During the past decades, the composition and distribution of marine species have changed due to multiple anthropogenic pressures. Monitoring these changes in a cost-effective manner is of high relevance to assess the environmental status and evaluate the effectiveness of management measures. In particular, recent studies point to a rise of jellyfish populations on a global scale, negatively affecting diverse marine sectors like commercial fishing or the tourism industry. Past monitoring efforts using underwater video observations tended to be time-consuming and costly due to human-based data processing. In this paper, we present Jellytoring, a system to automatically detect and quantify different species of jellyfish based on a deep object detection neural network, allowing us to automatically record jellyfish presence during long periods of time. Jellytoring demonstrates outstanding performance on the jellyfish detection task, reaching an F1 score of 95.2%; and also on the jellyfish quantification task, as it correctly quantifies the number and class of jellyfish on a real-time processed video sequence up to a 93.8% of its duration. The results of this study are encouraging and provide the means towards a efficient way to monitor jellyfish, which can be used for the development of a jellyfish early-warning system, providing highly valuable information for marine biologists and contributing to the reduction of jellyfish impacts on humans.

Recent studies have shown evidence of a significant decline of the Posidonia oceanica (P.O.) meadows on a global scale. The monitoring and mapping of these meadows are fundamental tools for measuring their status. We present an approach based on a deep neural network to automatically perform a high-precision semantic segmentation of P.O. meadows in sea-floor images, offering several improvements over the state of the art techniques. Our network demonstrates outstanding performance over diverse test sets, reaching a precision of 96.57% and an accuracy of 96.81%, surpassing the reliability of labelling the images manually. Also, the network is implemented in an Autonomous Underwater Vehicle (AUV), performing an online P.O. segmentation, which will be used to generate real-time semantic coverage maps.Francisco Bonin-Font received the degree in telecommunications engineering from the Polytechnical University of Catalonia, Barcelona, in 1996, and the Ph.D. degree in computer engineering from the University of the Balearic Islands in 2012. He has been ten years with the industry of information technology services addressed to bank business, before he initiated his academic activities. He has participated as a Technician and a Researcher in nine projects

No abstract

During the past few decades, the need to intervene in underwater scenarios has grown due to the increasing necessity to perform tasks like underwater infrastructure inspection and maintenance or archaeology and geology exploration. In the last few years, the usage of Autonomous Underwater Vehicles (AUVs) has eased the workload and risks of such interventions. To automate these tasks, the AUVs have to gather the information of their surroundings, interpret it and make decisions based on it. The two main perception modalities used at close range are laser and video. In this paper, we propose the usage of a deep neural network to recognise pipes and valves in multiple underwater scenarios, using 3D RGB point cloud information provided by a stereo camera. We generate a diverse and rich dataset for the network training and testing, assessing the effect of a broad selection of hyperparameters and values. Results show F1-scores of up to 97.2% for a test set containing images with similar characteristics to the training set and up to 89.3% for a secondary test set containing images taken at different environments and with distinct characteristics from the training set. This work demonstrates the validity and robust training of the PointNet neural in underwater scenarios and its applicability for AUV intervention tasks.

The morphological analysis of dendritic spines is an important challenge for the neuroscientific community. Most state-of-the-art techniques rely on user-supervised algorithms to segment the spine surface, especially those designed for light microscopy images. Therefore, processing large dendritic branches is costly and time-consuming. Although deep learning (DL) models have become one of the most commonly used tools in image segmentation, they have not yet been successfully applied to this problem. In this article, we study the feasibility of using DL models to automatize spine segmentation from confocal microscopy images. Supervised learning is the most frequently used method for training DL models. This approach requires large data sets of high-quality segmented images (ground truth). As mentioned above, the segmentation of microscopy images is time-consuming and, therefore, in most cases, neuroanatomists only reconstruct relevant branches of the stack. Additionally, some parts of the dendritic shaft and spines are not segmented due to dyeing problems. In the context of this research, we tested the most successful architectures in the DL biomedical segmentation field. To build the ground truth, we used a large and high-quality data set, according to standards in the field. Nevertheless, this data set is not sufficient to train convolutional neural networks for accurate reconstructions. Therefore, we implemented an automatic preprocessing step and several training strategies to deal with the problems mentioned above. As shown by our results, our system produces a high-quality segmentation in most cases. Finally, we integrated several postprocessing user-supervised algorithms in a graphical user interface application to correct any possible artifacts.