Use of Neural Networks and Computer Vision for Spill and Waste Detection in Port Waters: An Application in the Port of Palma (MaJorca, Spain)

Abstract

Water quality and pollution is the main environmental concern for ports and adjacent coastal waters. Therefore, the development of Port Environmental Management systems often relies on water pollution monitoring. Computer vision is a powerful and versatile tool for an exhaustive and systematic monitoring task. An investigation has been conducted at the Port of Palma de Mallorca (Spain) to assess the feasibility and evaluate the main opportunities and difficulties of the implementation of water pollution monitoring based on computer vision. Experiments on surface slicks and marine litter identification based on random image sets have been conducted. The reliability and development requirements of the method have been evaluated, concluding that computer vision is suitable for these monitoring tasks. Several computer vision techniques based on convolutional neural networks were assessed, finding that Image Classification is the most adequate for marine pollution monitoring tasks due to its high accuracy rates and low training requirements. Image set size for initial training and the possibility to improve accuracy through retraining with increased image sets were considered due to the difficulty in obtaining port spill images. Thus, we have found that progressive implementation can not only offer functional monitoring systems in a shorter time frame but also reduce the total development cost for a system with the same accuracy level.

The Contribution of Open Source Software in Identifying Environmental Crimes Caused by Illicit Waste Management in Urban Areas

“…Monitoring urban ecosystems through computer vision not only provides real-time data on environmental health, but also supports the design of targeted conservation strategies [52]. For example, the identification of critical habitats or the assessment of the impacts of human activities becomes more efficient, allowing for more careful and sustainable management of natural resources [53,54].…”

Novel tools for port environmental management systems

“…These algorithms were developed and refined in SPILLCONTROL project, in previous stages of research. The first algorithm (V1) is the compensated image algorithm previously designed and utilized in study case presented in Chapter 4 [Morell, 2023]. This algorithm forms the foundation of the algorithm development process, providing valuable insights and serving as a benchmark for comparison.…”

“…For this experiment, algorithms were trained with different sets of images, and the training cost was evaluated in terms of the number of training epochs required and the reliability of the resulting algorithm using the error index proposed in the previous case [Morell 2023]. The algorithms were trained with image sets of the following resolutions: 0.043 MP, 0.120 MP, 0.480 MP, and 1.080 MP.…”

“…In the domain of environmental monitoring, remote sensing technology has proven effective in providing spatially synoptic and near real-time measurements for the detection and management of pollutants, including suspended solid waste and suspended sediments [Ciappa 2022, Hafeez et al 2019]. Moreover, recent contributions in waste and pollutant detection have utilized Image Classification techniques based on deep convolutional networks [Panwar et al 2020, Jiao et al 2018, Morell 2023]. Such approaches have achieved success in addressing waste detection in large surface areas.…”

“…Also, the effect of the original resolution of the image on the classification accuracy has been analyzed. To evaluate the algorithm's performance specifically for port environmental management applications, a novel performance index proposed on a former work [Morell 2023] reproduced in chapter 4, the error index, has been considered. The image sets were collected from Es Portitxol Port, located in Palma de Majorca, a recreational port where users and citizen are highly sensitive to water quality degradation events.…”