CN116576782A - Underwater fish body length measurement method - Google Patents
Underwater fish body length measurement method Download PDFInfo
- Publication number
- CN116576782A CN116576782A CN202310294481.0A CN202310294481A CN116576782A CN 116576782 A CN116576782 A CN 116576782A CN 202310294481 A CN202310294481 A CN 202310294481A CN 116576782 A CN116576782 A CN 116576782A
- Authority
- CN
- China
- Prior art keywords
- fish body
- fish
- length
- image
- target detection
- 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
- 241000251468 Actinopterygii Species 0.000 title claims abstract description 227
- 238000000691 measurement method Methods 0.000 title claims abstract description 11
- 238000001514 detection method Methods 0.000 claims abstract description 72
- 238000000034 method Methods 0.000 claims abstract description 46
- 238000005259 measurement Methods 0.000 claims abstract description 21
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 6
- 238000004364 calculation method Methods 0.000 claims description 14
- 230000008569 process Effects 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 13
- 238000007781 pre-processing Methods 0.000 claims description 5
- 230000009182 swimming Effects 0.000 claims description 4
- 238000012360 testing method Methods 0.000 claims description 4
- 238000012795 verification Methods 0.000 claims description 4
- 238000009360 aquaculture Methods 0.000 abstract description 10
- 244000144974 aquaculture Species 0.000 abstract description 10
- 238000012545 processing Methods 0.000 abstract description 5
- 239000002028 Biomass Substances 0.000 abstract description 2
- 230000001276 controlling effect Effects 0.000 abstract description 2
- 230000001105 regulatory effect Effects 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 5
- 230000006872 improvement Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 206010002091 Anaesthesia Diseases 0.000 description 1
- 101000642315 Homo sapiens Spermatogenesis-associated protein 17 Proteins 0.000 description 1
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 102100036408 Spermatogenesis-associated protein 17 Human genes 0.000 description 1
- 230000037005 anaesthesia Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000005452 bending Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000003306 harvesting Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/02—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness
- G01B11/04—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness specially adapted for measuring length or width of objects while moving
- G01B11/043—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness specially adapted for measuring length or width of objects while moving for measuring length
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/02—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness
- G01B11/022—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness by means of tv-camera scanning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
- Y02A40/80—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in fisheries management
- Y02A40/81—Aquaculture, e.g. of fish
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Image Processing (AREA)
Abstract
The application discloses an underwater fish body length measuring method, which comprises the following steps: collecting video information, and constructing a fish body target detection model based on the video information; constructing a fish body target tracking model, and tracking and shooting the fish body in the video information based on the fish body target detection model and the fish body target tracking model to obtain a fish body image; and calculating the length of the fish body based on the fish body image. The application solves the problems that the existing non-contact measurement method based on machine vision and image processing has larger limitation in the measurement of the length of the fish body, can not measure multi-angle fish bodies, has low measurement efficiency and the like. The method provides decision basis for optimizing the feeding amount, regulating and controlling the water quality environment and grading the quality, promotes the evolution of a refined, intelligent and intensive aquaculture mode, and brings new challenges and opportunities for estimating the biomass of the aquaculture.
Description
Technical Field
The application relates to the technical field of aquaculture, in particular to an underwater fish body length measurement method.
Background
The aquaculture in China develops rapidly. The aquaculture industry occupies an important position in the fishery industry structure in China, and plays an important role in promoting the fishery economic development in China.
With the advent of the internet age and the rapid popularization of artificial intelligence in various industries, the automatic management and control of these elements are realized by means of equipment in the future, so that the risk of aquaculture is reduced and the cost of aquaculture is reduced. Therefore, future intelligentization is one of the trends in the aquaculture industry.
The measurement of the fish growth data is indispensable in the cultivation process, and relates to the throwing of fish fries, the feeding of feeds, disease control, harvesting and catching and the like of farmers, so that the method has important guiding effects on scientific decisions of the farmers and improvement of comprehensive benefits, and provides biological basis for the evaluation and reasonable utilization of fishery resources.
At present, the length measurement method adopted for the fish body is divided into contact measurement and non-contact measurement. The contact type measurement method generally needs steps of anesthesia, fishing, manual measurement and the like on the fish body, the measurement operation is time-consuming and labor-consuming, when the fish is in great charge, the fish is exposed to the skin and struggled and bounces, the fish can be in full measure, accurate and punctual measurement cannot be realized, the measurement efficiency is low, the accuracy is low, the fish can be subjected to negative effects such as stress reaction in the measurement process, the fish is delayed in growth and even dead, and the economic loss is caused. Therefore, there is an urgent need to develop a contactless measuring method instead of a contact measuring method to achieve rapid and accurate measurement of the length of the fish body.
The existing contactless measurement method is to measure the length of the fish body by combining machine vision with image processing, but the image processing process of the contactless measurement is difficult to achieve simplification and intellectualization, has larger limitation in the process of measuring the length of the fish body, generally depends on a high-quality fish body image, a perfect form of the fish body in the image and an orthogonal imaging angle of the fish body in the image and an optical axis, and often needs a large amount of manual intervention in the measurement process to participate in preprocessing of the image, and particularly needs to manually mark measurement key points (mouth and tail of the fish body) of the fish body so as to finish the measurement of the length of the fish body, thereby seriously affecting the measurement efficiency.
In summary, in the development process of the image measurement technology of the cultured fish, although the design aspect of measuring the length of the fish body has been greatly improved, the problems of bending the fish body, multiple angles of the fish body and the like still exist during free swimming. Therefore, the design of the underwater fish body measuring device and the underwater fish body measuring method suitable for multiple angles in the farm has important significance and application value.
Disclosure of Invention
The application provides an underwater fish body length measuring method which is used for solving the problems that the existing non-contact measuring method based on machine vision and image processing has larger limitation in the measurement of the length of the fish body, cannot measure multi-angle fish bodies, has low measuring efficiency and the like.
In order to achieve the above object, the present application provides the following solutions:
the underwater fish body length measuring method comprises the following steps:
collecting video information, and constructing a fish body target detection model based on the video information;
constructing a fish body target tracking model, and tracking and shooting the fish body in the video information based on the fish body target detection model and the fish body target tracking model to obtain a fish body image;
and calculating the length of the fish body based on the fish body image.
Preferably, the method for constructing the fish body target detection model comprises the following steps:
constructing a fish body target detection image data set;
and training a fish body target detection model based on the fish body target detection data set.
Preferably, the method for constructing a fish body target detection image dataset comprises the following steps:
collecting images above and beside the fish body, and carrying out data annotation on the images above and beside the fish body to obtain annotated image data;
dividing the marked image data into a training set, a verification set and a test set according to a preset proportion to obtain the fish object detection image data set.
Preferably, the training method of the fish body target detection model comprises the following steps:
preprocessing the fish object target detection image data set to obtain a preprocessed image;
and inputting the preprocessed image into a target detection network for training to obtain the fish body target detection model.
Preferably, the method for calculating the length of the fish body comprises the following steps:
acquiring a fish body top view by using a first waterproof camera, and acquiring a fish body front view by using a second waterproof camera;
acquiring first data based on the fish top view and acquiring second data based on the fish front view;
acquiring an image magnification based on the fish body image, the first data and the second data;
and calculating the length of the fish body based on the first data, the second data and the image magnification.
Preferably, the first data includes: length-thickness ratio beta of fish body and actual length-thickness ratio beta 0 Distance v from center point of target detection frame to upper side of image when fish body is at center position 0 And the distance v from the center point of the target detection frame to the upper side of the image in the calculation process;
the second data includes: fish body length-width ratio alpha and actual length-width ratio alpha 0 Distance u from center point of target detection frame to upper side of image when fish body is at center position 0 And the distance u from the center point of the target detection frame to the upper side of the image in the calculation process.
Preferably, the fish body length calculating method comprises the following steps:
dividing the fish into different calculation conditions according to the swimming state of the fish body in water;
and obtaining the length of the fish body based on different calculation conditions.
Preferably, the calculating case includes: the fish body is in a horizontal state in the first waterproof camera and the second waterproof camera;
the fish body is in an inclined state in the first waterproof camera, and the fish body is in a horizontal state in the second waterproof camera;
the fish body is in a horizontal state in the first waterproof camera, and the fish body is in an inclined state in the second waterproof camera;
the fish body is in an inclined state between the first waterproof camera and the second waterproof camera.
The beneficial effects of the application are as follows:
the application solves the problems that the existing non-contact measurement method based on machine vision and image processing has larger limitation in the measurement of the length of the fish body, can not measure multi-angle fish bodies, has low measurement efficiency and the like. The method provides decision basis for optimizing the feeding amount, regulating and controlling the water quality environment and grading the quality, promotes the evolution of a refined, intelligent and intensive aquaculture mode, and brings new challenges and opportunities for estimating the biomass of the aquaculture.
Drawings
In order to more clearly illustrate the technical solutions of the present application, the drawings that are needed in the embodiments are briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow diagram of a method according to an embodiment of the present application;
FIG. 2 is a schematic diagram of a YOLOv7 network in accordance with an embodiment of the present application;
fig. 3 is a schematic layout view of a first waterproof camera and a second waterproof camera according to an embodiment of the present application;
FIG. 4 is a first waterproof camera magnification scatter plot of an embodiment of the present application;
fig. 5 is a second waterproof camera magnification scatter plot of an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
In order that the above-recited objects, features and advantages of the present application will become more readily apparent, a more particular description of the application will be rendered by reference to the appended drawings and appended detailed description.
In this embodiment, as shown in fig. 1, a method for measuring the length of an underwater fish body includes the following steps:
s1, collecting video information, and constructing a fish body target detection model based on the video information.
The construction method of the fish body target detection model comprises the following steps: constructing a fish body target detection image data set; and training a fish body target detection model based on the fish body target detection data set.
Wherein the method for constructing the fish body target detection image data set comprises the following steps: collecting images above and beside the fish body, and marking the data of the images above and beside the fish body to obtain marked image data; dividing the marked image data into a training set, a verification set and a test set according to a preset proportion to obtain a fish body target detection image data set. In this embodiment, a fish body target detection model is constructed through a YOLOv7 network, collected upper and side fish body images are obtained by using edge equipment, artificial data annotation of fish body target detection is performed through labelimg software, annotation information of each image is matched from original annotation files of a data set according to names, the annotation information is converted into YOLOv annotation format files, finally, training sets, verification sets and test sets are divided for all image data according to the proportion of 80%, 10% and 10%, and finally a fish body target detection image data set is formed.
The training method of the fish body target detection model comprises the following steps: preprocessing a fish body target detection image data set to obtain a preprocessed image; and inputting the preprocessed image into a YOLOv7 network, and training the YOLOv7 network to obtain the fish body target detection model. In this embodiment, as shown in fig. 2, the YOLOv7 network is composed of three parts: input, backbox, and head, where backbox is used to extract features and head is used for prediction. The architecture diagram according to fig. 2 walks through the network flow: preprocessing an input picture, aligning the input picture into RGB pictures with 640 x 640 size, inputting the RGB pictures into a backbone network, continuously outputting three layers of feature maps (hereinafter referred to as fm) with different size through the backbone network at a head layer according to three layers of output in the backbone network, predicting three types of tasks (classification, front and back background classification and frame) of image detection through a RepVGG block and conv, and outputting a final result.
S2, constructing a fish body target tracking model, and tracking and shooting the fish body in the video information based on the fish body target detection model and the fish body target tracking model to obtain a fish body image.
In this embodiment, the fish object tracking model construction is based on StrongSort, which generates StrongSort by simply equipping advanced components in DeepSORT with new SOTAs on popular benchmarks MOT17 and MOT 20. There are two lightweight, plug and play, model independent, appearance independent algorithms proposed in StrongSORT to optimize the tracking results, firstly, in order to better exploit global information, there are several methods proposed to associate short trajectories with trajectories using global linking models, which usually generate accurate but incomplete trajectories and associate them with global information, which, although they significantly improve tracking performance, all rely on computationally intensive models, especially appearance features, in contrast to an appearance-free linking model (AFLink) that only uses spatio-temporal information to predict whether two incoming tracklets belong to the same ID. Second, linear interpolation is widely used to compensate for the omission, however, it ignores motion information, which limits the accuracy of interpolation position, and in order to solve this problem, a gaussian smoothing interpolation algorithm (GSI) is proposed, which uses a gaussian process regression algorithm to enhance interpolation.
S3, calculating the length of the fish body based on the fish body image.
The method for calculating the length of the fish body comprises the following steps: acquiring a fish body top view by using a first waterproof camera, and acquiring a fish body front view by using a second waterproof camera; acquiring first data based on a fish body top view and acquiring second data based on a fish body front view; acquiring an image magnification based on the fish body image, the first data and the second data; the fish body length is calculated based on the first data, the second data, and the image magnification.
Wherein the first data comprises: length-thickness ratio beta of fish body and actual length-thickness ratio beta 0 Distance v from center point of target detection frame to upper side of image when fish body is at center position 0 And the distance v from the center point of the target detection frame to the upper side of the image in the calculation process; the second data includes: fish body length-width ratio alpha and actual length-width ratio alpha 0 Distance u from center point of target detection frame to upper side of image when fish body is at center position 0 And the distance u from the center point of the target detection frame to the upper side of the image in the calculation process. In this embodiment, the arrangement of the first waterproof camera and the second waterproof camera is as shown in fig. 3, and it is specified that in the top view of the fish body photographed by the first waterproof camera, the ratio of the length to the width of the target detection frame is the length to thickness ratio β of the fish, and in the front view of the fish body photographed by the second waterproof camera, the ratio of the length to the width of the target detection frame is the aspect ratio α of the fish. The actual length-thickness ratio of the fish body is beta 0 The actual length-width ratio is alpha 0 . When the fish is in the central position, recording the distance v from the central point of the target detection frame to the upper side of the image in the fish body top view shot by the first waterproof camera 0 In the front view of the fish body shot by the second waterproof camera, the distance from the center point of the target detection frame to the upper side of the image is u 0 . In the procedure running process, in the top view of the fish body shot by the first waterproof camera, the distance from the center point of the target detection frame to the upper side of the image is v, and in the front view of the fish body shot by the second waterproof camera, the distance from the center point of the target detection frame to the upper side of the image is u.
Since the cameras will change the size of the fish in the water, we need to determine the magnification delta of the fish by both cameras before calculating the length. Firstly, placing the fish at the central position, namely, the distances between a fish body target detection frame in the first waterproof camera and the second waterproof camera and the left and right sides are equal, and the lengths of fish bodies in the two cameras are equalThe degrees are equal. By fixing the position of the fish in one of the cameras, the fish in the other camera moves two points upwards and two points downwards from the central position, and the five sets of data are recorded. Likewise, five additional sets of data were recorded. y is 0 And (3) for the fish length in the central position, y is the fish length after moving, and the moving distance of the fish is in direct proportion to the fish body length, so that the photographed image in the first waterproof camera meets the formula (1), and the photographed image in the second waterproof camera meets the formula (2).
y-y0＝Δ1(u-u0) (1)
y-y0＝Δ2(v-v0) (2)
The difference between Δ1 and Δ2 is small by calculation, so Δis averaged, i.e., Δ= (Δ1+Δ2)/2.
Fish length of first waterproof camera and fish center and image upper side distance u and standard fish center and image upper side distance u in second waterproof camera 0 The data of the difference are shown in table 1, the data are plotted as a scatter diagram, and the slope of the obtained trend line is the magnification of the first camera, as shown in fig. 4.
TABLE 1
| u-u0 | l_up |
| x | y |
| -368 | 371 |
| -315 | 261 |
| -232 | 181 |
| -147 | 113 |
| -129 | 88 |
| -76 | 51 |
| 0 | 0 |
| 53 | -15 |
| 149 | -89 |
| 221 | -163 |
| 270 | -226 |
| 357 | -309 |
The fish length of the second waterproof camera and the distance v between the fish center and the upper side of the image in the first waterproof camera 0 The data of the difference are shown in table 2, the data are plotted into a scatter diagram, and the slope of the obtained trend line is the magnification of the second camera, as shown in fig. 5.
TABLE 2
| v-v0 | l_down |
| x | y |
| -463 | 443 |
| -382 | 300 |
| -283 | 232 |
| -204 | 149 |
| -115 | 87 |
| 0 | 0 |
| 86 | -67 |
| 167 | -137 |
| 266 | -203 |
| 359 | -258 |
| 427 | -324 |
The fish body length calculating method comprises the following steps: dividing the fish into different calculation conditions according to the swimming state of the fish body in water; based on different calculation conditions, the fish body length is obtained. The calculation condition comprises the following steps: the fish body is in a horizontal state in the first waterproof camera and the second waterproof camera; the fish body is in an inclined state in the first waterproof camera, and the fish body is in a horizontal state in the second waterproof camera; the fish body is in a horizontal state in the first waterproof camera, and the fish body is in an inclined state in the second waterproof camera; the fish body is in an inclined state between the first waterproof camera and the second waterproof camera.
In this embodiment, the first state is a horizontal state of the fish body in the first waterproof camera and the second waterproof camera, and the determination condition is that when the difference between the measured aspect ratio and the actual aspect ratio is smaller than ε 1 (ε1=0.1), as shown in formula (3).
α-α0＜ε1 (3)
At this time, the length of the fish body is the length of a fish body target detection frame in the image shot by the second waterproof camera, namely, the abscissa of the upper right corner of the target detection frame is subtracted by the abscissa xd of the upper left corner, and then the obtained length is corrected, as shown in a formula (4).
length＝xd+Δ(v-v0) (4)
The second state is that the fish body is in an inclined state in the first waterproof camera, the fish body is in a horizontal state in the second waterproof camera, when the fish is in an inclined state in the image, the fish length is the length of the diagonal line of the target detection frame, and when the fish is in a horizontal state in the image, the fish length is the width multiplied by the length-to-width ratio of the target detection frame;
length2＝y2×α (6)
wherein x1 and y1 are the length and width of the target detection frame in the first waterproof camera, and y2 is the width of the target detection frame in the second waterproof camera.
The second state is determined under the condition that the absolute value of the difference between the fish length1 in the first waterproof camera and the fish length2 in the second waterproof camera is smaller than epsilon 2 (epsilon 2=50) pixels as shown in formula (7), and the fish length is the average value of the two lengths at this time as shown in formula (8).
abs(length1-length2)＜ε2 (7)
length＝(length1+length2)/2 (8)
The third state is that the fish body is in a horizontal state in the first waterproof camera, the fish body is in an inclined state in the second waterproof camera, when the fish is in a horizontal state in the image, the fish length is the width of the target detection frame multiplied by the length-width ratio, and when the fish is in an inclined state in the image, the fish length is the length of the diagonal line of the target detection frame. The determination condition is that the fish length1 in the first waterproof camera is equal to the fish length2 in the second waterproof camera.
length1＝y1*β (9)
Wherein y1 is the width of the target detection frame in the first waterproof camera, and x2 and y2 are the length and width of the target detection frame in the second waterproof camera.
The third state is determined under the condition that the absolute value of the difference between the fish length1 in the first waterproof camera and the fish length2 in the second waterproof camera is smaller than epsilon 2 (epsilon 2=50) pixels as shown in formula (7), and the fish length is the average value of the two lengths at this time as shown in formula (8).
abs(length1-length2)＜ε2 (7)
length＝(legth1+length2)/2 (8)
The fourth state is that the fish body is in an inclined state between the first waterproof camera and the second waterproof camera, and the length of the fish is calculated by calculating the body diagonal, as shown in a formula (11).
Wherein x1 and y1 are the length and width of the target detection frame in the first waterproof camera, and y2 is the width of the target detection frame in the second waterproof camera.
The fish body length calculated in the four modes is the pixel number, a constant C is calculated according to the image resolution and the fish length at the center position, and the real fish body length is obtained by multiplying the pixel number by C.
The above embodiments are merely illustrative of the preferred embodiments of the present application, and the scope of the present application is not limited thereto, but various modifications and improvements made by those skilled in the art to which the present application pertains are made without departing from the spirit of the present application, and all modifications and improvements fall within the scope of the present application as defined in the appended claims.
Claims (8)
1. The underwater fish body length measuring method is characterized by comprising the following steps of:
collecting video information, and constructing a fish body target detection model based on the video information;
constructing a fish body target tracking model, and tracking and shooting the fish body in the video information based on the fish body target detection model and the fish body target tracking model to obtain a fish body image;
and calculating the length of the fish body based on the fish body image.
2. The underwater fish body length measurement method according to claim 1, wherein the method for constructing the fish body target detection model comprises the steps of:
constructing a fish body target detection image data set;
and training a fish body target detection model based on the fish body target detection data set.
3. A method of underwater fish length measurement as claimed in claim 2, wherein the method of constructing a fish object detection image dataset comprises:
collecting images above and beside the fish body, and carrying out data annotation on the images above and beside the fish body to obtain annotated image data;
dividing the marked image data into a training set, a verification set and a test set according to a preset proportion to obtain the fish object detection image data set.
4. The underwater fish body length measurement method according to claim 2, wherein the training method of the fish body target detection model comprises:
preprocessing the fish object target detection image data set to obtain a preprocessed image;
and inputting the preprocessed image into a target detection network for training to obtain the fish body target detection model.
5. The method for measuring the length of an underwater fish body according to claim 1, wherein the method for calculating the length of the fish body comprises the steps of:
acquiring a fish body top view by using a first waterproof camera, and acquiring a fish body front view by using a second waterproof camera;
acquiring first data based on the fish top view and acquiring second data based on the fish front view;
acquiring an image magnification based on the fish body image, the first data and the second data;
and calculating the length of the fish body based on the first data, the second data and the image magnification.
6. The method of claim 5, wherein the first data comprises: length-thickness ratio beta of fish body and actual length-thickness ratio beta 0 Distance v from center point of target detection frame to upper side of image when fish body is at center position 0 And the distance v from the center point of the target detection frame to the upper side of the image in the calculation process;
the second data includes: fish body length-width ratio alpha and actual length-width ratio alpha 0 Distance u from center point of target detection frame to upper side of image when fish body is at center position 0 And calculatedAnd the distance u from the center point of the target detection frame to the upper side of the image in the process.
7. The underwater fish body length measurement method of claim 6, wherein the fish body length calculation method comprises:
dividing the fish into different calculation conditions according to the swimming state of the fish body in water;
and obtaining the length of the fish body based on different calculation conditions.
8. The method for measuring the length of an underwater fish according to claim 7, wherein the calculating comprises: the fish body is in a horizontal state in the first waterproof camera and the second waterproof camera;
the fish body is in an inclined state in the first waterproof camera, and the fish body is in a horizontal state in the second waterproof camera;
the fish body is in a horizontal state in the first waterproof camera, and the fish body is in an inclined state in the second waterproof camera;
the fish body is in an inclined state between the first waterproof camera and the second waterproof camera.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310294481.0A CN116576782B (en) | 2023-03-24 | 2023-03-24 | Underwater fish body length measurement method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310294481.0A CN116576782B (en) | 2023-03-24 | 2023-03-24 | Underwater fish body length measurement method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116576782A true CN116576782A (en) | 2023-08-11 |
| CN116576782B CN116576782B (en) | 2024-02-06 |
Family
ID=87540221
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310294481.0A Active CN116576782B (en) | 2023-03-24 | 2023-03-24 | Underwater fish body length measurement method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116576782B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109145808A (en) * | 2018-08-17 | 2019-01-04 | 郑州轻工业学院 | A kind of tuna recognition methods based on adaptive fish body bending die |
| CN111696150A (en) * | 2020-05-19 | 2020-09-22 | 杭州飞锐科技有限公司 | Method for measuring phenotypic data of channel catfish |
| US20210289759A1 (en) * | 2016-12-15 | 2021-09-23 | University Of The Philippines | Estimating fish size, population density, species distribution and biomass |
| CN114049577A (en) * | 2021-11-17 | 2022-02-15 | 中国水利水电科学研究院 | Fish specification measuring method and system |
| CN115082777A (en) * | 2022-04-26 | 2022-09-20 | 北京林业大学 | Binocular vision-based underwater dynamic fish form measuring method and device |
| JP7155375B1 (en) * | 2021-10-13 | 2022-10-18 | マルハニチロ株式会社 | Measurement system, information processing device, information processing method and program |
-
2023
- 2023-03-24 CN CN202310294481.0A patent/CN116576782B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210289759A1 (en) * | 2016-12-15 | 2021-09-23 | University Of The Philippines | Estimating fish size, population density, species distribution and biomass |
| CN109145808A (en) * | 2018-08-17 | 2019-01-04 | 郑州轻工业学院 | A kind of tuna recognition methods based on adaptive fish body bending die |
| CN111696150A (en) * | 2020-05-19 | 2020-09-22 | 杭州飞锐科技有限公司 | Method for measuring phenotypic data of channel catfish |
| JP7155375B1 (en) * | 2021-10-13 | 2022-10-18 | マルハニチロ株式会社 | Measurement system, information processing device, information processing method and program |
| CN114049577A (en) * | 2021-11-17 | 2022-02-15 | 中国水利水电科学研究院 | Fish specification measuring method and system |
| CN115082777A (en) * | 2022-04-26 | 2022-09-20 | 北京林业大学 | Binocular vision-based underwater dynamic fish form measuring method and device |
Non-Patent Citations (1)
| Title |
|---|
| 王振宇: "利用水下摄影测量技术测量鱼类体长的可行性研究", 大连海洋大学学报 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116576782B (en) | 2024-02-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113537106B (en) | Fish ingestion behavior identification method based on YOLOv5 | |
| CN114724022B (en) | Method, system and medium for detecting farmed fish shoal by fusing SKNet and YOLOv5 | |
| CN112883861B (en) | Feedback type bait casting control method based on fine-grained classification of fish school feeding state | |
| CN114241031B (en) | Fish body ruler measurement and weight prediction method and device based on double-view fusion | |
| CN113920453A (en) | Pig body size weight estimation method based on deep learning | |
| CN110163798B (en) | Method and system for detecting damage of purse net in fishing ground | |
| CN111887853B (en) | Fish body length measuring device and method based on binocular vision | |
| CN113349111A (en) | Dynamic feeding method, system and storage medium for aquaculture | |
| CN114627554A (en) | Automatic aquaculture feeding centralized management method and system for aquatic products | |
| CN114049577A (en) | Fish specification measuring method and system | |
| CN113313833A (en) | Pig body weight estimation method based on 3D vision technology | |
| CN111696079A (en) | Surface defect detection method based on multi-task learning | |
| Shi et al. | Underwater fish mass estimation using pattern matching based on binocular system | |
| CN116576782B (en) | Underwater fish body length measurement method | |
| CN117029673B (en) | Fish body surface multi-size measurement method based on artificial intelligence | |
| CN113484867A (en) | Imaging sonar-based fish school density detection method in closed space | |
| CN116452967B (en) | Fish swimming speed identification method based on machine vision | |
| CN109840498B (en) | Real-time pedestrian detection method, neural network and target detection layer | |
| CN115601301B (en) | Fish phenotype characteristic measurement method, system, electronic equipment and storage medium | |
| CN115843733A (en) | Machine vision-based electronic feeding table device for river crab cultivation and working method | |
| CN115049692A (en) | Intelligent regulation method and system for marine aquaculture illumination simulating natural ecology | |
| CN111145344A (en) | Structured light measuring method for snow carving 3D reconstruction | |
| Deng et al. | An automatic body length estimating method for Micropterus salmoides using local water surface stereo vision | |
| CN116912671A (en) | Intelligent fish feeding method and device | |
| US20240029289A1 (en) | System and method for estimating a length of underwater creatures |
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 |