WO2021112798A1 - Fish counting machine and system - Google Patents

Fish counting machine and system Download PDF

Info

Publication number
WO2021112798A1
WO2021112798A1 PCT/TR2020/051191 TR2020051191W WO2021112798A1 WO 2021112798 A1 WO2021112798 A1 WO 2021112798A1 TR 2020051191 W TR2020051191 W TR 2020051191W WO 2021112798 A1 WO2021112798 A1 WO 2021112798A1
Authority
WO
WIPO (PCT)
Prior art keywords
fish
counting
fishes
camera
image
Prior art date
Application number
PCT/TR2020/051191
Other languages
French (fr)
Inventor
Ali BARAN
Onur Sahin
Riza DONMEZ
Ahad AGHAPOUR
Hatice BUKEN
Derya TURAN
Muhammed Vuslat OZEL
Original Assignee
Yonga Teknoloji Mikroelektronik Arge Tic. Ltd. Sti.
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Yonga Teknoloji Mikroelektronik Arge Tic. Ltd. Sti. filed Critical Yonga Teknoloji Mikroelektronik Arge Tic. Ltd. Sti.
Priority to EP20895243.2A priority Critical patent/EP4068953A4/en
Publication of WO2021112798A1 publication Critical patent/WO2021112798A1/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06MCOUNTING MECHANISMS; COUNTING OF OBJECTS NOT OTHERWISE PROVIDED FOR
    • G06M11/00Counting of objects distributed at random, e.g. on a surface
    • AHUMAN NECESSITIES
    • A01AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01KANIMAL HUSBANDRY; AVICULTURE; APICULTURE; PISCICULTURE; FISHING; REARING OR BREEDING ANIMALS, NOT OTHERWISE PROVIDED FOR; NEW BREEDS OF ANIMALS
    • A01K61/00Culture of aquatic animals
    • A01K61/90Sorting, grading, counting or marking live aquatic animals, e.g. sex determination
    • A01K61/95Sorting, grading, counting or marking live aquatic animals, e.g. sex determination specially adapted for fish

Landscapes

  • Life Sciences & Earth Sciences (AREA)
  • Zoology (AREA)
  • Environmental Sciences (AREA)
  • Marine Sciences & Fisheries (AREA)
  • Animal Husbandry (AREA)
  • Biodiversity & Conservation Biology (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Farming Of Fish And Shellfish (AREA)

Abstract

The present invention relates to fish counting machine and system which becomes prominent thanks to the reporting and archiving characteristics and precise fish counting by means of detection through line scanning camera (K) in a real-time manner by using machine learning and deep learning and the morphological characteristics of fishes, in all areas where fish counting must be realized, primarily in fish firms, fish production plants and fish sales areas.

Description

FISH COUNTING MACHINE AND SYSTEM
TECHNICAL FIELD
The present invention relates to fish counting machine and system which becomes prominent thanks to the reporting and archiving characteristics and precise fish counting by means of detection through line scanning camera in a real-time manner by using machine learning and deep learning and the morphological characteristics of fishes, in all areas where fish counting must be realized, primarily in fish firms, fish production plants and fish sales areas.
KNOWN STATE OF THE ART
Fishes are vertebrate animals which are poikilothermic and which nearly completely only live in water and which make respiration by means of gills and which are cold-blooded and which have double-chambered hearts and which have nearly completely scaled body and which generally reproduce with eggs.
Fish is a low-oil including food which has high protein and which has various advantages for health. Particularly the oil proportion of white meaty fishes is lower than the oil proportion of other animal protein sources. The oily fishes have higher proportion of omega-3 oil acid, in other words, they have higher proportion of good oil. Heart attacks are at the first among the death reasons in the world.
Consuming fish regularly is one of the best ways for protecting heart health in accordance with American Heart Organization which recommends consuming fish at least 2 times per week. Various studies made on fish consumption and heart health until today have shown that regular fish consumption decreases heart attack and paralysis risk. Consuming fish regularly once per week decreases heart disease risk by 15% on average. Experts mention that preference of oily fishes with the omega-3 oil acid is more effective.
Omega-3 oil acids of oily fishes like salmon provide an important base for the normal development of brains of babies. Experts recommend fish consumption at least 2 times per week for providing contribution to the eye and brain development of babies and for pregnant women and for nursing mothers. The researches made also show that insufficient omega-3 consumption has unfavorable effects on eye development in babies.
Brain functions begin to deteriorate with ageing. This condition known as cognitive regression is a normal process of ageing. However, serious diseases like Alzheimer may occur in this process.
The studies made on fish consumption and brain health show that the cognitive regression depending on age is slower as fish consumption increases. This condition is associated with the gray substance which exists in the brain.
There are studies which show that the gray substance in the brain, which is a basic functional tissue, is greater in persons who consume fish.
Depression, which is one of the most frequent spiritual problems today, leads to the person being more insensible, having lower energy and being sadder.
After the heart diseases and obesity which threaten the world, depression is one of the biggest health problems in the world. Experts mention that regular fish consumption decreases depression risk.
Omega-3 oil acid of fish and particularly oily fish can be effective in the fight with depression and moreover, it also increases the efficiency of depression treatment drugs.
Vitamin D which functions as happiness hormone in the body is a vitamin which is frequently lacking in the public particularly in winter. Vitamin D does not exist much in food but fish is a good vitamin D source. Particularly oily fishes like salmon are substantially rich in terms of vitamin D.
A portion of grilled salmon nearly meets all of your daily need for vitamin D.
The destruction of healthy tissues in the body may lead to some auto-immune diseases like diabetes. The studies made shows that fish consumption is substantially important for decreasing type 1 diabetes risk in children. The researches show that fish is also important for the other immune system diseases like MS disease. Some problems occur in eye health as a result of ageing. The studies made on eye diseases and feeding show that the omega-3 oil acid of fish provides protection against various eye diseases.
Sleep disorders are one of the biggest problems of modern life. The studies made on sleep disorders show that vitamin D insufficiency may play a role on sleep disorders.
Consumption of oily fishes like salmon regularly 2-3 times per week and increase of daily consumed vitamin D increase both sleeping quality and life quality.
Because of the abovementioned basic advantages and reasons, fish consumption takes an important place in the food chain of today. Within this context, since our country is encircled by seas from four sides, it is a substantially rich country in terms of fish production and consumption. Within this context, fish is produced also in fish firms for providing fishing in seas in a controlled manner and particularly for meeting the need for trout.
In such production, it is substantially difficult to realize counting and number controls since tagging of fishes is not possible in the related places and in fish wholesale areas. Within this context, workers realize this counting process manually. However, this process leads to effort and time loss.
For meeting the abovementioned need, fish counting machines have been developed. One of these fish counting machines is the patent with application number DK 3143559 (T3) and which belongs to the company Vaki Fiskeldiskerfi Hf. Even though a fish counting machine and the operation principle thereof have been described in this application, the most basic deficiency of this study is that said machine cannot realize archiving and reporting. Thus, since the information cannot be stored and the reports cannot be taken afterwards, the fishes are counted again in case of an error, and comparison with the prior counts cannot be realized. Said invention with number DK 3143559 (T3) which belongs to the company Vaki describes a method for controlling the concentration of viable organisms which exist in fluid for facilitating carrying of organisms with functions like graduation and counting, and a device which realizes this function. Said apparatus comprises a feeding channel for the viable organism flow in the fluid (1), a second feeding channel related to the fluid (2) flow and a chamber (4) for the arrangement of the counted organism. The abstract describes that the organism, which goes to the outlet (15) through the first and second feeding channel, is counted by means of a computer, and here, a sensor (9) is placed to the first feeding channel for detecting the density of the viable organism therein. As can also be understood from the abstract, the related mechanism cannot realize an instantaneous archiving and reporting.
In the file with number CN208459234 (U), biological counting is described which passes through the fluid. However, in the related mechanism, counting processes are realized in the same manner, but learning cannot be realized and therefore, it is close to manual usage and at the same time, it cannot realize archiving and reporting.
The present systems cannot meet the basic requirements as mentioned below and include the problems mentioned below;
• Error proportions of equivalent products are high.
• Calibration is substantially needed by the user. User errors were occurring during calibration. The companies which have similar solution in the world cannot give technical support to Turkey.
DEFINITION OF THE INVENTION
Said invention eliminates the disadvantages described in the known state of the art and meets the needs.
Said invention relates to a fish counting machine which becomes prominent thanks to the reporting and archiving characteristics and precise fish counting by means of detection through line scanning camera in a real-time manner by using machine learning and deep learning and the morphological characteristics of fishes, in all areas where fish counting must be realized, primarily in fish firms, fish production plants and fish sales areas.
In said invention, camera and illumination placement angles are changed mechanically. The length, angle (radius) and width of the machine have been designed for solving the problems. All corners in the system which the fishes contact are designed in a chamfered form.
New machine learning methods have been used for making the error proportion independent from the fish and fish dimension in software. The block diagram of the software has been shown in Figure 1.
The basic advantages of the present invention can be described as follows:
• Easy installation and usage.
• Prevention of the probable user errors. • Correct counting and correct result.
• Usage of machine learning methods.
• Reporting and archiving of the results and transferring of the results through the lot.
REFERENCE NUMBERS
1. Machine body
2. Tank
3. Channels
4. Camera and illumination unit
5. Water/fish discharge part
K Camera
A. Intermediate memory capturing array
GA. Image analysis
G. Image
M. Median
GB. Image combining . Periphery scanning
MH. Morphological calculation characteristic
T. Estimation characteristic
GKA. Graphical user interface
S. Result
SY. Result capturing
SA. Result analysis
BRIEF DESCRIPTION OF THE DRAWINGS
In Figure 1 , a representative general view of the block diagram of said system is given.
In Figure 2, a representative general lateral view of said invention is given.
DESCRIPTION OF THE INVENTION
The present invention is a fish counting machine comprising:
• Machine body (1 ) which is the chassis of the whole system,
• Tank (2) where water is received and where excessive water is discharged and which is the inlet area of the system, • Channels (3) provided between the illumination unit (4) and the camera where counting process is realized and the tank (2) and which comprises guide paths for fishes,
• Camera and illumination unit (4) which is the area where the cameras (K) and the illumination units, providing the required luminous medium for taking the images (G) of the fishes by the camera, are existing,
• Water/fish discharge part (5) which forms the outlet of the system and which provides transfer of the water and the fishes, which remain in the system, to another desired medium.
In said invention, the system essentially operates by passing and counting of the water and fishes, transferred to the tank (2) provided on the machine body (1), through the camera and illumination unit (4) via channels (3) after the water and fishes enter to the system, and by discharging the counted fishes through the water/fish discharge part (5).
For the operation of the system, the machine body (1) must be placed on a straight and robust floor.
The height of the system from the floor must be adjusted such that the passages of water and fishes through the channels (3) are equal by means of the feet of the machine body (1).
The machine body (1) must be positioned such that user accesses to the touch-panel freely and comfortably and such that no exposure to sun occurs. High temperatures may damage the computer.
After the system is operated, water and fishes begin to enter to the tank (2) by means of pumps. Since excessive accumulation of the water and fishes in the tank (2) will affect the counting process in an unfavorable manner, the speed of the pump must be correctly adjusted. The water and fishes, provided in the tank (2), advance towards the camera and the illumination unit (4), where counting will be realized, by means of channels (3). The cameras transfer the image (G) scanned in line form in the camera and illumination unit (4) where the required illumination is provided to the cameras, and provide realization of the counting process by the software. The water and fishes which leave the camera and illumination unit (4) leave the system in the desired direction through the water/fish discharge part (5). Therefore, the front of the water/fish discharge part (5) must not be blocked and must not be subjected to obstacle. Since the inverse pressure, which may occur at the pumps, may lead to overloading to the system, water and fish provision should be ensured by means of the free flow of the system.
In said invention, the operation of the system is as follows;
1 . Power is given to the fish counting machine through UPS. (The power feeding of the counting machine must be realized through UPS. The grounding connections of the network must not be deficient and must be compliant to the international standards.)
2. After power is applied to the system, the system will be completely turned on within 2 minutes at the latest. As the system is turned on, the counting program will also be turned on and will take its place on the screen.
3. Water must be given to the system by realizing the required connections for discharging excessive water by means of the pump. The pump speed must be between 27-30 br. In case the given water is excessive, the required connections must be realized for removing the excessive water from under the main body. For being able to control the water which is to be discharged, it is recommended that a valve is fixed under the main body.
4. In the turned on fish counting program, the program is started by pressing the “start” button. After starting of the program, together with discharging of the excessive water, only water is passed through the machine. Here, it must be checked that the excessive water is removed from under the body and the program does not realize any counting by means of the effect of water.
5. In order to ensure that the program is ready for counting fish and does not have water effect, it is checked in GUI that the “frame count” number increases and the fish number does not increase.
6. The system is ready for counting fishes. The fishes can be sent in a compliant manner to the capacity of the machine.
The explanation of the diagram seen in Figure 1 ;
After the images (G), scanned by the cameras (the camera and illumination unit (4)), are passed through the median filter (M), the image is combined (GB) in the system and is sent for image analysis (GA) to the image (G) processing algorithm.
The algorithm analyzes the coming photograph and finds the contours and separates the fishes from the background by means of morphological calculation characteristic (MH), and the pre-trained dataset is based on, and separates to numerical classes by the machine learning algorithm. The fish images separated to the numerical class estimates the number of fishes in the examined region (estimation characteristic (T)) and adds this number to the total fish number. By means of machine learning, the number of fishes corresponding one on the other and side by side is decomposed, and the result is obtained in total number which is closest to the real value. The total number of fishes is displayed to the user through screens and the whole total is shown to user at the end of counting.
In the system, camera systems using line scanning is used. Thus, the fish passing through the scanner is counted once.
In the invention, operation by using the image (G) mode or the real-time mode directly by means of camera is possible.

Claims

1. The present invention is a fish counting system wherein the following steps are provided:
• After the images (G), scanned by the cameras (K) (the camera and illumination unit (4)), are passed through the median filter (M), the image is combined (GB) in the system and is sent for image analysis (GA) to the image (G) processing algorithm in parts,
• The algorithm analyzes the coming photograph and finds the contours and separates the fishes from the background by means of “morphological calculation characteristic” (MH), and the pre-trained dataset is based on, and separates to numerical classes by the machine learning algorithm,
• The fish images separated to the numerical class estimates the number of fishes in the examined region (estimation characteristic (T)) and adds this number to the total fish number,
• By means of machine learning, the number of fishes corresponding one on the other and side by side is decomposed, and the result is obtained in total number which is closest to the real value,
• Displaying the total fish number to the user through screens and displaying the whole total to the users at the end of counting.
2. The fish counting system according to claim 1 , wherein the system realizes counting process by instantaneously transferring the image (G), scanned in line form in the camera and illumination unit (4) where illumination is provided for the cameras, to the computer.
3. The fish counting machine mentioned in the invention, wherein a camera is used which realizes linear scanning.
4. The fish counting system according to claim 1 , wherein counting can be realized in real-time mode or by using image (G) mode directly by means of camera.
PCT/TR2020/051191 2019-12-03 2020-11-30 Fish counting machine and system WO2021112798A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
EP20895243.2A EP4068953A4 (en) 2019-12-03 2020-11-30 Fish counting machine and system

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
TR201919104 2019-12-03
TR2019/19104 2019-12-03

Publications (1)

Publication Number Publication Date
WO2021112798A1 true WO2021112798A1 (en) 2021-06-10

Family

ID=76221809

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/TR2020/051191 WO2021112798A1 (en) 2019-12-03 2020-11-30 Fish counting machine and system

Country Status (2)

Country Link
EP (1) EP4068953A4 (en)
WO (1) WO2021112798A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115953725A (en) * 2023-03-14 2023-04-11 浙江大学 Fish egg automatic counting system based on deep learning and counting method thereof

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110114029A1 (en) * 2009-11-17 2011-05-19 Chu Chien-Hung Aquatic-animal counting system, aquatic-animal counting method, and recording medium
WO2015173838A1 (en) * 2014-05-12 2015-11-19 Vaki Fiskeldiskerfi Hf. Optical system for counting objects
KR20180076083A (en) * 2016-12-27 2018-07-05 부산대학교 산학협력단 Aquatic animals counter
JP2019153077A (en) * 2018-03-02 2019-09-12 日本電気株式会社 Individual counting device, individual counting method, individual counting program, and individual counting system
WO2019198701A1 (en) * 2018-04-13 2019-10-17 日本電気株式会社 Analysis device and analysis method

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111868472A (en) * 2018-03-20 2020-10-30 吉利海洋科技有限公司 System and method for extracting statistical samples of multiple moving objects

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110114029A1 (en) * 2009-11-17 2011-05-19 Chu Chien-Hung Aquatic-animal counting system, aquatic-animal counting method, and recording medium
WO2015173838A1 (en) * 2014-05-12 2015-11-19 Vaki Fiskeldiskerfi Hf. Optical system for counting objects
KR20180076083A (en) * 2016-12-27 2018-07-05 부산대학교 산학협력단 Aquatic animals counter
JP2019153077A (en) * 2018-03-02 2019-09-12 日本電気株式会社 Individual counting device, individual counting method, individual counting program, and individual counting system
WO2019198701A1 (en) * 2018-04-13 2019-10-17 日本電気株式会社 Analysis device and analysis method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4068953A4 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115953725A (en) * 2023-03-14 2023-04-11 浙江大学 Fish egg automatic counting system based on deep learning and counting method thereof

Also Published As

Publication number Publication date
EP4068953A4 (en) 2023-12-27
EP4068953A1 (en) 2022-10-12

Similar Documents

Publication Publication Date Title
Franks et al. Animal welfare risks of global aquaculture
Pettersen et al. Salmon welfare index model 2.0: an extended model for overall welfare assessment of caged Atlantic salmon, based on a review of selected welfare indicators and intended for fish health professionals
Liu et al. Measuring feeding activity of fish in RAS using computer vision
Papadakis et al. A computer-vision system and methodology for the analysis of fish behavior
Blonk et al. Heritability of shape in common sole, Solea solea, estimated from image analysis data
Corrêa et al. Effect of the establishment of dominance relationships on cortisol and other metabolic parameters in Nile tilapia (Oreochromis niloticus)
Nordgreen et al. Behavioural effects of the commonly used fish anaesthetic tricaine methanesulfonate (MS-222) on zebrafish (Danio rerio) and its relevance for the acetic acid pain test
TW201118761A (en) Aquatic animal counting system, method and recording medium thereof
EP4068953A1 (en) Fish counting machine and system
KR101931998B1 (en) Discrimination method of healthy seed using phenotype of farmed fish
CN113598098B (en) Fish disease recognition processing device based on machine vision
Holtze et al. Monitoring health and reproductive status of olms (Proteus anguinus) by ultrasound
Almansa et al. The laser scanner is a reliable method to estimate the biomass of a Senegalese sole (Solea senegalensis) population in a tank
Rey et al. Farmed fish welfare practices: salmon farming as a case study
Nevitt et al. Comparison of air sac volume, lung volume, and lung densities determined by use of computed tomography in conscious and anesthetized Humboldt penguins (Spheniscus humboldti) positioned in ventral, dorsal, and right lateral recumbency
White et al. Pedomorphosis in the ancestry of marsupial mammals
JP7129675B2 (en) Poultry farming system, poultry farming method and program
Lyles Genetic variation and susceptibility to parasites: Poeclia reticulata infected with Gyrodactylus turnbulli
Nilsson et al. Welfare Indicators for farmed Atlantic salmon–Part A. Knowledge and theoretical background
Styrishave et al. Diurnal variations in physiology and behaviour of the noble crayfish Astacus astacus and the signal crayfish Pacifastacus leniusculus
Hoey et al. To feed or to breed: morphological constraints of mouthbrooding in coral reef cardinalfishes
Galeana et al. Mother-young spatial association and its relation with proximity to a fence separating ewes and lambs during enforced weaning in hair sheep (Ovis aries)
Guo et al. Social isolation affects intra‐specific interaction behaviour and reduces the size of the cerebellar brain region in juvenile Atlantic salmon Salmo salar
Frommel et al. Organ health and development in larval kingfish are unaffected by ocean acidification and warming
Kraskura et al. Body size and temperature affect metabolic and cardiac thermal tolerance in fish

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20895243

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2020895243

Country of ref document: EP

Effective date: 20220704