Suchergebnisse

L'H2020 712949 está vinculat amb el programa ACCIO (Agència per a la Competitivitat de la Empresa). Programa ACCIÓ per fomentar la mobilitat d'investigadors amb un enfocament en la recerca aplicada i la transferència de tecnologia ; Due to absorption and scattering effects the under-water scenes are characterized by poor contrast, color shifting, additional noise and hazy appearance. In this paper we introduce a novel solution that estimates locally the backscattered light. While in general the existing solutions estimate a global backscattered light value over the entire scene, our local strategy is able to deal effectively to the more challenging non-uniform illumination generated by multiple light-sources. Our solution computes two complementary estimates of the local backscattered light, covering a large and a small patch size. The optimal local backscattered light is computed as the mean of the outputs processed with the small and the large patches while the transmission map, is estimated based on the dark-channel prior (DCP) [1]. Finally, our restored results are computed by simply inverting the optical model using the transmission and the local backscattered light estimates. The qualitative evaluation demonstrates the effectiveness of our approach compared with the recent underwater enhancing techniques ; Part of this work has been funded by the Romanian Government UEFISCDI, project PN-III-P2-2.1-PED-2016-0940. Part of this work has been funded from the 2020 European Union Research and Innovation Horizon 2020 under the grant agreement Marie Sklodowska-Curie No 712949 (TECNIOspring PLUS), as well as the Agency for the Competitiveness of the Company of the Generalitat de Catalunya - ACCIO: TECSPR17-1-0054

13 pages, 9 figures, 2 tables.-- Data Availability Statement: The time series of specimen counts per species, obtained through the visual inspection of the image dataset, is provided as a supplementary material (only the images containing at least one specimen are reported). The image datasets analysed for this study can be accessed by contacting the OBSEA observatory [https://www.obsea.es/] on reasonable request ; The marine science community is engaged in the exploration and monitoring of biodiversity dynamics, with a special interest for understanding the ecosystem functioning and for tracking the growing anthropogenic impacts. The accurate monitoring of marine ecosystems requires the development of innovative and effective technological solutions to allow a remote and continuous collection of data. Cabled fixed observatories, equipped with camera systems and multiparametric sensors, allow for a non-invasive acquisition of valuable datasets, at a high-frequency rate and for periods extended in time. When large collections of visual data are acquired, the implementation of automated intelligent services is mandatory to automatically extract the relevant biological information from the gathered data. Nevertheless, the automated detection and classification of streamed visual data suffer from the "concept drift" phenomenon, consisting of a drop of performance over the time, mainly caused by the dynamic variation of the acquisition conditions. This work quantifies the degradation of the fish detection and classification performance on an image dataset acquired at the OBSEA cabled video-observatory over a one-year period and finally discusses the methodological solutions needed to implement an effective automated classification service operating in real time ; This research activity was partially funded by the "ENDURUNS - Development and demonstration of a long-endurance sea surveying autonomous unmanned vehicle with gliding capability powered by hydrogen fuel cell project", Horizon 2020, Grant Agreement H2020-MG-2018-2019-2020 n.824348 and by the "Joint European Research Infrastructure of Coastal Observatories: Science, Service, Sustainability - JERICO-S3'' project, Horizon 2020, Grant Agreement no. 871153. This research was also funded within the framework of the following project activities: ARIM (Autonomous Robotic sea-floor Infrastructure for benthopelagic Monitoring; MarTERA ERA-Net Cofound); RESBIO (TEC2017-87861-R; Ministerio de Ciencia, Innovación y Universidades). We also profited from the funding from the Spanish Government through the 'Severo Ochoa Centre of Excellence' accreditation (CEX2019-000928-S)

Jellyfish can form erratic blooms in response to seasonal and irregular changes in environmental conditions with often large, transient effects on local ecosystem structure as well as effects on several sectors of the marine and maritime economy. Early warning systems able to detect conditions for jelly fish proliferation can enable management responses to mitigate such effects providing benefit to local ecosystems and economies. We propose here the creation of a research team in response to the EU call for proposal under the European Maritime and Fisheries Fund called "Blue Labs: innovative solutions for maritime challenges". The project will establish a BLUELAB team with a strong cross-sectorial component that will benefit of the expertise of researchers in IT and Marine Biology, Computer Vision and embedded systems, which will work in collaboration with Industry and Policy maker to develop an early warning system using a new underwater imaging system based on Time of Flight Laser cameras. The camera will be combined to machine learning algorithm allowing autonomous early detection of jellyfish species (e.g. polyp, ephyra and planula stages). The team will develop the system and the companion software and will demonstrate its applications in real case conditions.

Background. The Life+ Indemares project (www.indemares.es) aimed to better understand the natural and socioeconomic values of several marine areas along the territorial waters of the Spanish State, leading to an informed decision-making process for the designation of new protected areas for the marine environment. One of the 10 areas selected corresponded to the "South-West Gulf of Lions Canyon System" (https://eunis.eea.europa.eu/sites/ESZZ16001), located in the north-eastern part of the Iberian Peninsula. It includes the submarine canyons of Cap de Creus and Lacaze-Duthiers and their adjacent continental shelf. The Marine Biodiversity, Ecology and Conservation Research Group from the Institute of Marine Sciences of Barcelona (ICM-CSIC) produced a detailed evaluation of the physical and ecological characteristics of the seabed and water column habitats, providing the necessary scientific information for its proposal as a Site of Community Importance to the European Union. Faunistic catalogue. An ecosystem-based approach to the management of Cap de Creus marine area after protection measures are put in place requires a comprehensive knowledge of its benthic ecosystem, which includes the continental shelf and submarine canyon. The methodology employed to identify and characterize its main benthic communities down to 400 m depth was based on images collected through underwater video platforms, such as Remotely Operated Vehicles (ROVs) and manned submersibles. The involvement of taxonomists allowed for the identification of most of the organisms observed in the images down to species or genus level, a task that could only be achieved due to the high-quality footage recorded and the set of biological samples collected. This detailed work led to the development of a photographic catalogue that served as basis for the set of ecological studies developed subsequently. Acknowledging that the identification of benthic species from imagery still has its caveats, we felt that this faunistic inventory should be made ...

Archaeological data are usually inherently incomplete, heterogeneous, discontinuous and require frequent updates and possible adjustments. It is important to constantly create detailed documentation, which will precisely represent the actual situation. However, even the most precise figure is only an estimated representation of the documented object. Therefore, it is necessary to collect fully metric documentation and its professional archaeological interpretation. Acquiring correct and valuable underwaterdigital images for the archaeology purposes is not easy due to specific shooting conditions. It should be noted a number of limitations are unique to this type of imaging environment — the apparent extension of the focal length, the "disappearance" of colours, as well as a significant reduction in the transparency of the water environment. Therefore, the authors have made attempts to describe changes, in a much broader sense, in the quality of photogrammetric images that had been taken in various shooting conditions. Underwater and aerial images of two test fields were tested. First, the ground sampling distance of the INTOVA IC500 digital camera and the geometric accuracy of the acquired images were examined. Then, the impact of changes to the imaging conditions on the radio-metric resolution and colour projection were designated. In the last stage, the acquired images were used in practice — to assess the progress of the erosion process of an archaeological object, and also to comply its documentation in the form of vector drawing with the accuracy of mxy = ±0.5 mm. Keywords: photogrammetry and remote sensing, archaeology, underwater photogrammetry, resolution, ground resolved distance

This thesis proposes new techniques as well as considers state-of-the-art and recently developed methods to analyse images acquired through different turbulent media, such as atmospheric and underwater turbulence. The techniques include some challenging image processing tasks, such as image restoration, object detection and identification. In particular, it deals with both space-variant and space-invariant blur removal and supress the distortions. In addition, two effective techniques are proposed for object detection in images acquired through long-range atmospheric turbulent path, and object identification in images acquired through underwater turbulent media. Possible applications of these techniques can include different forms, such as military surveillance, detecting and identifying objects, imaging problems in remote sensing, and various ocean applications.A faster image pre-processing technique by employing the k-means clustering technique is proposed in this thesis for selecting the important frames rather than considering random frames. This technique is used in imaging for both turbulent media. A recent restoration method is reviewed and improved by employing a spatio-temporal kernel regression based fusion method for removing space-invariant blur from the images and supress deformations. To further improve the image quality a blind deconvolution procedure is employed. In practice, the images acquired through turbulent media are distorted due to space-variant blur problem. In-order to address this problem a patch-wise deconvolution process is carried out on the distorted image, and the turbulence corrected image is given as input to the dictionary learning algorithm for further denoising. Motivated by the saliency technique, object detection technique is proposed considering the small cluttered objects present in the images acquired. These objects makes other detection problems difficult such as moving object detection. Thus, it is important to detect the small cluttered objects.This thesis also focuses on the imaging issues considering underwater turbulence media. Images are distorted mainly due to non-uniform illumination that occurs as a result of scattering and absorption of various submerged organic substances, and space-variant blur. The Retinex model is employed to correct the non-uniform illumination problem prior to space-variant blur removal. A patch-wise operation is then carried out for further restoration purpose. Image segmentation algorithm by considering the efficiency of suppression factor is then implemented on the restored image for identifying the objects. The potential and the performance of the proposed approaches are investigated considering both real-world and synthetic datasets, and shows better restoration, detection and identification results. The results show practical interest for considering the proposed approaches in different computer vision and image processing systems, and machine learning and pattern analysis applications.

Imaging through a turbulent medium, such as the atmosphere or the wavy surface of water, is highly desired in many scientific and military applications. This is a very challenging task due to the time-varying shifts and blurs captured in the images. This thesis deals with the geometrical restoration of such images captured as video sequences. These ordinarily undesirable geometrical distortions also act as information compressors and can be exploited to extract further bandwidth from the images to produce high-quality images from their lower resolution counterparts. The research investigations cover both the atmospheric as well as underwater imaging.First, a simple and robust method is reviewed and improved upon to restore warped frames using motion vector fields (shiftmaps) obtained through a motion estimation technique. The centroid of the pixel shiftmaps is then calculated to generate individual restoration shiftmaps for each warped frame. The centroid shiftmap is updated iteratively to take the restored frames closer to their likely ground-truth. Furthermore, the image restoration method is made predictive by the use of a generalized regression neural network (GRNN), where the pixel shiftmaps amongst successive frames are used for training the network to determine the underlying warping functions, which in turn, are used to predict the upcoming warped frame. Moreover, the accurate motionanalysis along with video stabilization method is utilized for reliable segmentation of video frames into stable and moving components and subsequently stabilizing frames, keeping real moving objects unaltered. Motivated by the successful application of GRNN in warp prediction, finally, a new and more efficient target tracking algorithm is proposed that works based on determining the centre and the area of moving objects, using those features for GRNN training, and employing the trained network to estimate the objects' locations in the next frame. Both the accuracy and the potential of the proposed algorithms have been investigated. The results presented are of both theoretical and practical interest and offer new efficient tools for substantial improvement of infrastructure of machine vision-based systems in general and of intelligent surveillance systems in particular.