Resilience of riverbed vegetation to uprooting by flow Perona, P.; Crouzy, B.
Proceedings of the Royal Society. A, Mathematical, physical, and engineering sciences,
03/2018, Letnik:
474, Številka:
2211
Journal Article
Recenzirano
Odprti dostop
Riverine ecosystem biodiversity is largely maintained by ecogeomorphic processes including vegetation renewal via uprooting and recovery times to flow disturbances. Plant roots thus heavily ...contribute to engineering resilience to perturbation of such ecosystems. We show that vegetation uprooting by flow occurs as a fatigue-like mechanism, which statistically requires a given exposure time to imposed riverbed flow erosion rates before the plant collapses. We formulate a physically based stochastic model for the actual plant rooting depth and the time-to-uprooting, which allows us to define plant resilience to uprooting for generic time-dependent flow erosion dynamics. This theory shows that plant resilience to uprooting depends on the time-to-uprooting and that root mechanical anchoring acts as a process memory stored within the plant–soil system. The model is validated against measured data of time-to-uprooting of Avena sativa seedlings with various root lengths under different flow conditions. This allows for assessing the natural variance of the uprooting-by-flow process and to compute the prediction entropy, which quantifies the relative importance of the deterministic and the random components affecting the process.
Humans move their eyes while looking at scenes and pictures. Eye movements correlate with shifts in attention and are thought to be a consequence of optimal resource allocation for high-level tasks ...such as visual recognition. Models of attention, such as "saliency maps," are often built on the assumption that "early" features (color, contrast, orientation, motion, and so forth) drive attention directly. We explore an alternative hypothesis: Observers attend to "interesting" objects. To test this hypothesis, we measure the eye position of human observers while they inspect photographs of common natural scenes. Our observers perform different tasks: artistic evaluation, analysis of content, and search. Immediately after each presentation, our observers are asked to name objects they saw. Weighted with recall frequency, these objects predict fixations in individual images better than early saliency, irrespective of task. Also, saliency combined with object positions predicts which objects are frequently named. This suggests that early saliency has only an indirect effect on attention, acting through recognized objects. Consequently, rather than treating attention as mere preprocessing step for object recognition, models of both need to be integrated.
One-shot learning of object categories Li Fei-Fei; Fergus, R.; Perona, P.
IEEE transactions on pattern analysis and machine intelligence,
04/2006, Letnik:
28, Številka:
4
Journal Article
Recenzirano
Odprti dostop
Learning visual models of object categories notoriously requires hundreds or thousands of training examples. We show that it is possible to learn much information about a category from just one, or a ...handful, of images. The key insight is that, rather than learning from scratch, one can take advantage of knowledge coming from previously learned categories, no matter how different these categories might be. We explore a Bayesian implementation of this idea. Object categories are represented by probabilistic models. Prior knowledge is represented as a probability density function on the parameters of these models. The posterior model for an object category is obtained by updating the prior in the light of one or more observations. We test a simple implementation of our algorithm on a database of 101 diverse object categories. We compare category models learned by an implementation of our Bayesian approach to models learned from by maximum likelihood (ML) and maximum a posteriori (MAP) methods. We find that on a database of more than 100 categories, the Bayesian approach produces informative models when the number of training examples is too small for other methods to operate successfully.
We demonstrated nanoscale MRI of metabolism in single isolated mitochondria and within living cells using NV centers in diamond.
Free radicals play a vital role in all kinds of biological processes ...including immune responses. However, free radicals have short lifetimes and are highly reactive, making them difficult to measure using current methods. Here, we demonstrate that relaxometry measurement, or T1, inherited from the field of diamond magnetometry can be used to detect free radicals in living cells with subcellular resolution. This quantum sensing technique is based on defects in diamond, which convert a magnetic signal into an optical signal, allowing nanoscale magnetic resonance measurements. We functionalized fluorescent nanodiamonds (FNDs) to target single mitochondria within macrophage cells to detect the metabolic activity. In addition, we performed measurements on single isolated mitochondria. We were able to detect free radicals generated by individual mitochondria in either living cells or isolated mitochondria after stimulation or inhibition.
We propose a novel approach to learn and recognize natural scene categories. Unlike previous work, it does not require experts to annotate the training set. We represent the image of a scene by a ...collection of local regions, denoted as codewords obtained by unsupervised learning. Each region is represented as part of a "theme". In previous work, such themes were learnt from hand-annotations of experts, while our method learns the theme distributions as well as the codewords distribution over the themes without supervision. We report satisfactory categorization performances on a large set of 13 categories of complex scenes.
Fluorescent nanodiamonds are promising probes for nanoscale magnetic resonance measurements. Their physical properties predict them to have particularly useful applications in intracellular analysis. ...Before using them in intracellular experiments however, it should be clear whether diamond particles influence cell biology. While cytotoxicity has already been ruled out in previous studies, we consider the non-fatal influence of fluorescent nanodiamonds on the formation of reactive oxygen species (an important stress indicator and potential target for intracellular sensing) for the first time. We investigated the influence of different sizes, shapes and concentrations of nanodiamonds on the genetic and protein level involved in oxidative stress-related pathways of the HeLa cell, an important model cell line in research. The changes in viability of the cells and the difference in intracellular levels of free radicals, after diamond uptake, are surprisingly small. At lower diamond concentrations, the cellular metabolism cannot be distinguished from that of untreated cells. This research supports the claims of non-toxicity and includes less obvious non-fatal responses. Finally, we give a handhold concerning the diamond concentration and size to use for non-toxic, intracellular measurements in favour of (cancer) research in HeLa cells.
The establishment of riparian pioneer vegetation is of crucial importance within river restoration projects. After germination or vegetative reproduction on river bars juvenile plants are often ...exposed to mortality by uprooting caused by floods. At later stages of root development vegetation uprooting by flow is seen to occur as a consequence of a marked erosion gradually exposing the root system and accordingly reducing the mechanical anchoring. How time scales of flow-induced uprooting do depend on vegetation stages growing in alluvial non-cohesive sediment is currently an open question that we conceptually address in this work. After reviewing vegetation root issues in relation to morphodynamic processes, we then propose two modelling mechanisms (Type I and Type II), respectively concerning the uprooting time scales of early germinated and of mature vegetation. Type I is a purely flow-induced drag mechanism, which causes alone a nearly instantaneous uprooting when exceeding root resistance. Type II arises as a combination of substantial sediment erosion exposing the root system and resulting in a decreased anchoring resistance, eventually degenerating into a Type I mechanism. We support our conceptual models with some preliminary experimental data and discuss the importance of better understanding such mechanisms in order to formulate sounding mathematical models that are suitable to plan and to manage river restoration projects.
We investigate a method for learning object categories in a weakly supervised manner. Given a set of images known to contain the target category from a similar viewpoint, learning is translation and ...scale-invariant; does not require alignment or correspondence between the training images, and is robust to clutter and occlusion. Category models are probabilistic constellations of parts, and their parameters are estimated by maximizing the likelihood of the training data. The appearance of the parts, as well as their mutual position, relative scale and probability of detection are explicitly described in the model. Recognition takes place in two stages. First, a feature-finder identifies promising locations for the model"s parts. Second, the category model is used to compare the likelihood that the observed features are generated by the category model, or are generated by background clutter. The flexible nature of the model is demonstrated by results over six diverse object categories including geometrically constrained categories (e.g. faces, cars) and flexible objects (such as animals).PUBLICATION ABSTRACT
Abstract Skin equivalents (SE) that recapitulate biological and mechanical characteristics of the native tissue are promising platforms for assessing cosmetics and studying fundamental biological ...processes. Methods to achieve SEs with well‐organized structure, and ideal biological and mechanical properties are limited. Here, the combination of melt electrowritten PCL scaffolds and cell‐laden Matrigel to fabricate SE is described. The PCL scaffold provides ideal structural and mechanical properties, preventing deformation of the model. The model consists of a top layer for seeding keratinocytes to mimic the epidermis, and a bottom layer of Matrigel‐based dermal compartment with fibroblasts. The compressive modulus and the biological properties after 3‐day coculture indicate a close resemblance with the native skin. Using the SE, a testing system to study the damage caused by UVA irradiation and evaluate antioxidant efficacy is established. The effectiveness of Tea polyphenols (TPs) and L‐ascorbic acid (Laa) is compared based on free radical generation. TPs are demonstrated to be more effective in downregulating free radical generation. Further, T1 relaxometry is used to detect the generation of free radicals at a single‐cell level, which allows tracking of the same cell before and after UVA treatment.
•We run four water stress experiments on a drip irrigation system in Burkina Faso.•Eggplant water stress was related to soil matric potential (SMP), biomass and yields.•Numerical simulations unravel ...SMP thresholds dependence on root and sensor depth.•Optimality for water and yield is obtained with evolving thresholds during growth.•SMP sensor location at 10cm depth reveals optimal independently of soil texture.
Field experiments were combined with a numerical model to optimize drip irrigation management based on soil matric potential (SMP) measurements. An experimental crop of eggplant was grown in Burkina Faso from December 2014 to March 2015 and plant response to water stress was investigated by applying four different irrigation treatments. Treatments consisted in using two different irrigation depths (low or high), combined with a water provision of 150%, 100% or 66% (150/100/66) of the maximum crop evapotranspiration (T150low, T66low, T100high, T66high). Soil matric potential measurements at 5, 10 and 15cm depth were taken using a wireless sensor network and were compared with measurements of plant and root biomass and crop yields. Field data were used to calibrate a numerical model to simulate triggered drip irrigation. Different simulations were built using the software HYDRUS 2D/3D to analyze the impact of the irrigation depth and frequency, the irrigation threshold and the soil texture on plant transpiration and water losses. Numerical results highlighted the great impact of the root distribution on the soil water dynamics and the importance of the sensor location to define thresholds. A fixed optimal sensor depth of 10 cm was found to manage irrigation from the vegetative state to the end of fruit development. Thresholds were defined to minimize water losses while allowing a sufficient soil water availability for optimal crop production. A threshold at 10cm depth of −15kPa is recommended for the early growth stage and −40kPa during the fruit formation and maturation phase. Simulations showed that those thresholds resulted in optimal transpiration regardless of the soil texture so that this management system can constitute the basis of an irrigation schedule for eggplant crops and possibly other vegetable crops in semi-arid regions.
PDF
