NO347369B1 - Automatic Scanning System for Fish - Google Patents
Automatic Scanning System for FishInfo
- Publication number
- NO347369B1 NO347369B1 NO20201404A NO20201404A NO347369B1 NO 347369 B1 NO347369 B1 NO 347369B1 NO 20201404 A NO20201404 A NO 20201404A NO 20201404 A NO20201404 A NO 20201404A NO 347369 B1 NO347369 B1 NO 347369B1
- Authority
- NO
- Norway
- Prior art keywords
- duct
- fish
- probe
- subject
- region
- Prior art date
Links
- 241000251468 Actinopterygii Species 0.000 claims description 148
- 239000000523 sample Substances 0.000 claims description 136
- 239000007788 liquid Substances 0.000 claims description 59
- 238000002604 ultrasonography Methods 0.000 claims description 44
- 241001465754 Metazoa Species 0.000 claims description 33
- 238000003384 imaging method Methods 0.000 claims description 26
- 238000000034 method Methods 0.000 claims description 14
- 230000007246 mechanism Effects 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 9
- 210000001835 viscera Anatomy 0.000 claims description 5
- 235000019688 fish Nutrition 0.000 description 148
- 230000033001 locomotion Effects 0.000 description 14
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 11
- 230000008878 coupling Effects 0.000 description 10
- 238000010168 coupling process Methods 0.000 description 10
- 238000005859 coupling reaction Methods 0.000 description 10
- 241000277331 Salmonidae Species 0.000 description 5
- 238000012544 monitoring process Methods 0.000 description 5
- 230000003287 optical effect Effects 0.000 description 5
- 230000037361 pathway Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 241000972773 Aulopiformes Species 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 4
- 210000002149 gonad Anatomy 0.000 description 4
- 210000000056 organ Anatomy 0.000 description 4
- 235000019515 salmon Nutrition 0.000 description 4
- 230000005484 gravity Effects 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000003307 slaughter Methods 0.000 description 3
- 241000894007 species Species 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 239000003814 drug Substances 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000010255 response to auditory stimulus Effects 0.000 description 2
- 230000001568 sexual effect Effects 0.000 description 2
- 238000010408 sweeping Methods 0.000 description 2
- 238000011144 upstream manufacturing Methods 0.000 description 2
- 238000002255 vaccination Methods 0.000 description 2
- 241000238557 Decapoda Species 0.000 description 1
- 241000277263 Salmo Species 0.000 description 1
- FAPWRFPIFSIZLT-UHFFFAOYSA-M Sodium chloride Chemical compound [Na+].[Cl-] FAPWRFPIFSIZLT-UHFFFAOYSA-M 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 239000000654 additive Substances 0.000 description 1
- 210000003484 anatomy Anatomy 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 238000009530 blood pressure measurement Methods 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 239000000356 contaminant Substances 0.000 description 1
- 239000013078 crystal Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000030808 detection of mechanical stimulus involved in sensory perception of sound Effects 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 230000008713 feedback mechanism Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 239000000499 gel Substances 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000007373 indentation Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000035800 maturation Effects 0.000 description 1
- 235000013372 meat Nutrition 0.000 description 1
- 230000003278 mimic effect Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012634 optical imaging Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000009372 pisciculture Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 239000013535 sea water Substances 0.000 description 1
- 238000005204 segregation Methods 0.000 description 1
- 230000036301 sexual development Effects 0.000 description 1
- 239000011780 sodium chloride Substances 0.000 description 1
- 239000007921 spray Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 210000001550 testis Anatomy 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000007306 turnover Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01K—ANIMAL HUSBANDRY; AVICULTURE; APICULTURE; PISCICULTURE; FISHING; REARING OR BREEDING ANIMALS, NOT OTHERWISE PROVIDED FOR; NEW BREEDS OF ANIMALS
- A01K61/00—Culture of aquatic animals
- A01K61/90—Sorting, grading, counting or marking live aquatic animals, e.g. sex determination
- A01K61/95—Sorting, grading, counting or marking live aquatic animals, e.g. sex determination specially adapted for fish
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
- G01N29/04—Analysing solids
- G01N29/06—Visualisation of the interior, e.g. acoustic microscopy
-
- 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
- Life Sciences & Earth Sciences (AREA)
- Zoology (AREA)
- Environmental Sciences (AREA)
- Marine Sciences & Fisheries (AREA)
- Animal Husbandry (AREA)
- Biodiversity & Conservation Biology (AREA)
- Physics & Mathematics (AREA)
- Analytical Chemistry (AREA)
- Immunology (AREA)
- Chemical & Material Sciences (AREA)
- Acoustics & Sound (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Pathology (AREA)
- Farming Of Fish And Shellfish (AREA)
- Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)
- Processing Of Meat And Fish (AREA)
- Mechanical Means For Catching Fish (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
Description
AUTOMATIC SCANNING SYSTEM FOR FISH
The invention relates to an automatic scanning system, and in particular to an automatic scanning system for fish. The scanning system may be configured to determine characteristics of a subject using an ultrasound probe.
Fish farming is becoming increasingly important economically. This is particularly true in Norway where the vast majority of farmed fish is exported, and where a large proportion of the world’s total production of Atlantic salmon is sourced. Exports of farmed salmon and trout produced in Norway have continued to grow year on year.
Salmon and other fish are regularly farmed in large offshore (or onshore) installations in which the fish are contained and subjected to carefully controlled conditions in order to optimise growth and maximise profit. This means controlling the temperature, salinity, and pH of the water, monitoring how much the fish are eating, and providing the proper medicines if required. In order to determine the optimum parameters within a fish cage the growth and development of the fish must also be monitored closely. This has traditionally involved either removal and sacrifice of a number of the fish within the cage, or visual monitoring with optical equipment or other sensors located at a distance from the fish inside of the fish cage itself.
Gender segregation of fish also has a number of significant advantages in that it allows growth rate, slaughter weight, and weight class of the fish to be better controlled. Optimization of slaughter time, for example, can help to achieve a higher turnover rate of fish in the cages, and so a more efficient production. Fish growing quickly and well use the feed better, which means lower levels of pollution. A better quality of meat from the fish will also be possible, as the number of fish which have become sexually mature at the time of slaughter can be reduced.
NO-A-343355 describes a system for automatic sorting and vaccination of fish. The apparatus moves fish past a detection unit on a conveyer belt. The detection unit includes a camera which obtains an image of the fish which can be used to determine size, length, and so on. Fish are held in place by a suction device during the vaccination and sorting process, and are dropped into a sorting line which is selected based on information extracted from the camera image.
In WO-A-2020/002997 a sorting system suitable for use in the natural habitat of an animal, which can be a fish (e.g. in a lake), is described. The use of this system is primarily to remove invasive species from the environment in order to protect native species. The fish swims through the sorting apparatus on its own and is led into a holding pen if sensor data indicates that it belongs to the invasive species. Automatic sorting systems have previously made use of optical and other imaging systems, which are located away from the fish during scanning. GB-A-2201772 describes a scanning system including a channel along which a fish can swim, and having a source of radiation on one side and a series of cameras on the other. Blocking of the light by the fish can be detected at the cameras. US-A-2018/177169 describes a fishing system including a sorting mechanism for classifying shrimp or fish collected within the system.
It is known to analyse the sexual maturity of male salmonids using a handheld ultrasound probe, and this process is described in an article by Næve et al.
(Physiological Reports ISSN 2051-817X). The salmonids, which have already been sorted by gender to select males, are examined to determine the size of the left testes. Using this measurement, some information regarding the development of the fish can be extracted. The paper analyses the suitability of ultrasound as a replacement for previous more invasive methods of maturation monitoring.
Ultrasound or sonography works by the detection of sound waves having frequencies above the range detectable by the human ear. Probes comprise a transducer which emits ultrasound waves towards an object to be imaged. Sound waves which have been reflected from the interfaces between different materials return to the detector carrying with them information about the type and location of the materials forming the interfaces. Detailed images of the anatomy of a fish or another animal can be built up in this way.
Coupling between the transducer of the probe and the area to be imaged is an important factor in achieving a clear ultrasound image showing internal structure. Coupling gels are usually applied to the probe when used on a human or animal for imaging internal structures of the body, and the probe is pressed against the skin to take the image. Ultrasound will only be effective with the probe at close range and with adequate acoustic coupling, which requires the probe to be manipulated and held manually.
According to a first aspect of the present invention, there is provided a scanning system for imaging the internal parts of a subject, the system comprising: a duct having an inlet and an outlet, and a region in a base of the duct configured to contain a liquid and extending from the base of the duct less than 1cm into the duct, an ultrasound probe located in the region; and mechanical means configured to move the subject from the inlet to the outlet so that the subject passes through the region and past the probe.
In embodiments, the system comprises a liquid inlet for liquid to flow into and fill the region. This liquid inlet may be the same inlet or a different inlet to the duct inlet. When water fills the region, and the subject passes through the region during scanning, the space between the subject and the probe will be filled with water.
In embodiments, the region is an indent in the base of the duct. An indented region is simple to manufacture and allows liquid to collect, at least to an extent, over the probe head which protrudes through the base of the duct within the region.
In embodiments, the subject is an animal, and the system is a scanning system for imaging the internal organs of the animal. In embodiments, the subject is a fish. In embodiments, the subject is an anaesthetised fish.
In examples, the region extends from the base of the duct less than 30cm into the duct, preferably less than 10cm into the duct, more preferably less than 5cm into the duct, and most preferably less than 1cm into the duct. In some embodiments, the region extends between 1mm and 30cm, between 1mm and 5cm, or between 1mm and 1cm into the duct. If the region extends less than 30cm into the duct this means that the region is less than 30cm deep, as measured from the base of the duct toward the geometric centre, and may be 30cm deep across its whole area. The fish therefore passes within 30cm of the probe, and preferably much closer. When the region is full of liquid, this also means that there will be a layer of liquid in the region of a depth equal to the depth of the region, and the subject will pass through this layer of liquid as it travels past the probe. The depth of the water may be, for example, between 1mm and 30cm, preferably between 1cm and 10cm deep, and preferably between 1cm and 5cm deep.
In embodiments, the ultrasound probe operates in the frequency range 1MHz to 50MHz. This frequency range is ideal for imaging internal structure or organs.
In embodiments, the probe extends through the base of the duct within 30cm of the lowest point on the duct inner wall, as measured along the inner wall of the duct.
In embodiments, the probe is located at the lowest part of the inner duct wall. The lowest point refers to the part of the inner duct wall that is closest to the floor or platform on which the scanning system is supported in use.
In embodiments, the system comprises an outlet for liquid to flow from the region and out of the duct.
In embodiments, the mechanical means comprises a robot arm moveable to push or carry the subject along the duct.
In embodiments, the robot arm comprises one or more extensions configured to grip the subject.
In embodiments, the system comprises a processor configured to classify one or more images of the subject taken with the ultrasound probe into one or more groups.
In embodiments, the outlet comprises two or more outlet channels, and the system comprises a mechanism for selectively opening one of the outlet channels based on the group classification.
In embodiments, the system comprises a drainage device at the inlet to the duct for removal of excess water. The drainage device may comprise a grid with openings sized so that liquid can pass through for disposal or reuse in another part of the system, but the subject cannot.
In embodiments, the mechanical means comprises one or more conveyor belts forming the base of at least a portion of the duct.
In embodiments, the mechanical means comprises a downward slope in the duct from the inlet to the outlet and, in use, the subject slides along the duct under the force of gravity.
In embodiments, the system comprises a braking device configured to slow the speed of travel of the subject as it passes the probe.
In embodiments, the braking device comprises one or more of an opening through which the subject travels having a flexible diameter, a hinged door through which the subject passes that is biased in the closed position, and a rough portion of the duct surface.
In embodiments, the mechanical means comprises a stream of liquid which flows along the base of the duct and carries the subject. This liquid may also fill the region between the probe and subject in the imaging region of the duct.
According to a second aspect of the present invention, there is provided a method for automatic imaging and sorting of an animal, the method comprising: obtaining one or more ultrasound images of the internal organs of the animal using a system comprising: a duct having an inlet and an outlet, and a region in a base of the duct configured to contain a liquid; an ultrasound probe located in the region; and mechanical means configured to move the subject from the inlet to the outlet so that the subject passes through the region and past the probe; processing, by an image processor, the one or more ultrasound images and classifying the images as belonging to one of two or more groups based on at least one property of the images, characterized in that the region extends from the base of the duct less than 1cm into the duct.
In embodiments, the method comprises the step of directing the subject to an area based on the group that it is classified as belonging to.
The invention will be described in more detail with reference to the figures in which:
Figure 1 shows a scanning system;
Figure 2 shows a cross-section through a circular duct;
Figure 3A illustrates the position of a fish and probe as the fish moves past the probe;
Figure 3B illustrates the position of a fish and probe as the fish moves past the probe in an example where the fish contacts the probe as it passes.
The scanning system described below, and shown in figure 1, comprises a channel or duct 11 along which a fish can be guided in direction 18 past a scanner (in this case an ultrasound probe 15 comprising a transducer). The fish 14 may be anaesthetised prior to being guided or placed into the duct, at which point it will be carried along automatically by mechanical or other means. Prior to entering the duct, excess liquid may be removed from around the subject. This may be by way of a grating over which the subject passes prior to reaching the duct inlet. Water flows down through the grating leaving the fish on top of the grating with excess water removed.
The duct may be a closed or open channel as described in more detail below. The term “duct” is intended to refer simply to a pathway of some sort for the subject to pass through the imaging system and past the ultrasound probe. This duct can in some embodiments be a pipe, a half-pipe, or a closed or open-topped channel. The duct may have a curved or flat base. The probe is located in a wall, roof, or base of the duct and a layer 13 of water or another liquid is present between the probe and the subject (in this case the fish) as it is guided past the probe, so that proper acoustic coupling can be achieved without the requirement of manual intervention. This means that, while the system is in use, the part of the probe that is exposed within the duct and which is directed towards the subject during imaging is completely covered with a layer of liquid. There is a region of the duct surrounding the probe which is filled with liquid in use, and through which a part of the body of the subject passes (reference to the subject passing through a region of the duct refers to a part of the body of the subject passing through that region). The head of the probe may extend through the wall, roof, or base of the duct and may be surrounded, as is shown in figure 1, by a waterproof gasket 16 which prevents the liquid from leaking through the join between the duct and the probe. The ultrasound transducer may be configured to emit and receive high frequency sound waves in the range 1MHz to 50MHz, and preferably 1MHz to 30MHz.
In figure 1, the mechanical means by which a fish is encouraged to move along the duct comprises a robot arm 17 coupled to one or more movable extensions or fingers 19. The extensions, all or part of the arm itself, or both are movable to contact the fish 14 and to push it along the duct. The movement of the fingers 19 may encourage the fish along using a sweeping motion. This may be as a result of frictional forces between the contact point(s) and the body of the fish or may be aided by use of a suction cup or a similar device in order to grip the fish more forcefully. Extensions 19 may be provided on a robot arm as shown, or may be directly coupled to the top, bottom, or one or both sides of the duct. These extensions can be movable from side to side and forwards and backwards to provide the desired sweeping effect. The movement can be continuous and repetitive, and extensions may be provided all of the way along the duct.
Alternatively, the movement of the individual extensions or of the arm can be adapted depending on various parameters of the fish or where in the duct the extensions are located. Alternatively, the finger(s) 19 can be fixed relative to the arm. An attachment to the fish (a suction cup or other means) can be provided at the end of the fingers and the arm, with the fingers attached, can move to carry the fish along. The arm then needs to move in one direction only to sweep the fish over or past the probe. The arm may also be configured to move in a direction perpendicular to a direction along the duct (a sideways and/or up and down movement) to allow the arm to adjust the position of the subject in a direction across the duct. Whether this is required in each case will depend on the size and orientation of the subject, and the movement of the arm can be adjusted via a feedback mechanism in some cases (i.e. an image of the subject’s position causing an adjustment of the position using the arms). A single arm of this type can couple to the fish and carry or push it all of the way from the inlet to the duct to the outlet or can be provided to carry the fish past the probe with other means provided earlier and/or later during the fish’s journey along the duct. The arm can repeatedly couple to a fish, carry it along, decouple and move back towards the inlet to move another fish along the duct for scanning.
In other examples, the robot arm may be replaced with an alternative mechanical means for moving and/or guiding the fish through the system. The fish can be carried on a conveyor belt forming at least a portion of the base of the duct in one embodiment, in which case one conveyor belt may be provided downstream of the probe and another upstream if the probe is also located in the base of the duct. The duct may alternatively slope slightly from an entry point for the fish to an exit point from the duct. This way the fish is caused to slide down the duct from one end to another. A combination of one or more of the different means described above for moving the fish through the system may also be used together to provide the optimum speed of travel in the different parts of the system. It may be advantageous for the fish to move quickly through the system, but to slow as it passes the probe to give an efficient overall process while still achieving a high spatial resolution in the ultrasound images.
Whichever mechanism is used to move and guide the fish along the duct, the speed at which the fish moves, and in some cases the path of the fish and the position of the fish within the duct, can be controllable either manually or automatically in response to input. An optical camera can be used to take a picture of the fish entering the duct, and the position of the fish adjusted based on its apparent position as shown in the camera image. The same instrument can be used to measure a speed, for example by taking two pictures one after another and comparing the position of the fish in both, in which case the speed may also be adjusted to an optimum based on the measurement. One or more video cameras can also be used for the same purpose.
The speed at which the fish moves along the duct is important in terms of achieving a desired image resolution, especially in the portion of the duct near to the probe. In order to accurately determine the gender of a salmonid, which is one example of a potential use of the system, a spatial resolution of 2 mm (0.2 cm) or less for the ultrasound image is required. A spatial resolution of 1 mm (0.1 cm) is preferred. For such resolutions to be achievable, firstly the head of the probe must be close enough to the body of the fish to keep the distance that the sound waves need to travel before entering the body to a minimum, secondly the liquid filling the region between the probe and the fish body must be fairly free of bubbles and contaminants, and thirdly the fish must move slowly enough past the probe for image slices through the body to be able to be taken roughly every 2 mm (0.2 cm) along the length of the imaged section. This may necessitate a slowing of the speed of travel as the fish passes the probe.
To provide a precise control of speed as the fish passes the probe, as well as to provide better control in general, the position of the fish may be monitored as it moves along the duct. Monitoring may be by way of the optical cameras or video cameras described above, which can be located near to or within the duct as mentioned. The images of the fish entering and/or passing along the duct may be used to reposition the fish (i.e. using the robot arm) prior to it reaching the ultrasound probe. Other mechanisms for monitoring the movement of the fish can replace or enhance the action of the optical cameras. A measurement of speed can, for example, use light sensors located at two points within the duct to detect a drop in light intensity as the fish passes over, or one light sensor which emits light and detects a doppler shift in reflected light to determine a speed of movement of the passing subject.
Depending on the mechanical means used to move the fish along the duct, the movement may be controlled in the vicinity of the probe by a number of different mechanisms. For a system using robot arms and/or moveable extensions some feedback is provided to the robot arm and extensions to cause these to move so as to encourage the fish along the duct at the correct speed. If the duct slopes, at least in the portion where the probe is located, then some type of mechanical braking device can be included. This can consist of a hinged panel with a biasing mechanism for biasing the panel in a first position. As the fish passes along the duct it contacts one side of the panel and forces it outwards against the biasing force. The biasing force can be provided by a spring or a similar mechanism, and can be adjusted depending on the slope of the duct and the material of the duct base and walls, or the speed of movement of the conveyor belts, and so on, to achieve the correct speed as the fish passes over or past the probe.
The fish may also be forced through a narrower region of ducting as it passes the probe. The walls of the duct in this region can be somewhat flexible so that the fish is squeezed through the narrower region and is simultaneously pressed against the probe and its speed slowed slightly. A narrower region of this type may be provided by a number of hinged panels surrounding an opening and biased to close the opening in some examples, or by including a flexible walled opening in the duct or a portion of the ducting itself which is flexible to expand and contract the size of a section of tubing that the fish must pass through as it passes the probe. The fish is squeezed through the opening as a result of the force provided by the mechanical means acting to move the fish along the duct. The size of the opening can be automatically controlled in some examples, for example in response to a pressure measurement so that the opening can expand slightly once a fish is present at the opening to allow it to slide through.
A similar effect may be provided by providing a rougher surface on the base and/or one or more sides of the duct in the region preceding or surrounding the probe in order to increase the frictional force in that region and slow the movement of the fish as it passes over. If conveyor belts are used to move the fish, then the rough area of ducting will obviously not form part of a conveyor belt but will represent a static region of the duct either between to conveyor belts or directly following a conveyor belt.
Mechanical arms, and in some examples the same mechanical arm as is used to move the fish along the duct can also act to provide some pressure between the body of the fish and the probe. This should be a fairly light pressure to avoid damage to the fish while minimising the distance between the probe and the fish body.
The ultrasound probe is located at the base of the channel in the example shown, so that the fish passes over the probe as it moves through the apparatus along the transport duct. The opening through which the probe extends into the duct is protected using a waterproof seal, which prevents egress of water or liquid from the channel. It may be necessary for the head of the probe, where sound waves are detected, to be exposed to the liquid in the duct in order to provide proper acoustic coupling, but the regions around this and any gaps between the probe and the duct walls will be protected with a waterproof seal (in this case a gasket 16).
The probe can be located anywhere along the travel path of the fish from the duct inlet to the duct outlet and in any position on the duct in a transverse cross-section, however it is preferable that whatever its location and orientation, there should be a fairly small distance between the probe head and the body of the fish or other animal when the images are taken. There should also preferably be a layer of liquid present between the head of the probe and the body of the fish at the time of imaging.
Between the probe head and the body of the fish, therefore, there may be a small volume containing liquid. Preferably the thickness of this volume when the fish passes closest to the probe, and at the point where the body of the fish is closest to the probe, will be less than or equal to 300mm (30cm), more preferably less than or equal to 50 mm (5 cm), more preferably less than or equal to 10 mm (1 cm), more preferably less than or equal to 5 mm (0.5 cm), and most preferably less than or equal to 1 mm (0.1 cm). This distance may depend to some extent on the size, shape, and position of the fish, but these parameters will be less decisive if the probe is located in the base of the duct. This position for the probe directly underneath the subject will help to provide the minimum distance between the probe head and the body of the fish as it moves along the duct, because the force of gravity will help to press the body of the fish against the probe. In most cases with the probe in this position the minimum distance will be zero and the fish will contact the probe as it passes.
It is preferred that the position of the fish as it passes the probe is uniform for all fish imaged with the system. This makes comparison of images and classification simpler in terms of processing and more accurate. Preferably, the fish will lie on its side as it passes along the duct, and adjustment means can be provided to ensure that this is the case. The adjustment means can be the same as the mechanical means which are configured to move the fish along the duct. The shape of the duct itself, particularly in a region near to the inlet side, may also be configured to encourage the subject into the desired position.
The duct can be open or closed, and simply represents a pathway along which the fish, or other animal or object to be imaged, can travel. In some embodiments the duct will be in the form of a closed pipe which may have a circular cross-section, an elliptical cross-section, a square cross-section, or any other shape of cross-section. An elliptical cross-section, with the longer axis orientated horizontally when the system is in use, can be preferable as it can help to encourage the fish into the correct position while still allowing liquid to collect easily in the base of the duct around the probe (in a case where the probe is located below the subject). In some embodiments, the duct comprises the open base of a tube and may be in the form of a half-pipe, for example. The duct can also represent a flat surface, without sides, along which the animal or object to the imaged is carried. It is preferable for the shape of the duct to correspond at its base at least to some extent with the shape of the animal or object to be imaged. The configuration of the duct may change along the path through the system. The duct may be open-topped initially to facilitate additional optical imaging of the fish as it enters the system and in the form of a closed tube in the region where the probe is located to protect the animal or parts of the system, and/or to squeeze the fish against the probe to some extent as it moves over or past it.
If the duct has sides or sides and a top as well as a base in the region where the probe is located, then the probe may protrude through the base, one or more sides, or the top of the duct. As mentioned, providing the desired layer of liquid between the probe and the animal during imaging will be easier if the probe protrudes through the base of the duct as shown in figure 1, or if it is at least located within the lower half of the duct.
Figure 2 shows a cross-section through a duct which has a semi-circular crosssectional profile, and which is open-topped. The most preferable option will generally be to locate the probe at 0 degrees, so that it is centered on the vertical line running through the geometric centre of the duct as is shown in figure 2 (directly below the subject in the base of the channel and centered on the dotted line C representing the vertical). The vertical will be perpendicular to a horizontal platform on which the system is supported during use. This position for the probe at the lowest part of the duct will help to achieve the minimum possible distance between the probe and the subject during imaging, and will usually result in contact between the probe and the subject, using gravity to help push the subject against the probe and maintain this contact, all of which will help to achieve a clearer picture. Any free volume around the contact point between the probe and the subject, and located in front of the face of the probe such that sound waves emitted by the probe travel through it on a path to the subject, will be filled with liquid, so that an optimum acoustic coupling can be provided.
The probe can be located anywhere on the duct inner wall or within the duct and may still achieve a clear enough image, depending on the desired application and the size and shape of the animal to be imaged. The probe may be located at an angle from the vertical, measured in a transverse cross-section with the vertex at the geometric centre of the duct, of less than or equal to 90 degrees, more preferably at an angle from the vertical of less than or equal to 60 degrees, and most preferably at an angle from the vertical of less than or equal to 20 degrees. The region of the duct wall represented by this last range of angles is illustrated in figure 2 as region B. The geometric centre of the duct is represented in the figure as the point at which the dotted lines meet.
The shape and size of the duct itself can be selected in order to minimise distance between the probe and the body of animals passing the probe. The duct may need to be adapted to the particular size and shape of the subjects to be scanned using the system.
It is preferable, as explained above, to ensure that a layer of liquid is present between the ultrasound probe and the object or animal to be imaged. This can be achieved in a number of different ways. An indented region or dip can be formed in the base of the duct to form a pool of liquid as shown in figure 1. The probe can be located at the lowest part of this indented region where the liquid layer is deepest.
The liquid may collect to an extent within the indented region, which can help to minimise the formation of unwanted bubbles. It is preferable for the liquid in the duct to be recycled at least to a degree to keep the system clean and the fish healthy. The duct can include a spray system providing a steady stream of liquid in the duct at least in the vicinity of the probe. The entire base of the duct can be provided with a stream of liquid, and in some cases the movement of the liquid itself can help to push the fish along the duct.
Inlets for the liquid can be provided at intervals along the duct, at one end of the duct, or there can be a single inlet provided close to the probe. If a pool or reservoir of liquid directly above and covering the probe is provided by an indented duct portion, then this might be provided through one or more inlets within or above the indentation itself. There can be one or more outlets further provided within the indented portion, and the liquid in the pool formed by the indent can be continuously recycled in this way.
The positioning of the probe within the duct and the pathways provided for liquid to form pools or streams within the duct will preferably coincide so that the probe is completely covered by a layer of liquid at all times when the scanner is in use (while subjects are being passed along the duct by mechanical means). In any case, the probe should be covered by a layer of liquid at least at the time when a subject is at its closest to the probe and is being imaged. The layer of liquid covering the probe may be at least 0.1 cm deep, preferably at least 1 cm deep at its deepest point. This ensures that when the animal or object passes the probe, a layer of liquid is always present between the animal or object and the probe. The acoustic coupling between the liquid and the tissue of the fish or other animal being imaged is important and the system will not be usable if only air is present in the intervening space. The liquid may be specifically chosen to have an acoustic impedance that it as close as possible to that of the tissue of the subject. The liquid should fill the space or gap between the probe and the animal or object to be imaged in order to provide proper acoustic coupling for the sound waves passing from the probe into the body of the animal and back.
The space between the probe and the subject refers to the space directly in front of the part of the probe through which sound waves are emitted and received. This is shown in figures 3A and 3B in a case where the subject passes close to the probe with a shortest distance x between the probe and the subject (3A), and were the subject contacts the probe as it passes (3B). In both cases, the space which must be filled with liquid when the subject is at its closest point to the probe is shaded and marked as area D. The volume making up the liquid-filled space is therefore demarcated on one side by the face of the probe, or at least the region of the probe through which sound waves are emitted and received, and has the same size and shape as this region, and is demarcated on an opposite side by the subject itself as it passes closest to the probe (at which point one or more images are collected). A similar cross-section to that shown in figures 3A and 3B can therefore be provided at 90 degrees to the cross-section shown in the figures, and the liquid-filled space represents a volume having a lateral cross-section in a plane parallel to the probe face that is the same size and shape as the part of the probe through which sound waves are received and emitted.
The liquid is preferably water. This may be saline (e.g. seawater) or fresh. The liquid (water or otherwise) may comprise additives to achieve a response to sound waves that is as similar as possible to the tissue or the animal to be imaged in order to maximise acoustic coupling. A slightly different composition of the liquid may therefore be used depending on what animal or object is intended to be imaged. Where the system is used to image the internal organs of fish, the liquid will be chosen to mimic the properties of fish tissue in terms of the propagation of sound waves therethrough.
The ultrasound machine, which includes both the transducer and additional components 20, receives reflected sound waves from the body of the fish and converts motion of parts caused by the pressure waves to electronic signals. This can be, for example, by way of piezoelectric crystals which produce a voltage when a force is applied or a thin diaphragm which vibrates in response to sound waves, its movement causing a change in capacitance between the diaphragm and a back plate. The electrical signals are passed to processing software making up part of the additional components 20 and can be transformed there into a 2D images, raw data, or 3D images representing one or more slices through the body of the fish. The processing software may also receive meta-data from the ultrasound machine, along with the image data. The various parts of the ultrasound machine are standard and any suitable model for the transducer and additional components can be used in the scanning system described herein. It is preferred for the ultrasound transducer to be medical grade.
A processor 21 is connected to the ultrasound machine to receive and process the images in order to categorise in some way the subject passing through the system. The processor receives a number of 2D images representing slices through the body of the subject, preferably in regions spaced 1 mm (0.1 cm) apart or less along the body of the fish. These images will usually each also have a spatial resolution of 1 mm (0.1 cm) or less because of the proximity of the probe to the fish during imaging and the use of liquid for proper acoustic coupling.
Either each 2D image for a particular animal or object can be examined in turn, or the 2D slices can be merged to a single 3D image and this can be analysed to classify the image. Classification may use a particular predetermined algorithm or can use machine learning to develop the classification algorithm based on training data (images of fish with a known gender, for example). The algorithms can continue to develop during use of the system if the performance is checked or confirmed at least at intervals, for example by manual classification of the images. In an example 3D convolutional neural networks (CNN) can be used as part of the image analysis process.
Predetermined algorithms can use pattern recognition to locate particular organs within the images. Such algorithms can, for example, work by simply pinpointing the expected location of a gonad using recognisable markers on the body (the point of the noise, a fin, etc) and then identifying any large contiguous area of darker pixels. The size of this region and its presence or absence can be noted. If absence is determined, the fish is classified as male, and if a gonad is detected then the fish is classified as female. Similar processing methods can be used to locate other organs and measure the size of these. The size of the heart may be an indicator of health or illness, for example.
The imaging system can represent a unit to be used as a module within a larger system for treatment and sorting of fish or other animals. The fish/animal may need to be anaesthetised, for example, prior to feeding them into the imaging unit via the duct in order to avoid excessive stress or damage. This may require machinery designed to retain and inject the fish, which can be coupled to the imaging unit by feeding the fish from an outlet from the anaesthetising unit directly to the inlet of the duct of the imaging system described herein. Other modules, for cleaning and/or providing medicine to the animals, for example, can be included in the system upstream or downstream of the imaging system.
In some examples, the imaging system can also perform a sorting function, and the sorting may be based on the results of the image processing. One important use of the imaging system is to separate fish, and in particular salmonids or salmon or other fish at around 20g-10kg weight, according to gender. Larger fish between 30g and several kg, fish between 20g and 10,000g, or salmon or other fish from 30g up may be sorted using the system described herein. The gender of a fish can be very accurately determined using ultrasound because the sexual organs of the fish (and in particular the female fish) are clearly visible in an ultrasound image of the animal. Image processing to locate and isolate the gonad within a 3D image of 2D slices can also be achieved by fairly simple image processing techniques, as described above. The gender of the fish being imaged can be determined quickly by the processing software of the processor while the fish is passing through the duct past the probe. This determination can be used to control a subsequent pathway for the fish in order to sort the fish by gender. The duct could be bifurcated, for example, and one pathway closed off depending on the gender determined during processing.
Alternatively, a hole in the base of the duct, which may lead to a holding pen or back into the fish cage, can be caused to open or close based on the gender determination.
This type of sorting mechanism can be used with other information resulting from the image processing. The image can be processed to pick out diseased regions in the body, defects, or areas where damage has occurred, and fish can be sorted based on the presence or absence of these diseased and/or damaged regions. The ultrasound images can also be used to predict future sexual development of the fish (based on the size of the gonads in comparison to the total size, for example) and fish which are expected to develop more quickly can be separated from those which are expected to develop more slowly.
The system can include more than one duct and more than one ultrasound probe. The two or more probes can be coupled to the same or to a different ultrasound scanner. The ultrasound scanners (if more than one) can be coupled to the same or to a different processing unit. The operation of the two or more similar system parts can be coordinated so as to maximise efficiency during use. The mechanical means can be configured so that one fish reaches the probe within the first duct and is being imaged concurrently with a second fish in the second duct being positioned at the inlet of the duct correctly ready to be carried to the scanner using the mechanical means. This will ensure that the workload for the processor is more consistent.
Although the apparatus described above is particularly suited for use as a fish scanning system, it is also usable for scanning other animals, either marine or otherwise, provided that they can be passed along the duct and near enough to the ultrasound probe to provide a clear anatomical picture. The system can also be used for scanning inanimate objects.
In addition, while the system has been described as including an ultrasound probe as the imaging device, other imaging devices, such as x-ray scanners, light sensors, pressure or heat sensors, and so on may be swapped for or used in addition to the ultrasound probe in any of the examples described above. Although most suited as an ultrasound scanning system, the proximity between the subject and the sensor, and the automatic nature of the set-up, will still provide inherent advantages if other imaging devices are used.
Claims (17)
1. A scanning system for imaging the internal parts of a subject (14), the system comprising:
a duct (11) having an inlet and an outlet, and a region in a base of the duct configured to contain a liquid (13);
an ultrasound probe (15) located in the region; and
mechanical means configured to move the subject (14) from the inlet to the outlet so that the subject passes through the region and past the probe (15), characterized in that the region extends from the base of the duct (11) less than 1cm into the duct.
2. The scanning system of any of claims 1, wherein the system comprises a liquid inlet for liquid to flow into and fill the region.
3. The scanning system of any of claims 1 and 2, wherein the region is an indent in the base of the duct (11).
4. The scanning system of any of claims 1 to 3, wherein the subject (14) is an animal and the system is a scanning system for imaging the internal organs of the animal.
5. The scanning system of claim 4, wherein the subject (14) is a fish.
6. The scanning system of claim 4, wherein the subject (14) is an anaesthetised fish.
7. The scanning system of any of claims 1 to 6, wherein the ultrasound probe (15) operates in the frequency range 1MHz to 50MHz.
8. The scanning system of any of claims 1 to 7, wherein the ultrasound probe (15) extends through the base of the duct (11) within 30cm of the lowest point on the inner duct wall, as measured along the wall of the duct.
9. The scanning system of claim 8, wherein the ultrasound probe (15) extends through the inner duct wall at its lowest point.
10. The scanning system of claim 9, comprising an outlet for liquid to flow from the region and out of the duct (15).
11. The scanning system of any of claims 1 to 10, wherein the mechanical means comprises a robot arm (17) moveable to push or carry the subject (14) along the duct (15).
12. The scanning system of claim 11, wherein the robot arm (17) comprises one or more extensions (19) configured to grip the subject (14).
13. The scanning system of any of claims 1 to 12, wherein the system comprises a processor (21) configured to classify one or more images of the subject (14) taken with the ultrasound probe (15) into one or more groups.
14. The scanning system of any of claims 1 to 13, wherein the outlet comprises two or more outlet channels, and the system comprises a mechanism for selectively opening one of the outlet channels based on the group classification.
15. The scanning system of any of claims 1 to 14, comprising a drainage device at the inlet to the duct for removal of excess liquid.
16. A method for automatic imaging and sorting of an animal (14), the method comprising:
obtaining one or more ultrasound images of the internal organs of the animal using a system comprising:
a duct (11) having an inlet and an outlet, and a region in a base of the duct configured to contain a liquid (13);
an ultrasound probe (15) located in the region; and
mechanical means configured to move the subject (14) from the inlet to the outlet so that the subject passes through the region and past the probe (15);
processing, by an image processor (21), the one or more ultrasound images and
classifying the images as belonging to one of two or more groups based on at least one property of the images, characterized in that the region extends from the base of the duct (11) less than 1cm into the duct.
17. The method of claim 16, wherein the method comprises the step of directing the subject (14) to an area based on the group that it is classified as belonging to.
Priority Applications (6)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
NO20201404A NO347369B1 (en) | 2020-12-18 | 2020-12-18 | Automatic Scanning System for Fish |
CA3202150A CA3202150A1 (en) | 2020-12-18 | 2021-12-17 | Automatic scanning system for fish |
EP21848306.3A EP4240146A1 (en) | 2020-12-18 | 2021-12-17 | Automatic scanning system for fish |
AU2021403126A AU2021403126A1 (en) | 2020-12-18 | 2021-12-17 | Automatic scanning system for fish |
PCT/NO2021/050271 WO2022131929A1 (en) | 2020-12-18 | 2021-12-17 | Automatic scanning system for fish |
CL2023001767A CL2023001767A1 (en) | 2020-12-18 | 2023-06-15 | Automatic scanning system for fish |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
NO20201404A NO347369B1 (en) | 2020-12-18 | 2020-12-18 | Automatic Scanning System for Fish |
Publications (2)
Publication Number | Publication Date |
---|---|
NO20201404A1 NO20201404A1 (en) | 2022-06-20 |
NO347369B1 true NO347369B1 (en) | 2023-10-02 |
Family
ID=80050941
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
NO20201404A NO347369B1 (en) | 2020-12-18 | 2020-12-18 | Automatic Scanning System for Fish |
Country Status (6)
Country | Link |
---|---|
EP (1) | EP4240146A1 (en) |
AU (1) | AU2021403126A1 (en) |
CA (1) | CA3202150A1 (en) |
CL (1) | CL2023001767A1 (en) |
NO (1) | NO347369B1 (en) |
WO (1) | WO2022131929A1 (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
NO20230610A1 (en) | 2019-08-20 | 2022-02-16 | Aquaticode Ltd | Methods and systems for recognizing fish and traits of fish through non-invasive means |
NO20221145A1 (en) | 2021-11-23 | 2023-05-24 | Aquaticode Ltd | Sorting animals based on non-invasive determination of animal characteristics |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS5737387A (en) * | 1980-11-27 | 1982-03-01 | Fuji Electric Co Ltd | Sex discriminating method for fish and shellfish |
GB2201772A (en) * | 1986-10-16 | 1988-09-07 | Papirind Forskningsinst | An opto-electronic method for determining by length-measurement the quality of cultured fish and a device for implementing the method |
US20180177169A1 (en) * | 2015-06-10 | 2018-06-28 | Melbo International B.V. | Fishing System for Fishing for Shrimps, Fish Processing Line and Vessel Provided Therewith, and Method Therefor |
NO20170365A1 (en) * | 2017-03-13 | 2018-09-14 | Skala Maskon As | Method and system for automated vaccination and sorting alive fish |
JP6826356B1 (en) * | 2020-03-03 | 2021-02-03 | 樹 川邉 | Fish sex discriminator |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS5429797A (en) * | 1977-08-02 | 1979-03-05 | Fuji Electric Co Ltd | Male and female distinguishing of fish and shellfish |
JP2000004708A (en) * | 1998-06-23 | 2000-01-11 | Kobe Steel Ltd | Device for distinguishing male and female of fish |
NO331843B1 (en) * | 2010-06-24 | 2012-04-16 | Maskon As | Method and system for vaccination and sorting of fish |
WO2020002997A2 (en) | 2018-06-27 | 2020-01-02 | Radmantis Llc | Animal sensing system |
-
2020
- 2020-12-18 NO NO20201404A patent/NO347369B1/en unknown
-
2021
- 2021-12-17 WO PCT/NO2021/050271 patent/WO2022131929A1/en unknown
- 2021-12-17 EP EP21848306.3A patent/EP4240146A1/en active Pending
- 2021-12-17 AU AU2021403126A patent/AU2021403126A1/en active Pending
- 2021-12-17 CA CA3202150A patent/CA3202150A1/en active Pending
-
2023
- 2023-06-15 CL CL2023001767A patent/CL2023001767A1/en unknown
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS5737387A (en) * | 1980-11-27 | 1982-03-01 | Fuji Electric Co Ltd | Sex discriminating method for fish and shellfish |
GB2201772A (en) * | 1986-10-16 | 1988-09-07 | Papirind Forskningsinst | An opto-electronic method for determining by length-measurement the quality of cultured fish and a device for implementing the method |
US20180177169A1 (en) * | 2015-06-10 | 2018-06-28 | Melbo International B.V. | Fishing System for Fishing for Shrimps, Fish Processing Line and Vessel Provided Therewith, and Method Therefor |
NO20170365A1 (en) * | 2017-03-13 | 2018-09-14 | Skala Maskon As | Method and system for automated vaccination and sorting alive fish |
JP6826356B1 (en) * | 2020-03-03 | 2021-02-03 | 樹 川邉 | Fish sex discriminator |
Also Published As
Publication number | Publication date |
---|---|
AU2021403126A9 (en) | 2024-02-08 |
NO20201404A1 (en) | 2022-06-20 |
WO2022131929A1 (en) | 2022-06-23 |
CA3202150A1 (en) | 2022-06-23 |
AU2021403126A1 (en) | 2023-07-06 |
EP4240146A1 (en) | 2023-09-13 |
CL2023001767A1 (en) | 2023-12-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Nasirahmadi et al. | Implementation of machine vision for detecting behaviour of cattle and pigs | |
US11528885B2 (en) | Generating consensus feeding appetite forecasts | |
US8036429B2 (en) | System and method for measuring animals | |
EP2178362B1 (en) | Means and method for average weight determination and appetite feeding | |
Miranda et al. | A prototype to measure rainbow trout's length using image processing | |
CA3202150A1 (en) | Automatic scanning system for fish | |
US11089762B1 (en) | Methods for generating consensus biomass estimates | |
EP2027770B1 (en) | Method and apparatus for the automatic grading of condition of livestock | |
Olst et al. | Some aspects of the organization of fish schools | |
Puig-Pons et al. | Automatic Bluefin Tuna (Thunnus thynnus) biomass estimation during transfers using acoustic and computer vision techniques | |
Bhoj et al. | Image processing strategies for pig liveweight measurement: Updates and challenges | |
Vetter et al. | Characterization and management implications of silver carp (Hypophthalmichthys molitrix) jumping behavior in response to motorized watercraft. | |
Kugler et al. | Echo-acoustic scanning with noseleaf and ears in phyllostomid bats | |
JP3747308B2 (en) | Live fish measurement device in fish tank | |
NO20201382A1 (en) | A system for monitoring of dead fish | |
EP4183491A1 (en) | Sorting animals based on non-invasive determination of animal characteristics | |
Soltanzadeh et al. | A prototype system for real-time monitoring of Arctic char in indoor aquaculture operations: possibilities & challenges | |
JP7233688B2 (en) | Method and system for measuring substances in liquid | |
RU2777572C2 (en) | System for monitoring external fish parasite in aquaculture | |
WO2023027647A1 (en) | System and process in monitoring aquatic animals' growth and health status | |
Nasirahmadi | Development of automated computer vision systems for investigation of livestock behaviours | |
Browers et al. | The use of computed tomography in cetacean research air sac determination of Lagenorhyncus albirostris. Part 1 | |
Duan et al. | Comprehensive Analysis of Aquatic Animal Behavior Research—A Review | |
NO347741B1 (en) | System and method for sorting animals | |
Cui et al. | Fish Tracking, Counting, and Behaviour Analysis in Digital Aquaculture: A Comprehensive Review |