Learning features from georeferenced seafloor imagery with location guided autoencoders

Although modern machine learning has the potential to greatly speed up the interpretation of imagery, the varied nature of the seabed and limited availability of expert annotations form barriers to its widespread use in seafloor mapping applications. This motivates research into unsupervised methods that function without large databases of human annotations. This paper develops an unsupervised feature learning method for georeferenced seafloor visual imagery that considers patterns both within the footprint of a single image frame and broader scale spatial characteristics. Features within images are learnt using an autoencoder developed based on the AlexNet deep convolutional neural network. Features larger than each image frame are learnt using a novel loss function that regularises autoencoder training using the Kullback–Leibler divergence function to loosely assume that images captured within a close distance of each other look more similar than those that are far away. The method is used to semantically interpret images taken by an autonomous underwater vehicle at the Southern Hydrates Ridge, an active gas hydrate field and site of a seafloor cabled observatory at a depth of 780 m. The method's performance when applied to clustering and content‐based image retrieval is assessed against a ground truth consisting of more than 18,000 human annotations. The study shows that the location based loss function increases the rate of information retrieval by a factor of two for seafloor mapping applications. The effects of physics‐based colour correction and image rescaling are also investigated, showing that the improved consistency of spatial information achieved by rescaling is beneficial for recognising artificial objects such as cables and infrastructures, but is less effective for natural objects that have greater dimensional variability.     

A Master Curve for the Shear Degradation of Lubricating Greases with a Fibrous Structure

In this article, the mechanical aging behavior of lubricating greases with a fibrous structure is studied by aging fresh samples at controlled shear rates and temperatures in an in-house-made Couette aging machine. The rheological properties of fresh and aged samples were evaluated in a plate–plate rheometer. In the absence of oxygen, no chemical reactions occurred. The results showed that shear degradation is accelerated by increasing the temperature. This thermal effect can be described by an Arrhenius law. A grease aging master curve was created to describe the influence of shear and temperature on the mechanical aging of fibrous structured grease. Next, the model was validated using a conventional grease worker test. Finally, the model was applied to full ball bearings using an R0F?+?bearing test rig. The master curve forms an important contribution to the development of grease life models.     

The role of the substrate on the mechanical and thermal stability of Pd thin films during hydrogen (de)sorption

In this work, we studied the mechanical and thermal stability of ~100 nm Pd thin films magnetron sputter deposited on a bare oxidized Si(100) wafer, a sputtered Titanium (Ti) intermediate layer, and a spin-coated Polyimide (PI) intermediate layer. The dependence of the film stability on the film morphology and the film-substrate interaction was investigated. It was shown that a columnar morphology with elongated voids at part of the grain boundaries is resistant to embrittlement induced by the hydride formation (α?β phase transitions). For compact film morphology, depending on the rigidity of the intermediate layer and the adherence to the substrate, complete transformation (Pd-PI-SiO₂/Si) or partly suppression (Pd-Ti-SiO₂/Si) of the α to β-phase was observed. In the case of Pd without intermediate layer (Pd-SiO₂/Si), buckling delamination occurred. The damage and deformation mechanisms could be understood by the analysis of the stresses and dislocation (defects) behavior near grain boundaries and the film-substrate interface. From diffraction line-broadening combined with microscopy analysis, we showed that in Pd thin films, stresses relax at critical stress values via different relaxation pathways depending on film-microstructure and film-substrate interaction. On the basis of the in-situ hydriding experiments, it was concluded that a Pd film on a flexible PI intermediate layer exhibits free-standing film-like behavior besides being strongly clamped on a stiff SiO₂/Si substrate.     

Integrating water and agricultural management: Collaborative governance for a complex policy problem

This paper examines governance requirements for integrating water and agricultural management (IWAM). The institutional arrangements for the agriculture and water sectors are complex and multi-dimensional, and integration cannot therefore be achieved through a simplistic ‘additive’ policy process. Effective integration requires the development of a new collaborative approach to governance that is designed to cope with scale dependencies and interactions, uncertainty and contested knowledge, and interdependency among diverse and unequal interests. When combined with interdisciplinary research, collaborative governance provides a viable normative model because of its emphasis on reciprocity, relationships, learning and creativity. Ultimately, such an approach could lead to the sorts of system adaptations and transformations that are required for IWAM.     

MRI-determined carotid artery flow velocities and wall shear stress in a mouse model of vulnerable and stable atherosclerotic plaque

Objectives  

We report here on the pre-clinical MRI characterization of an apoE−/− mouse model of stable and vulnerable carotid artery atherosclerotic plaques, which were induced by a tapered restriction (cast) around the artery. Specific focus was on the quantification of the wall shear stress, which is considered a key player in the development of the plaque phenotype.     

Tobacco imagery on New Zealand television 2002-2004

Considerable emphasis has been placed on the importance of tobacco imagery in the movies as one of the "drivers" of smoking among young people. Findings are presented from a content analysis of 98 hours of prime-time programming on New Zealand television 2004, identifying 152 scenes with tobacco imagery, and selected characteristics of those scenes. About one in four programmes contained tobacco imagery, most of which might be regarded as "neutral or positive". This amounted to about two scenes containing such imagery for every hour of programming. A comparison with our earlier content analysis of programming in 2002 indicated little change in the level of tobacco imagery. The effect of this imagery in contributing to young viewers taking up smoking, and sustaining the addiction among those already smoking, deserves more research attention.