Abstract
With the advent of upcoming new patterns to handle (e.g., big data and semantic annotation), to encourage research to detect and identify objects, the development of the internet of things in the ocean requires the interconnection of all equipment (sensors) to observe the oceans. By serving as a common conceptualization among these tools, marine ontologies can lead to lower costs and better flexibility in marine data recognition and classification. To that end, marine pattern analysis literature (1991-2021) is used to create a sample network of records, comprising visual and textual features that can be annotated from video and image sequences, with the underwater parameters as the target of interest. The sample is split into ontological and machine learning (ML) datasets to build a prediction of the importance of data visualization techniques. The predicted suitability is strong with data classification that belongs to the machine learning dataset. However, the initial results from the study are encouraging, because ontologies' tools are proposed as automatic reasoning mechanisms.
TopBackground
Application of ontologies in ocean data grows out of an Artificial Intelligence (AI) engagement with marine data metrics of interoperability and reuse. Ontologies serve as such a tool and method to assess the added value robotic technology brings to the marine environment (autonomous underwater vehicles (AUVs) or (ocean floor observation systems) OFOSs). From a pattern recognition point of view, ontologies for describing sensors and sensor networks work in the context of Sensor Web applications. Knowledge representation in the Internet of Things (IoT) presents a general architecture of Sensor Web applications. And that is why it provides huge numbers of interconnected data across an extended variety of various ocean regions, which classifications depend on the specific context and resources of LinkedData (Fang, 2021).
By using ontological representation, the best technical progress, undertaken by a community to unambiguously set definitions and interconnect concepts in various fields, is captured. The use of ontologies for representing database entities has proven to be advantageous in the field of Pattern Analysis and Machine Intelligence (PAMI) (see Table 1).
Table 1. The main features provided by ontologies in support of PAMI
| Ontology feature | Utility in PAMI |
| Classes and relations | When ontology reasoning is applied to sensor data, RDF: type will be connected to a class name of an ontology |
| Domain vocabulary | Ontologies provide a domain vocabulary that can be exploited to create a dense network of relationships among the entities and server software applications and GIS |
| Metadata and descriptions | Biodiversity data, especially in the marine domain, have database entities represented as ontologies where these last are primarily used for metadata that describe raw data providing contextual information |
| Axioms and formal declarations | Ontology axioms and applied reasoning on them are related to the recognition of object
presence in a time interval |
Key Terms in this Chapter
Ontology: a structured set of terms and concepts that illustrate the meaning of an information field.
Spatial Data: Data related to georeferenced features, including points, lines, and polygons.
Data Classification: algorithmic categorization of data.
Deep Learning: a type of artificial intelligence based on a network of artificial neurons inspired by the human brain.
Computer Vision: a branch of artificial intelligence that deals with how computers can get a general idea of digital images or videos.
Industry 5.0: A long-term strategy to support the transformation of manufacturing sites into more digital and greener industrial solutions and processes.
Data Recognition: branch of artificial intelligence aiming to identify computational patterns from raw data.
Machine Learning: artificial intelligence technology allowing computers to learn without being programmed.
Image Analysis: recognition of the elements and information contained in an image.
Pattern Analysis: a set of techniques aimed at identifying rules that describe specific patterns in the data.