WO2021125923A1 - Method for automatically calibrating fish, combining image processing and artificial intelligence - Google Patents
Method for automatically calibrating fish, combining image processing and artificial intelligence Download PDFInfo
- Publication number
- WO2021125923A1 WO2021125923A1 PCT/MA2019/000021 MA2019000021W WO2021125923A1 WO 2021125923 A1 WO2021125923 A1 WO 2021125923A1 MA 2019000021 W MA2019000021 W MA 2019000021W WO 2021125923 A1 WO2021125923 A1 WO 2021125923A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- fish
- artificial intelligence
- image processing
- combining image
- automatically calibrating
- Prior art date
Links
- 241000251468 Actinopterygii Species 0.000 title claims abstract description 32
- 238000000034 method Methods 0.000 title claims abstract description 16
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 9
- 238000004364 calculation method Methods 0.000 claims abstract description 6
- 241000894007 species Species 0.000 description 5
- 241000276707 Tilapia Species 0.000 description 3
- 235000013305 food Nutrition 0.000 description 2
- 238000003306 harvesting Methods 0.000 description 2
- 238000009360 aquaculture Methods 0.000 description 1
- 244000144974 aquaculture Species 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000004513 sizing Methods 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
-
- 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
Definitions
- the present invention relates to the field of aquaculture. It relates to automatic fish calibration processes.
- Sorting by size of the fish may prove to be necessary in several situations, in particular those relating to the separation of faster growing fish from slower growing fish, (for example male tilapia and female tilapia). Also; Sizing of fish is important early in the life cycle of predatory fish, when the size range becomes too large, or when harvesting a few species such as fry, before storing them in grow-out ponds. Or quite simply, to choose fish that have reached marketing size.
- Sorting by size has various advantages such as reducing the loss of fish due to cannibalism, improving the efficiency of the distribution of complementary foods through an adequate food ration, improving the precision of the estimates of the stock for management purposes, reducing the proportion of small fish at harvest time from grow-out ponds and ultimately increasing production (e.g. through an increased proportion of faster growing males in growing ponds. 'tilapia farm)
- the species recognition and selective sorting system is built on a network of neurons. By this technique, the system makes decisions as could an operator who would have been taught to sort the fish by species and size.
- the recognition of a fish of a given species also makes it possible, via several memorized correction factors, to determine an approximate value of the weight of the fish and therefore to sort the fish by size.
- Recognition software can sort new species by learning.
- the method is based on a linear camera which makes it possible to record the images of the fish one by one. The system then analyzes these images, and calculates several descriptors relating to (calculation of size areas, regularity of contours, etc.) and makes it possible to deduce the classification of the fish.
- the present invention makes it possible to improve the prior state by proposing a method for sorting fish by size based on a calculation of the perimeters of the eyes.
- Said method is based on the driving of a batch of fish spread by a scraper adjustable in height according to the type to be treated, by means of a conveyor.
- a camera then makes it possible to capture images of the batch which will be analyzed based on artificial intelligence and image processing algorithms in order to classify the fish according to the perimeter of the eyes.
- a large fish will have a larger eye perimeter.
- the method makes it possible to determine the percentage of each size in a batch of fish.
- This method will have the advantage of treating a whole batch of fish at the same time efficiently (instead of a passage one by one), and of minimizing the execution time since the calculation of the perimeters is much less expensive. in terms of time than the calculation of the size areas. This then implies increasing the rate and improving the performance of automatic graders.
- FIGURE 1 is a block diagram of the overall system
- FIGURE 2 is a view of the method of detecting fish eyes and their classification.
- FIGURE 3 is a view of the result of sorting
- the batch of fish (1) passes through a conveyor (2), below a line camera (3) which makes it possible to capture images (4).
- the artificial intelligence and image processing algorithm then analyzes the captured images by detecting the eyes of the fish (5) which are then bypassed.
- the system calculates the perimeters (6) and determines as a result the percentage of each size existing in the batch (7).
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Zoology (AREA)
- Environmental Sciences (AREA)
- Marine Sciences & Fisheries (AREA)
- Animal Husbandry (AREA)
- Biodiversity & Conservation Biology (AREA)
- Image Processing (AREA)
Abstract
The invention relates to an automatic calibration method of fish, combining image processing and artificial intelligence based on the calculation of the perimeters of the eyes. This allows the rate of automatic calibration machines to be increased and their performance to be improved.
Description
Description : Description:
Procédé de calibrage automatique des poissons combinant le traitement d'image et l'intelligence artificielle. Automatic fish calibration process combining image processing and artificial intelligence.
Domaine Technique : Technical area :
[001] La présente invention concerne le domaine de l'aquaculture. Elle est relative aux procédés de calibrage automatique de poissons. The present invention relates to the field of aquaculture. It relates to automatic fish calibration processes.
Technique antérieure : Prior technique:
[002] Le triage par taille des poissons peut s'avérer nécessaire dans plusieurs situations notamment celles qui concerne la séparation des poissons à croissance plus rapide des poissons à croissance plus lente, (par exemple les tilapias mâles et les tilapias femelles). Egalement ; Le calibrage des poissons est important au début du cycle de vie de poissons prédateurs, lorsque l'éventail des tailles devient trop important ou lors de la récolte de quelques espèces comme les alevins, avant de les stocker dans des étangs de grossissement. Ou tout simplement, pour choisir les poissons ayant atteint la taille de commercialisation. [002] Sorting by size of the fish may prove to be necessary in several situations, in particular those relating to the separation of faster growing fish from slower growing fish, (for example male tilapia and female tilapia). Also; Sizing of fish is important early in the life cycle of predatory fish, when the size range becomes too large, or when harvesting a few species such as fry, before storing them in grow-out ponds. Or quite simply, to choose fish that have reached marketing size.
[003] Le triage par taille présente différents avantages comme la réduction des pertes de poissons dues au cannibalisme, l'amélioration de l'efficacité des distributions d'aliments de complément par une ration alimentaire adéquate , l'amélioration de la précision des estimations du stock pour des fins de gestion , la réduction de la proportion de petits poissons au moment de la récolte d'étangs de grossissement et finalement l'augmentation de la production (par exemple par une proportion accrue de mâles à croissance plus rapide dans les étangs d'élevage de tilapias) [003] Sorting by size has various advantages such as reducing the loss of fish due to cannibalism, improving the efficiency of the distribution of complementary foods through an adequate food ration, improving the precision of the estimates of the stock for management purposes, reducing the proportion of small fish at harvest time from grow-out ponds and ultimately increasing production (e.g. through an increased proportion of faster growing males in growing ponds. 'tilapia farm)
[004] Il est connu que plusieurs procédés ont automatisé la tâche du triage en proposant une solution intelligente alliant reconnaissance et intelligence artificielle. Le système de reconnaissance, et de tri sélectif des espèces, est bâti sur un réseau de neurones. Par cette technique, le système prend des décisions comme pourrait le faire un opérateur auquel on aurait appris à trier les poissons par espèce et par taille. La reconnaissance d'un poisson d'une espèce donnée permet également, via plusieurs facteurs de correction mémorisés, de déterminer une valeur approchée du poids du poisson et donc de trier les poissons par calibre. Le logiciel de reconnaissance peut trier de nouvelles espèces par apprentissage.
[005] Le procédé repose sur une caméra linéaire qui permet de relever les images des poissons un par un. Le système analyse ensuite ces images, et calcul plusieurs descripteurs relatifs aux (calcul d'aires de tailles, régularité des contours ...) et permet de déduire la classification des poissons. It is known that several methods have automated the task of sorting by proposing an intelligent solution combining recognition and artificial intelligence. The species recognition and selective sorting system is built on a network of neurons. By this technique, the system makes decisions as could an operator who would have been taught to sort the fish by species and size. The recognition of a fish of a given species also makes it possible, via several memorized correction factors, to determine an approximate value of the weight of the fish and therefore to sort the fish by size. Recognition software can sort new species by learning. The method is based on a linear camera which makes it possible to record the images of the fish one by one. The system then analyzes these images, and calculates several descriptors relating to (calculation of size areas, regularity of contours, etc.) and makes it possible to deduce the classification of the fish.
Exposé de l'invention : Disclosure of the invention:
[006] La présente invention permet d'améliorer l'état antérieur en proposant un procédé de triage par taille des poissons basé sur un calcul des périmètres des yeux. Ledit procédé repose sur l'entrainement d'un lot de poissons étalé par une raclette réglable en hauteur selon le type à traiter, grâce à un convoyeur. Une caméra permet ensuite de capturer des images du lot qui seront analysées en se basant sur des algorithmes d'intelligence artificielle et de traitement d'images afin de classifier les poissons selon le périmètre des yeux. Un poisson de grande taille aura un périmètre d'œil plus grand. Finalement, Le procédé permet de déterminer le pourcentage de chaque calibre dans un lot de poissons. The present invention makes it possible to improve the prior state by proposing a method for sorting fish by size based on a calculation of the perimeters of the eyes. Said method is based on the driving of a batch of fish spread by a scraper adjustable in height according to the type to be treated, by means of a conveyor. A camera then makes it possible to capture images of the batch which will be analyzed based on artificial intelligence and image processing algorithms in order to classify the fish according to the perimeter of the eyes. A large fish will have a larger eye perimeter. Finally, the method makes it possible to determine the percentage of each size in a batch of fish.
[007] Cette méthode aura l'avantage de traiter tout un lot de poissons à la fois efficacement (au lieu d'un passage un par un), et de minimiser le temps d'exécution vu que le calcul des périmètres est bien moins coûteux en termes de temps que le calcul des aires de tailles. Ceci implique alors, l'augmentation de la cadence et l'amélioration de la performance des calibreuses automatiques. This method will have the advantage of treating a whole batch of fish at the same time efficiently (instead of a passage one by one), and of minimizing the execution time since the calculation of the perimeters is much less expensive. in terms of time than the calculation of the size areas. This then implies increasing the rate and improving the performance of automatic graders.
[008] Dans les dessins qui illustrent l'invention, In the drawings which illustrate the invention,
La FIGURE 1 est un schéma synoptique du système global FIGURE 1 is a block diagram of the overall system
La FIGURE 2 est une vue de la méthode de détection des yeux de poissons et leur classification. FIGURE 2 is a view of the method of detecting fish eyes and their classification.
La FIGURE 3 est une vue du résultat du triage FIGURE 3 is a view of the result of sorting
[009] En se référant aux dessins, on verra que le lot de poissons (1) passe à travers un convoyeur (2), en dessous d'une caméra linéaire (3) qui permet de capturer des images (4). L'algorithme d'intelligence artificielle et traitement d'images analyse ensuite les images capturées en détectant les yeux des poissons (5) qui sont ensuite contournés. Le système calcul les périmètres (6) et détermine comme résultat le pourcentage de chaque calibre existant dans le lot (7).
Referring to the drawings, it will be seen that the batch of fish (1) passes through a conveyor (2), below a line camera (3) which makes it possible to capture images (4). The artificial intelligence and image processing algorithm then analyzes the captured images by detecting the eyes of the fish (5) which are then bypassed. The system calculates the perimeters (6) and determines as a result the percentage of each size existing in the batch (7).
Claims
1. Procédé de calibrage automatique de poissons combinant le traitement d'image et l'intelligence artificielle caractérisé en ce que l'algorithme d'analyse se base sur le calcul des périmètres des yeux de poissons. 1. Method for automatic calibration of fish combining image processing and artificial intelligence, characterized in that the analysis algorithm is based on the calculation of the perimeters of the eyes of fish.
2. Procédé de calibrage automatique de poissons combinant le traitement d'image et l'intelligence artificielle selon la revendication 1 caractérisé en ce que les poissons passent en dessous de la caméra sous forme de lot. 2. A method of automatically calibrating fish combining image processing and artificial intelligence according to claim 1, characterized in that the fish pass below the camera in the form of a batch.
3. Procédé de calibrage automatique de poissons combinant le traitement d'image et l'intelligence artificielle selon la revendication 1 caractérisé en ce que l'algorithme donne comme résultat le pourcentage de chaque calibre dans le lot.
3. A method of automatically calibrating fish combining image processing and artificial intelligence according to claim 1, characterized in that the algorithm gives as a result the percentage of each caliber in the batch.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/MA2019/000021 WO2021125923A1 (en) | 2019-12-20 | 2019-12-20 | Method for automatically calibrating fish, combining image processing and artificial intelligence |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/MA2019/000021 WO2021125923A1 (en) | 2019-12-20 | 2019-12-20 | Method for automatically calibrating fish, combining image processing and artificial intelligence |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2021125923A1 true WO2021125923A1 (en) | 2021-06-24 |
Family
ID=69467621
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/MA2019/000021 WO2021125923A1 (en) | 2019-12-20 | 2019-12-20 | Method for automatically calibrating fish, combining image processing and artificial intelligence |
Country Status (1)
Country | Link |
---|---|
WO (1) | WO2021125923A1 (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2559336A1 (en) * | 2011-08-19 | 2013-02-20 | Asociación Industrial de Óptica, Color e Imagen - AIDO | System and method for automatic classification of fish alevins by means of optical methods |
WO2018222048A1 (en) * | 2017-05-29 | 2018-12-06 | Ecotone As | Method and system for underwater hyperspectral imaging of fish |
US20190228218A1 (en) * | 2018-01-25 | 2019-07-25 | X Development Llc | Fish biomass, shape, and size determination |
WO2020076147A1 (en) * | 2018-10-11 | 2020-04-16 | Universite Internationale De Rabat | Method for automatically calibrating fish combining image processing and artificial intelligence |
-
2019
- 2019-12-20 WO PCT/MA2019/000021 patent/WO2021125923A1/en active Application Filing
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2559336A1 (en) * | 2011-08-19 | 2013-02-20 | Asociación Industrial de Óptica, Color e Imagen - AIDO | System and method for automatic classification of fish alevins by means of optical methods |
WO2018222048A1 (en) * | 2017-05-29 | 2018-12-06 | Ecotone As | Method and system for underwater hyperspectral imaging of fish |
US20190228218A1 (en) * | 2018-01-25 | 2019-07-25 | X Development Llc | Fish biomass, shape, and size determination |
WO2020076147A1 (en) * | 2018-10-11 | 2020-04-16 | Universite Internationale De Rabat | Method for automatically calibrating fish combining image processing and artificial intelligence |
Non-Patent Citations (1)
Title |
---|
ZHUHUA HU ET AL: "Fish eye recognition based on weighted constraint AdaBoost and pupil diameter automatic measurement with improved Hough circle transform", NONGYE GONGCHENG XUEBAO = TRANSACTIONS CHINESE SOCIETY OF AGRICULTURAL ENGINEERING, vol. 33, no. 23, 1 December 2017 (2017-12-01), CN, pages 226 - 232, XP055728521, ISSN: 1002-6819, DOI: 10.11975/j.issn.1002-6819.2017.23.029 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Payne et al. | Estimation of mango crop yield using image analysis–segmentation method | |
Amatya et al. | Detection of cherry tree branches with full foliage in planar architecture for automated sweet-cherry harvesting | |
Arakeri et al. | Computer vision based robotic weed control system for precision agriculture | |
US10130039B2 (en) | Agricultural harvester including feedback control of chopping parameters | |
KR101598898B1 (en) | The system for precaution for paralichthys olivaceus disease using and an analysis of image and sound | |
EP3226672B2 (en) | System and method for the automatic adjustment of the height of an agricultural implement using 3d reconstruction | |
RU2004132543A (en) | METHOD AND SYSTEM OF AUTOMATIC CONTROL OF AGRICULTURAL MACHINE, HARVESTING MACHINE AND MACHINE FOR TILLING SOIL | |
WO2016131124A1 (en) | Systems, devices, and methods for detecting fertility and gender of unhatched eggs | |
CN112400773B (en) | Greenhouse fry intelligent feeding device and method based on machine vision technology | |
JP2022506699A (en) | How to control the operation of the machine that harvests root vegetables | |
JP2022509749A (en) | How to control the operation of the machine that harvests root vegetables | |
Fontaine et al. | Development of line-detection algorithms for local positioning in densely seeded crops | |
WO2020076147A1 (en) | Method for automatically calibrating fish combining image processing and artificial intelligence | |
JP2022506703A (en) | How to control the operation of the machine that harvests root vegetables | |
FR3071644A1 (en) | METHOD AND DEVICE FOR CLASSIFYING PLANTS | |
FR3036830A1 (en) | SYSTEM AND METHOD FOR ESTIMATING A HARVEST VOLUME IN A WINE FARMING | |
TW202002772A (en) | Intelligent deep learning agricultural and fishery cultivation system to make processing operations, such as modification of cultivation environmental status corresponding response for growth cultivation, according to the organism growth and environment statuses | |
CN111008970A (en) | Hyperspectral image acquisition system-based method for detecting impurities in preserved vegetables | |
WO2021125923A1 (en) | Method for automatically calibrating fish, combining image processing and artificial intelligence | |
Baur et al. | Noninvasive monitoring system for Tenebrio molitor larvae based on image processing with a watershed algorithm and a neural net approach | |
WO2019055405A1 (en) | Grain handling system and method | |
Khatri et al. | Computer Vision and Image Processing for Precision Agriculture | |
Setyawan et al. | Machine vision-based urban farming growth monitoring system | |
US11785889B2 (en) | System and method for determining an indicator of processing quality of an agricultural harvested material | |
Dyshekov et al. | Principles of functioning of the autonomous device for weed control for precision agriculture |
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: 19848823 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 19848823 Country of ref document: EP Kind code of ref document: A1 |