile amblycephalic bream (4±0.5 g, 6±0.5 cm in length) in each cage. Record the phenotypic data of each fish before placement and number them sequentially from 1 to 175. Before the experiment begins, the juvenile amblycephalic bream are temporarily housed in a culture vat for two weeks to acclimate to the environment. The bottom of the vat is cleaned twice weekly.

[0046] Step 2: Feed the fish with a floating feed of 2 mm particle size and record the daily feed intake of each fish. The daily feed intake is 4% of the fish body weight, and the fish are fed twice daily. The remaining feed is collected and placed on white dry paper for photography.

[0047] Step 3, image preprocessing: import the feed pellet image into Image J software and perform grayscale processing;

[0048] Step 4: Threshold Segmentation: Adjust the threshold slider to ensure clear particle edges and select Don't set range before applying. For clinging particle processing: If clinging feed particles are present, use the watershed algorithm for segmentation. For particle counting: In the Analyze Particles dialog box, set Size = 10-Infinity (adjust based on the actual particle size). Select Outline to display particle outlines. Finally, obtain the final particle count and outline image. The average weight of each feed pellet is calculated by averaging the weights of 10,000 pellets.

[0049] The experimental period was 75 days. At the end of the experimental period, the residual feed intake phenotypic data of each fish were obtained according to the formula and statistical analysis was performed. Figure 3 This is a statistical chart of the residual feed intake (RFI) of 175 individual amblycephalic bream. SPSS was used to analyze the residual feed intake of each amblycephalic bream. Figure 4 The results of SPSS software analysis of the residual feed intake (RFI) of 175 individuals of Megalobrama amblycephala showed that the residual feed intake data were in accordance with the normal distribution, which reflected the accuracy and feasibility of the method.

[0050] Obviously, the above embodiments of the present invention are merely examples for the purpose of clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Those skilled in the art will appreciate that other variations or modifications can be made based on the above description. It is not necessary and impossible to enumerate all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the claims of the present invention.

Claims

1. A device for calculating the remaining feed intake of Megalobrama amblycephala based on image processing, characterized in that: The invention comprises a circular culture barrel (1), a plurality of water pipes (2), a plurality of net cages (3) and an aeration device (6), wherein the plurality of water pipes (2) are fixed perpendicularly to each other on the edge of the culture barrel (1), the plurality of net cages (3) are fixed on the water pipes (2) in a 4-6-6-4 symmetrical arrangement, and the aeration device (6) is arranged in the culture barrel (1), the aeration device (6) is composed of an aeration head connected to an aerator through a rubber hose, and each net cage (3) comprises a galvanized iron mesh (4) and an 80-mesh nylon mesh (5) attached to the outside of the galvanized iron mesh (4).

2. The device for calculating the residual feed intake of Megalobrama amblycephala based on image processing according to claim 1, characterized in that: The mesh (4-1) of the galvanized iron mesh (4) has a hole diameter of 1.6 cm, and the diameter of a single iron wire of the galvanized iron mesh (4) is 0.7 mm.

3. The device for calculating the residual feed intake of Megalobrama amblycephala based on image processing according to claim 1, characterized in that: The length of the cage (3) is more than 4 times the total length of the experimental fish, the width of the cage (3) is more than 2.5 times the total length of the experimental fish, and the height of the cage (3) is more than 40 cm.

4. The device for calculating the residual feed intake of Megalobrama amblycephala based on image processing according to claim 1, characterized in that: The water depth is four-fifths of the height of the cage (3); the bottom of the cage (3) is more than 20 cm higher than the bottom of the breeding barrel (1).

5. A method for calculating the residual feed intake of Megalobrama amblycephala based on image processing, characterized in that: The following steps are involved: Step 1: Place one young amblycephala bream in each cage, number the cages, and temporarily raise them for two weeks to adapt to the water environment; before measuring the weight, stop feeding for one day and record the weight (BW) of each young amblycephala bream, which is recorded as BW 初 ; Step 2: Place a feeder and fix it on the edge of the net cage, feed the fish with floating pellets daily. The particle size varies according to the size of the juvenile amblycephala, calculate the average weight of each pellet, and feed the fish at a daily rate of 4% of their body weight. Step 3: The experimental period was 75 days, and the fish were fed at a fixed time and place every day. After the feeding, the remaining bait was collected, placed on a white dry paper, photographed, and the number of feed pellets consumed by each fish was counted using Image J; Step 4: Image processing methods based on ImageJ are as follows: Image grayscale processing: Convert the original color image to an 8-bit grayscale image (Image → Type → 8-bit) to reduce computational complexity and unify the pixel range; Dynamic threshold segmentation: Use adaptive threshold adjustment (Image → Adjust → Threshold), check Don't set range and apply it to separate target particles from the background, retaining complete grayscale information to enhance segmentation robustness; Contiguous particle segmentation: Execute the watershed algorithm on the binary image after threshold segmentation (Process → Binary → Watershed) to accurately separate contiguous particles by simulating the principle of water erosion; Particle analysis and counting: Set the particle size range (Analyze → Analyze Particles), set Size to 10-Infinity (adjust according to the actual particle size) to filter out tiny noise particles; Check Display results, Clear results, and Add to Manager to generate a particle outline (Show: Outline) and output the statistical results; Step 5: Calculate the daily feed intake (DFI) of each fish according to the formula: feed intake = (total number of feed pellets fed - number of remaining feed pellets) × average weight of each feed pellet; Step 6: At the end of the experimental period, the fish were fasted for one day and the body weight (BW) of each fish was measured and recorded, which was recorded as BW 末 , calculate the average daily gain ADG during the experimental period, the calculation formula is: ADG=(BW 末 -BW 初 ) / test days; Step 7: Collect fish phenotypic data and calculate the residual feed intake (RFI) according to the formula: RFI=DFI-β0-β1×MW 0.8 -β2×ADG; DFI is daily feed intake; β0 is the regression intercept; β1 is the partial regression coefficient of DFI on metabolic body weight; MW is the mean body weight, calculated as MW = (BW 初 +BW 末 ) / 2,MW 0.8 is metabolic body weight (0.8 is the metabolic body weight coefficient developed by Lupatsch et al.), and β2 is the partial regression coefficient for ADG.

Images