Development of Coral Investigation System Based on Semantic Segmentation of Single-Channel Images

Abstract

Among aquatic biota, corals provide shelter with sufficient nutrition to a wide variety of underwater life. However, a severe decline in the coral resources can be noted in the last decades due to global environmental changes causing marine pollution. Hence, it is of paramount importance to develop and deploy swift coral monitoring system to alleviate the destruction of corals. Performing semantic segmentation on underwater images is one of the most efficient methods for automatic investigation of corals. Firstly, to design a coral investigation system, RGB and spectral images of various types of corals in natural and artificial aquatic sites are collected. Based on single-channel images, a convolutional neural network (CNN) model, named DeeperLabC, is employed for the semantic segmentation of corals, which is a concise and modified deeperlab model with encoder-decoder architecture. Using ResNet34 as a skeleton network, the proposed model extracts coral features in the images and performs semantic segmentation. DeeperLabC achieved state-of-the-art coral segmentation with an overall mean intersection over union (IoU) value of 93.90%, and maximum F1-score of 97.10% which surpassed other existing benchmark neural networks for semantic segmentation. The class activation map (CAM) module also proved the excellent performance of the DeeperLabC model in binary classification among coral and non-coral bodies.

Deteksi Area Kerusakan Pada Citra Terumbu Karang Akibat Coral Bleaching Berbasis Pengolahan Citra Digital

“…Namun akurasinya masih cukup rendah karena tidak mencapai 50%. Selanjutnya, (Song et al, 2021) melakukan penelitian terkait segmentasi citra terumbu karang dengan menggunakan citra RGB dan citra spectral terumbu karang. Metode segmentasi secara semantik dilakukan dalam penelitian ini menggunakan model jaringan syaraf tiruan.…”

Sensor Data Fusion Based on Deep Learning for Computer Vision Applications and Medical Applications

“…The work in [ 2 ] explored coral’s importance for marine ecosystems and proposed a convolutional neural network (CNN)-based model termed DeeperLabC to enhance the performance efficacy of underwater monitoring. The proposed DeeperLabC is one of the most effective techniques for automatically examining coral in underwater photos with the help of semantic segmentation to monitor marine pollution brought on by global environmental changes.…”

Investigating the rate of turbidity impact on underwater spectral reflectance detection

“…In the past decade, the concept of underwater spectral imaging technology has been developed by combining two-dimensional spatial information with one-dimensional spectral data, revealing the spectral features hidden in the narrow band (Polerecky et al, 2009). Because of its advantages of high spectral and spatial resolution, it has shown great potential in remote sensing applications (Yasir et al, 2023), exploration and mapping of seabed minerals, investigation of seabed ecological environment (Johnsen et al, 2013;Liu et al, 2020), marine archaeology (Ødegård et al, 2018), and classification of marine species such as sponges and corals (Song et al 2021b).…”