Prosjektnummer
901185
Anatomy of wild and farmed fish species
Results achieved
The main goal of this project has been to assemble a relevant dataset as a basis for development of processing equipment for the fish industry and revision of current processing methods.
The main goal of this project has been to assemble a relevant dataset as a basis for development of processing equipment for the fish industry and revision of current processing methods.
Further objectives of this project have been to image bones in whole fish and fillets in 9 different species and to provide detailed information about the size, orientation and location of pinbones and the walking stick bone in fillets. For each species 2–4 whole fish and 2–4 fillets were CT scanned and analysed. The bones and fillet were segmented and length, thickness, position and orientation of the bones were estimated.
Comparison with manual control measurements for some of the fillets showed that all the bones were detected, but there were some deviations in the length and thickness measures. The CT pinbone measures gives in average 0.2 mm thicker bone than the manual measures andt the CT length measures gives in average 3 mm shorter bone than the manual measures. In the project report a simple analysis of the recorded data is performed.
SINTEF presents in the report initial analysis of the data. However, the goal of this project has primarily been to assemble a relevant dataset as a basis for further analysis. To enable independent analysis, all data is made available electronically for download. See Appendix A1 in the report for download details.
Norsk fiskerinæring er avhengig av effektiv og automatisert produksjon for å være konkurransedyktig. Gjennom prosjektet har SINTEF IKT gjort CT-scanning av hel fisk og filét av 9 ulike arter av stor kommersiell verdi. Datasettene vil være utgangspunkt for å utvikle ny produksjonsteknologi som kan bidra til fullautomatiserte linjer for produksjon av hvitfiskfilét, og gjøres tilgjengelig for alle aktuelle teknologileverandører.
-
Faktaark: Anatomy of wild and farmed fish species
SINTEF ICT. Faktaark. 29. juni 2016. Av Marianne Bakken og Helene Schulerud.
-
Rapport: Apricot 2: CT imaging of hole fish and fillets
SINTEF ICT. Rapport A27733. Juni 2016. Av Marianne Bakken og Helene Schulerud.
-
Sluttrapport: Anatomy of wild and farmed fish species
SINTEF ICT. Juni 2016. Av Marianne Bakken og Helene Schulerud.
Background
Automation of fish processing has been recognized as a key factor in maintaining strong and competitive fish processing industry within the Nordic countries. 3-D imaging information of fish anatomy is an important tool in development of innovative processing methods and in adjusting new technology to anatomy of different fish species.
In the past, automation of manual operations has frequently been focused on a single processing step. The overall process perspective is sometimes lacking, and needed in deciding whether the whole process could and should be reorganized, for improving factors such as yield and value of products.
Whole fish / Gutted with head
3D-CT of whole fish anatomy will provide mechanical engineers essential tool to review current processing methods and identify new ways primary processing of the fish, such as decapitation and filleting. Digital information on structural alignment of bones with the fish, and provides view of proportion of different tissues.
Automation of fish processing has been recognized as a key factor in maintaining strong and competitive fish processing industry within the Nordic countries. 3-D imaging information of fish anatomy is an important tool in development of innovative processing methods and in adjusting new technology to anatomy of different fish species.
In the past, automation of manual operations has frequently been focused on a single processing step. The overall process perspective is sometimes lacking, and needed in deciding whether the whole process could and should be reorganized, for improving factors such as yield and value of products.
Whole fish / Gutted with head
3D-CT of whole fish anatomy will provide mechanical engineers essential tool to review current processing methods and identify new ways primary processing of the fish, such as decapitation and filleting. Digital information on structural alignment of bones with the fish, and provides view of proportion of different tissues.
Fillets
Automation of pin bone detection and removal by combined system of x-rays and water jet cutting (FleXicut) is one of the latest inventions in the whitefish industry. The main focus has been on cod but producers have emphasized the need to transfer the technology to processing of other species. The location of the pin bones, number and alignment in the fillets varies between different fish species. Therefore it is essential to implement for studies on fish anatomy, by techniques such as 3D CT-scanning (in similar way as done in the Apricot anatomy Project, ‘APRICOT anatomy: Avbildning og kvantifisering av tykkfiskbein i ulike fiskeslag’ (FHF-900814).
Objectives
To provide detailed information about fish anatomy, the skeleton of whole fish (with head and the size, orientation and location of internal bones (e.g. pin bones) in fillets.
To provide detailed information about fish anatomy, the skeleton of whole fish (with head and the size, orientation and location of internal bones (e.g. pin bones) in fillets.
Expected project impact
The project will assemble a relevant dataset as a basis for development in processing equipment for the fish industry and revision of current processing methods.
The project will assemble a relevant dataset as a basis for development in processing equipment for the fish industry and revision of current processing methods.
Project design and implementation
Different fish species, gutted and filleted (as listed in following table) will be CT-scanned (3D):
Different fish species, gutted and filleted (as listed in following table) will be CT-scanned (3D):
• Gutted fish with head for selected species- 4 fish per species
o 2 fishes of medium size (M)
o 2 fishes of small size (S)
• Fillet - 4 fillets from 4 different individuals per species
o 2 fillets of medium size (M)
o 2 fishes of small size (S)
o 2 fishes of medium size (M)
o 2 fishes of small size (S)
• Fillet - 4 fillets from 4 different individuals per species
o 2 fillets of medium size (M)
o 2 fishes of small size (S)
The aim is to cover higher number of species (average size) within this project rather than focus on individual variation. The purpose is to have basic data on parameters such as alignment of pin bones, number of pin bones etc. For example there is a significant variation in number of pin bones (length of pin bone frame) between different gadoid species. To cover variation due to size, condition factor and gender etc. a higher number such as 20 fishes per group would be needed, which is outside the scope of this project.
3D CT-scanning (16 hours / 64 images):
• Skeleton of the whole fish – distinguishing between bones and muscle.
• Pin bones in untrimmed fillets (number, size, position and orientation in fillets). Other internal bones (e.g. spåmannsbein) (number, size, position and orientation in fillets).
• Dimensions of fish (L, W, H) / 3D shape that can be used for model gutted fish and fillets, in development of processing equipment but also other purposes such is for estimation of cooling profiles (not part of this project).
• Skeleton of the whole fish – distinguishing between bones and muscle.
• Pin bones in untrimmed fillets (number, size, position and orientation in fillets). Other internal bones (e.g. spåmannsbein) (number, size, position and orientation in fillets).
• Dimensions of fish (L, W, H) / 3D shape that can be used for model gutted fish and fillets, in development of processing equipment but also other purposes such is for estimation of cooling profiles (not part of this project).
The fish should be scanned as fresh as possible (less than 5 days from catch). Gutted fish will be weighted and the fork length measured. Results on pin bones (and other internal bones) will be compared with manual control measurements (length, thickness) to check whether all the bones are detected and to evaluate relationship between bone size and detection by CT-scanning.
Marel will in collaboration with Norway Seafood plan and collect fishes for CT-scanning. SINTEF will be responsible for conducting CT-scanning, manual measurements, data analysis and report writing.
Project organization
SINTEF is responsible for the general project management, for monitoring the progress throughout the project lifetime and re-evaluation of project plan (with steering committee). SINTEF will summarize results from the measurements, with details of experimental materials and prepare a final report for FHF.
SINTEF is responsible for the general project management, for monitoring the progress throughout the project lifetime and re-evaluation of project plan (with steering committee). SINTEF will summarize results from the measurements, with details of experimental materials and prepare a final report for FHF.
Report describing main results of the project, will be delivered to the project consortium and FHF.
-
Sluttrapport: Anatomy of wild and farmed fish species
SINTEF ICT. Juni 2016. Av Marianne Bakken og Helene Schulerud.