Grenoble Institute of Technology

About: Grenoble Institute of Technology is a education organization based out in Grenoble, France. It is known for research contribution in the topics: Hyperspectral imaging & Geology. The organization has 3427 authors who have published 5345 publications receiving 137158 citations. The organization is also known as: Grenoble INP.

Taking Optimal Advantage of Fine Spatial Resolution: Promoting partial image reconstruction for the morphological analysis of very-high-resolution images

Abstract: Diverse sensor technologies have allowed us to measure different aspects of objects on Earth's surface [such as spectral characteristics in hyperspectral images and height in light detection and ranging (LiDAR) data] with increasing spectral and spatial resolutions. Remote-sensing images of very high geometrical resolution can provide a precise and detailed representation of the monitored scene. Thus, the spatial information is fundamental for many applications. Morphological profiles (MPs) and attribute profiles (APs) have been widely used to model the spatial information of very-high-resolution (VHR) remote-sensing images. MPs are obtained by computing a sequence of morphological operators based on geodesic reconstruction. However, both morphological operators based on geodesic reconstruction and attribute filters (AFs) are connected filters and, hence, suffer the problem of leakage (i.e., regions related to different structures in the image that happen to be connected by spurious links are considered as a single object). Objects expected to disappear at a given stage remain present when they connect with other objects in the image. Consequently, the attributes of small objects are mixed with their larger connected objects, leading to poor performances on postapplications (e.g., classification).

Quantum hall effect in polycrystalline graphene: The role of grain boundaries

Abstract: We use numerical simulations to predict peculiar magnetotransport fingerprints in polycrystalline graphene, driven by the presence of grain boundaries of varying size and orientation. The formation of Landau levels is shown to be restricted by the polycrystalline morphology, requiring the magnetic length to be smaller than the average grain radius. The nature of localization is also found to be unusual, with strongly localized states at the center of Landau levels (including the usually highly robust zero-energy state) and extended electronic states lying between Landau levels. These extended states percolate along the network of grain boundaries, resulting in a finite value for the bulk dissipative conductivity and suppression of the quantized Hall conductance. Such breakdown of the quantum Hall regime provoked by extended structural defects is also illustrated through two-terminal Landauer-B\"uttiker conductance calculations, indicating how a single grain boundary induces cross linking between edge states lying at opposite sides of a ribbon geometry.

Mobility Enhancement by Back-Gate Biasing in Ultrathin SOI MOSFETs With Thin BOX

Abstract: Carrier mobility (μ) at various back-gate biases is studied for nand p-channel ultrathin (8 nm) SOI MOSFETs with thin (10 nm) buried oxide (BOX) and ground plane (GP). We found that μ did not deteriorate for either thin BOX or GP structure, even in the back channel (BC). We also found the largest μ enhancement effect in p-channel devices by the back-gate bias. As this enhancement effect could conceal the superior μ at the Si/SiO2 interface, μ was maximized when both the front channel and BC were conducting. By contrast, μ in n-channel devices was maximized only when the BC was activated. This large μ gain in p-channel devices is promising for further CMOS scaling.

Improving Random Forest With Ensemble of Features and Semisupervised Feature Extraction

Abstract: In this letter, we propose a novel approach for improving Random Forest (RF) in hyperspectral image classification. The proposed approach combines the ensemble of features and the semisupervised feature extraction (SSFE) technique. The main contribution of our approach is to construct an ensemble of RF classifiers. In this way, the feature space is divided into several disjoint feature subspaces. Then, the feature subspaces induced by the SSFE technique are used as the input space to an RF classifier. This method is compared with a regular RF and an RF with the reduced features by the SSFE on two real hyperspectral data sets, showing an improved performance in ill-posed, poor-posed, and well-posed conditions. An additional study shows that the proposed method is less sensitive to the parameters.

Self-protection for distributed component-based applications

Abstract: The complexity of today's distributed computing environments is such that the presence of bugs and security holes is statistically unavoidable. A very promising approach to this issue is to implement a self-protected system, similarly to a natural immune system which has the ability to detect the intrusion of foreign elements and react while it is still in progress. This paper describes an approach relying on component-based software engineering to ease the protection of distributed systems. The knowledge of the application architecture is used to detect foreign activities and to trigger counter measures. We focus on a mean to recognize known and unknown attacks independently from legacy software and avoiding false positives. Hence, the scope of the detected attacks is, for the moment, limited to the detection of illegal communications. We describe how this approach can be applied to provide self-protection for clustered J2ee applications with a very low overhead.