Description: Megafauna play an important role in benthic ecosystem function and are sensitive indicators of environmental change. Non-invasive monitoring of benthic communities can be accomplished by seafloor imaging. However, manual quantification of megafauna in images is labor-intensive and therefore, this organism size class is often neglected in ecosystem studies. Automated image analysis has been proposed as a possible approach to such analysis, but the heterogeneity of megafaunal communities poses a non-trivial challenge for such automated techniques. Here, the potential of a generalized object detection architecture, referred to as iSIS (intelligent Screening of underwater Image Sequences), for the quantification of a heterogenous group of megafauna taxa is investigated. The iSIS system is tuned for a particular image sequence (i.e. a transect) using a small subset of the images, in which megafauna taxa positions were previously marked by an expert. To investigate the potential of iSIS and compare its results with those obtained from human experts, a group of eight different taxa from one camera transect of seafloor images taken at the Arctic deep-sea observatory HAUSGARTEN is used. The results show that inter-and intra-observer agreements of human experts exhibit considerable variation between the species, with a similar degree of variation apparent in the automatically derived results obtained by iSIS. Whilst some taxa (e. g. Bathycrinus stalks, Kolga hyalina, small white sea anemone) were well detected by iSIS (i.e. overall Sensitivity: 87%, overall Positive Predictive Value: 67%), some taxa such as the small sea cucumber Elpidia heckeri remain challenging, for both human observers and iSIS.

Description: Cold-water coral (CWC) reefs are heterogeneous ecosystems comprising numerous microhabitats. A typical European CWC reef provides various biogenic microhabitats (within, on and surrounding colonies of coral species such as Lophelia pertusa, Paragorgia arborea and Primnoa resedaeformis, or formed by their remains after death). These microhabitats may be surrounded and intermixed with non-biogenic microhabitats (soft sediment, hard ground, gravel/pebbles, steep walls). To date, studies of distribution of sessile fauna across CWC reefs have been more numerous than those investigating mobile fauna distribution. In this study we quantified shrimp densities associated with key CWC microhabitat categories at the Røst Reef, Norway, by analysing image data collected by towed video sled in June 2007. We also investigated shrimp distribution patterns on the local scale (〈40 cm) and how these may vary with microhabitat. Shrimp abundances at the Røst Reef were on average an order of magnitude greater in biogenic reef microhabitats than in non-biogenic microhabitats. Greatest shrimp densities were observed in association with live Paragorgia arborea microhabitat (43 shrimp m−2, SD = 35.5), live Primnoa resedaeformis microhabitat (41.6 shrimp m−2, SD = 26.1) and live Lophelia pertusa microhabitat (24.4 shrimp m−2, SD = 18.6). In non-biogenic microhabitat, shrimp densities were 〈2 shrimp m−2. CWC reef microhabitats appear to support greater shrimp densities than the surrounding non-biogenic microhabitats at the Røst Reef, at least at the time of survey.

Description: The increasing scientific and economic interest in the visual exploration and monitoring of marine areas is creating huge amounts of new underwater image and video data and approaches to computationally assisted analysis are desperately needed. In this paper we propose an image patch feature representation concept, the Bag of Prototypes (BoP), to cope with the individual problems in underwater image analysis. We consider the case of seafloor classification, which is relevant in many contexts such as habitat mapping or the exploration of mineral resources and show, that the BoP concept allows an efficient and accurate tile-wise estimation of poly-metallic nodule coverage in relation to two differently acquired gold standards.

Description: Detecting objects in underwater image sequences and video frames automatically, requires the application of selected algorithms in consecutive steps. Most of these algorithms are controlled by a set of parameters, which need to be calibrated for an optimal detection result. Those parameters determine the effectivity and efficiency of an algorithm and their impact is usually well known. There are however further non-algorithmic impact factors (or hidden parameters), which bias the training of a machine learning system as well as the subsequent detection process and thus need to be well understood and taken into account. In the context of megafauna detection in benthic images, we investigate the effects of some of these parameters on our machine learning based detection system iSIS. The images to be analyzed were taken at the deep-sea, long-term observatory HAUSGARTEN in which five experts labeled seven distinct object classes as an annotation gold standard. We found, that the hidden parameters from imaging as well as the fusion of expert knowledge could partly be compensated and were able to achieve detection performances of 0.67 precision and 0.87 recall. Despite the efforts to compensate the hidden parameters, the detection performance was still varying across the image transect. This poses the potential occurrence of further hidden parameters not taken into account so far.

Description: The analysis of architectural features in neural tissue sections and the identification of distinct regions is challenging for computer aided diagnosis (CAD) in neuropathology. Due to the difficulty of locating a tissue’s origin and alignment as well as the vast variety of structures within such images an orientation independent (i. e. rotation invariant) approach for tissue region segmentation has to be found to encode the structural features of neural layer architecture in the tissue. We propose to apply the Ripley’s-L function, originating from the field of plant ecol- ogy, to compute feature vectors encoding the spatial statistics of point patterns described by selectively stained cells. Combining the Ripley’s- L features with unsupervised clustering enables a segmentation of tissue sections into neuropathological areas.

Description: Far-sighted marine research institutions around the globe are capturing images from the seafloor at a scale of hundreds of thousands. Only a small part of these data have been accessed to date, as manual analyses are time-consuming and automated evaluation approaches are still under development. Machine learning and neural networks have been identified as a promising algorithmic approach to automate analysis of images from the seafloor. These algorithms need ground-truth data about the objects to be detected. As the information provided by one human expert lacks reproducibility, the expertise of a group of individuals has to be employed to collect training data as well as to evaluate the performance of an automated detection. In this paper we show that the inter-and intra-observer agreements of these human experts is a critical factor for the training of a learning architecture and has shown to be conditional to image quality for some object classes. A supervised automated detection approach is evaluated where five experts marked the positions of eight distinct object classes within seventy images taken at the HAUSGARTEN observatory (eastern Fram Strait, Arctic). Support Vector Machines were trained to detect and classify objects in the images with an overall sensitivity of 0.87 and precision of 0.67. A detailed comparison of the human expert agreements showed interesting correlations with the system's performance and pointed us towards new strategies for (semi-) automated underwater image analysis.

Description: Detecting objects in underwater image sequences and video frames automatically requires the application of selected algorithms in consecutive steps. Most of these algorithms are controlled by a set of parameters, which need to be calibrated for an optimal detection result. Those parameters determine the effectivity and efficiency of an algorithm and their impact is usually well known. There are however further non-algorithmic impact factors (or hidden parameters), which bias the training of a machine learning system as well as the subsequent detection process and thus need to be well understood and taken into account. In benthic imaging, one dominant, hidden parameter is the distance of the image acquisition device above the seafloor. Variations in the distance lead to variations in the benthic area size being captured, the relative size and position of an object within an image, the effect of the artificial light source and thus the recorded color spectrum. Image processing techniques that allow modeling the induced variations can be used to compensate for those effects and thus allow the exploration of initially biased data. Those processing techniques again require algorithmic parameters, which are influenced by the hidden parameters contained within the initial data. In supervised machine-learning architectures, further challenges arise from the inclusion of human expert knowledge used for the training of the learning algorithm. Utilizing the knowledge of only one expert can conceal the information needed for the generalization capability of an automated semantic image annotation system. Utilizing the knowledge of several experts requires explicit instruction of the participants to be able to produce comparable results. The fusion of individual expert knowledge poses further hidden parameters that impact the supervised learning architecture. Those could be an individual object specific expertise or the tendency to annotate with more or less self-criticism, which together can be expressed as the expert’s trustworthiness. In the context of megafauna detection in benthic images, we investigate the effects of some of these parameters on our machine learning based detection system iSIS [1] that consists of four succeeding steps: Imaging, expert annotation, training, and detection (see Figure 1). The images to be analyzed were taken at the deep-sea, long-term observatory HAUSGARTEN and five experts created an annotation gold standard. We found, that the hidden parameters from imaging as well as the fusion of expert knowledge could partly be compensated and were able to achieve detection performances of 67% precision and 87% recall. Despite the efforts to compensate the hidden parameters, the detection performance was still varying across the image transect. This poses the potential occurrence of further hidden parameters not taken into account so far. Here, we correlate the distance of the acquisition device with the image‐wise detection results (see Figure 2 A). Also, we show conformity of the automated detection results to the outcome of the manual detection consensus of human experts (see Figure 2 B). Finally, we show the impact of hidden parameters on subsequent steps by means of the effect of image illumination on the human expert annotation.

Description: In the field of underwater imaging it is often necessary to pre-process images to homogenize image quality across a whole image transect and to compensate variations in imaging conditions. A variety of pre-processing methods have been developed in the recent years to overcome different problems occurring in underwater imaging, hence performing different on different image sets. Protocols for an objective comparison and scoring of those methods are needed. Here we show how to use cluster indices to rank the different methods regarding their per- formances on different image sets. Our results show different ranking for four pre-processing methods for two chosen sets of benthic images from the deep seafloor.