Search

To study how axon growth is affected by the local environment biologists perform extensive experiments, watching the axons develop on different substrates. As axons grow from the neuron cell body they form tree-like structures, with branches forming and withering as they explore their surroundings. In this project, we designed a system which can track individual axons as they grow and branch over time, enabling quantitative evaluation of different aspects of the axon behaviour. The system includes a novel segmentation network that can be aware of not only intensity but also properties such as feature direction and scale. Such properties can be important when analysing images containing curvilinear structures such as vessels or fibres. We propose the General Multi-Angle Scale Convolution (G-MASC) network, whose kernels are arbitrarily rotatable and also fully differentiable. The model man- ages its directional detectors in sets, and supervises a set's rotational symmetry with a novel rotation penalty called PoRE. The algorithm works on pyramid representations to enable scale search. Direction and scale can be extracted from the output maps, encoded and analysed separately. Tests were conducted on three public datasets and the axon samples. Good performance is observed while the model requires 1% or fewer parameters compared to other approaches such as U-Net.

The Tibetan Plateau experienced tremendous climate change during the past four decades. Due to the large size, widely distribution of cryosphere, and diverse landforms, different parts of the plateau may experience different climate and cryosphere changing patterns. The changes of inland lakes within the plateau are important indicators of climate change as these lakes are fed by precipitation, permafrost degradation, and glacier melting that are all sensitive to climate change. To examine the spatial and temporal differences of lake variations across the Tibetan Plateau, Landsat images and ICESat/GLAS altimetry data were used to extract the changes in surface areas of 26 lakes selected from six different sub-regions during the 1970s-2010 and the changes in lake elevations of these lakes during 2003-2009. An automated model to extract lake surface area and elevation from Landsat and ICESat data is developed to improve the efficiency of processing the large amount of satellite data. By applying this model, the spatial and temporal changing patterns of selected 26 inland lakes across the Tibetan Plateau during the past four decades are revealed. The lakes from different parts of the Tibetan Plateau show different changing patterns. The lake expansion firstly started from the Central Tibetan Plateau in the 1980s, then moving northward and northwestward; the Northeastern and Northwestern Tibetan Plateau experienced obvious expansion after the late 1990s, and this expansion is still continuing in the northern part, whereas the rapid lake expansion either slowed down or stopped in the central and southern parts of the plateau. The differences in lake changing pattern are caused by diverse climatic regimes and the pattern of the cryospheric distribution in the Tibetan Plateau. For the southern part of the plateau, the change in precipitation and evaporation seems to be the dominating factor to control the lake changes; however, the cryospheric change caused by temperature increase is the most important factor influencing the lake fluctuations in the northern part. These patterns can provide insight into the mechanism of lakes dynamics in response to climate and cryospheric changes; and be applied to assess the potential impacts of climate change on water resources in the Tibetan Plateau.