Back to search

MAROFF-2-Maritim virksomhet og offsh-2

LIACi - Lifecycle Inspection, Analysis and Condition information system

Alternative title: LIACi - Beslutningsstøttesystem for videobasert inspeksjon, analyse og tilstandsvurdering

Awarded: NOK 11.8 mill.

Project Manager:

Project Number:

317854

Project Period:

2021 - 2023

Organisation:

The overall goal of LIACi is to develop an adaptive decision support system that improves efficiency of video-based inspections with automatic report generation and extended use of inspection data for decision support. POSICOM will use the leading competence of SINTEF and NTNU to extend its Seekuence video management system with new AI/ML features. This will give POSICOM a unique advantage in serving the maritime and fish farming market segments represented by DNV, VUVI, Island Offshore and Mainstay AS. The main objective is to take the lead in the transformation from manual inspection to remote inspection by offering the most efficient video-based solution for data gathering, storage, retrieval, and reporting with extended opportunities for contextualized analysis. The innovation will be realized by using Deep Learning techniques and emerging concepts related to Active Learning, Cooperative Machine Learning, and Explainable AI in combination with Knowledge Graphs. The planned objectives and innovations in the project are: 1. Machine Learning-based Collaborative Video Tagging Module: A new service that does real-time tagging of video sequences (e.g., classifying and labelling paint peel and marine growth in a ship inspection video). 2. Video Contextualization Module: A new service that integrates various data sources that can provide additional value and context to tagged video sequences. 3. Query Module for Report Generation: A new service that allows end users to generate tailor made reports for a specific inspection or maintenance procedure. 4. Business Model for offering LIACi on the Marketplace for Veracity by DNV.

Outcomes: 1. Machine Learning-based Collaborative Video Tagging Module: A new service that does real-time tagging of video sequences. Several Machine Learning (ML) models were developed to support in-water ship inspections and fish farm inspections. For ship inspections, ML models were developed to classify, detect and segment the presence of various objects (e.g., propeller and anode) and findings (e.g., corrosion, paint peel and marine growth) in ship inspection videos. For fish farm inspections, ML models were developed to detect holes in the fish nets, as well as marine growth. 2. Video Contextualization Module: A new service that integrates various data sources that can provide additional value and context to tagged video sequences. It uses a graph data model to store information about a ship, its inspections, inspection findings, images, and other related information (e.g., GA drawings) to support analytics that allows us to search and compare different findings of a ship over time. 3. Query Module for Report Generation: A new flexible reporting service that allows end users to generate tailor made reports for a specific inspection or maintenance procedure. 4. Business Model for offering LIACi on the Marketplace for Veracity by DNV. The LIACi-enhanced version of the Seekuence system is available on the Veracity markeplace (https://store.veracity.com/seekuence-video-management-system). Impacts: POSICOM has developed and introduced several new software features in their Seekuence system as part of the LIACi project. Some of the prototypes have been successfully integrated as new AI/ML software features in Seekuence. The prototypes have been tested and evaluated in five different business cases. 1. Intermediate In Water Survey for class renewal. 2. Marine growth estimation on ship hull. 3. Inspection of fish farm nets and mooring lines. 4. Inspection of offshore mooring lines (chain and rope). 5. Corals mapping in the Trondheim fjord. Through these demonstrators, POSICOM has been able to attract new customers in the shipping, aquaculture and healthcare sectors. LIACi presented at Sjøsikkerhetskonferansen 2022 and 2023, and Nor-Shipping 2023, resulting in several follow-up meetings with the Norwegian Maritime Authority and potential customers. POSICOM will continue to follow up other active leads in the maritime sector in 2024. We will use the Veracity marketplace to market our services to the DNV network of customers and service providers. Assuming 50% of them are relevant, we can reach some 2500 potential customers in shipping. In 2024, POSICOM signed a new contract with a service company that will install and use Seekuence on 30 vessels in the aquaculture sector. We expect that this will be a good reference in the fish farming market. There is also interest for AI/ML in the healthcare market, and we have started discussions with existing customers on how they could benefit from such technology.

Companies across oil & gas, maritime and fish farming value chains are seeking quicker, more accessible, and cost-effective ways to ensure technical safety and performance of projects and operations. Increasingly more of those companies are using digital technologies to virtually bring inspectors and surveyors to sites in order to witness and verify the quality and integrity of equipment and assets to company specifications or industry standards. The overall goal of LIACi is to develop an adaptive decision support system that improves efficiency of video-based inspections with automatic report generation and extended use of inspection data for decision support. Posicom will use the leading competence of SINTEF and NTNU to extend its video inspection system with these features; a) AI supported tagging of objects and events, and b) Contextualization of video with additional information. This will give Posicom a unique advantage in serving the maritime and oil & gas market segments represented by DNVGL, VUVI, Island Offshore and Kongsberg Ferrotech.

Publications from Cristin

No publications found

No publications found

Funding scheme:

MAROFF-2-Maritim virksomhet og offsh-2