ABS-FishCount: An Agent-Based Simulator of Underwater Sensors for Measuring the Amount of Fish

Abstract

Underwater sensors provide one of the possibilities to explore oceans, seas, rivers, fish farms and dams, which all together cover most of our planet’s area. Simulators can be helpful to test and discover some possible strategies before implementing these in real underwater sensors. This speeds up the development of research theories so that these can be implemented later. In this context, the current work presents an agent-based simulator for defining and testing strategies for measuring the amount of fish by means of underwater sensors. The current approach is illustrated with the definition and assessment of two strategies for measuring fish. One of these two corresponds to a simple control mechanism, while the other is an experimental strategy and includes an implicit coordination mechanism. The experimental strategy showed a statistically significant improvement over the control one in the reduction of errors with a large Cohen’s d effect size of 2.55.

Implementation of Whale Optimization for Budding Healthiness of Fishes with Preprocessing Approach

“…Since most of the process is implemented with sensors, a new system model is necessary; therefore, a semiconductor sensor model has been created [ 14 ], and it is integrated with machine vision algorithms. Besides, this system an agent-based simulator has also been introduced in the current system process that provides an inherent mechanism for monitoring fishes [ 15 ]. This type of model is having many hidden layers, and all these layers are fulfilled only using the cost of communication channels.…”

Environmental monitoring tools and strategies in salmon net‐pen aquaculture

“…Traditionally, biomass monitoring has been accomplished somewhat crudely by calculating the average size and weight of periodically collected fish from net pens minus countable dead fish (Costa et al, 2006). Accurate estimates of fish length, width, height, girth, thickness, and mass are now possible using noninvasive underwater video‐based sensors, stereo image analysis, and agent‐based and multiagent‐based simulators to interpret the data (García‐Magariño et al, 2017; da Silva Vale et al, 2020; Y. S. Yang et al, 2011). In turn, these parameters can be extrapolated or used in individual‐based growth models to gain greater insights into the overall fish condition that are relevant to animal welfare guidance found in aquaculture certification programs (Giménez‐Candela et al, 2020).…”

“…Traditionally, biomass monitoring has been accomplished somewhat crudely by calculating the average size and weight of periodically collected fish from net pens minus countable dead fish (Costa et al, 2006). Accurate estimates of fish length, width, height, girth, thickness, and mass are now possible using noninvasive underwater video-based sensors, stereo image analysis, and agent-based and multiagent-based simulators to interpret the data (García-Magariño et al, 2017;da Silva Vale et al, 2020;Y. S. Yang et al, 2011).…”

ABS‐DDoS: An Agent‐Based Simulator about Strategies of Both DDoS Attacks and Their Defenses, to Achieve Efficient Data Forwarding in Sensor Networks and IoT Devices

“…One of these application is to monitor crop fields to irrigate and fumigate some specific areas when necessary [1]. In addition, underwater SNs are useful for assessing amounts of fish in fish farms [2]. This can be useful for delivering the right amount of food for properly feeding fishes in aquaculture without generating unnecessary food wastage.…”