An Automatically Learning andDiscovering Human Fishing

An Automatically Learning and

Discovering Human Fishing Behaviors Scheme for CPSCN

ABSTRACT The cyber-physical-social (CPS) computing and Networking is a human centric and holistic computing framework that need convert the low-level data of physical, cyber, and social worlds into higherlevel information which can provide insights and help humans make complex decisions. Here we focus on human fishing behavior recognition for Vessel Monitoring Systems (VMS), an application of CPS. And the recognition of fishing behavior is the key task for studying human fishing activities, monitoring illegal fishing and protecting fishery resources.

However, VMS data basically consist of sequentially recorded position information and do not directly indicate whether a fisherman is fishing or not; thus, converting these low-level CPS data into intuitive information to humans is the primary task. In this paper, an identification model based on multi-step clustering algorithm (MSC-FBI) is proposed to automatically learn and discover fishing behaviors at sea. First, a temporal-spatial distance model is established; then, an improved multistep clustering algorithm is used to identify human fishing behaviors, and finally, the patterns of different behaviors are extracted from the trajectory and the unsupervised behavior learning model is established.

Using this method, many experiments on different fishing trajectory data were implemented and compared to a traditional identification method based on the Gaussian Mixture Model (GMM-FBI). The experimental results illustrate the proposed model’s superior performance.

I. INTRODUCTION fisherman is fishing or not. Those multi-modal data generated by heterogeneous sensors bring new challenges for human

physical devices, cyber systems and social networks to provide fishing vessel navigation, safety rescue, fishery production and maritime monitoring services.

In this distributed wireless sensor networks (DWSNs), the trajectory data is always massive and consumes a lot of resources [3]- [5]. And VMS data basically consist of sequentially recorded position information and do not directly indicate whether a portant to fisheries scientist, policy-makers and fishermen [8]. Because the technology of discovering human fishing behaviours from vessel trajectory is not only fundamental for studying the dynamic distribution of fishery resources and the impact of human fishing activities on marine environment, but it is also a key element for policy-makers to fight against

VOLUME 4, 2016 1

MS is an intelligent maritime transportation Cyberfishing event monitoring [6], [7]. In addition, discovering

Physical-social Systems [1], [2], which combines the

human fishing behaviours accurately and efficiently is im-

illegal fishing activities and introduce new fishing protected Areas. For fishermen, they may tend to learn about the fishing intensity of each area to make efficient fishing plans.

Nowadays, it is easier for people to share and obtain useful fishing information through CPS system than ever before. To analyze those large-scale trajectory data, the CPS computing platform need provide fault tolerant computing [9], high-performance computing [10], parallel methods [11], [12] to apply real-time service. And recognizing human fishing behaviors based on trajectory is a main category of CPS computing which convert low-level trajectory data into