Structure information is ubiquitous in natural scene images and it plays an important role in scene representation. In this paper, third order structure statistics (TOSS) and fourth order structure statistics (FOSS) are exploited to encode higher ord
er structure information. Afterwards, based on the radial and normal slice of TOSS and FOSS, we propose the high order structure feature: third order structure feature (TOSF) and fourth order structure feature (FOSF). It is well known that scene images are well characterized by particular arrangements of their local structures, we divide the scene image into the non-overlapping sub-regions and compute the proposed higher order structural features among them. Then a scene classification is performed by using SVM classifier with these higher order structure features. The experimental results show that higher order structure statistics can deliver image structure information well and its spatial envelope has strong discriminative ability.
In object recognition, Fisher vector (FV) representation is one of the state-of-art image representations ways at the expense of dense, high dimensional features and increased computation time. A simplification of FV is attractive, so we propose Spar
se Fisher vector (SFV). By incorporating locality strategy, we can accelerate the Fisher coding step in image categorization which is implemented from a collective of local descriptors. Combining with pooling step, we explore the relationship between coding step and pooling step to give a theoretical explanation about SFV. Experiments on benchmark datasets have shown that SFV leads to a speedup of several-fold of magnitude compares with FV, while maintaining the categorization performance. In addition, we demonstrate how SFV preserves the consistence in representation of similar local features.
We report on simulations and measurements of the optical absorption of silicon nanowires (NWs) versus their diameter. We first address the simulation of the optical absorption based on two different theoretical methods : the first one, based on the G
reen function formalism, is useful to calculate the scattering and absorption properties of a single or a finite set of NWs. The second one, based on the Finite Difference Time Domain (FDTD) method is well-adapted to deal with a periodic set of NWs. In both cases, an increase of the onset energy for the absorption is found with increasing diameter. Such effect is experimentally illustrated, when photoconductivity measurements are performed on single tapered Si nanowires connected between a set of several electrodes. An increase of the nanowire diameter reveals a spectral shift of the photocurrent intensity peak towards lower photon energies, that allows to tune the absorption onset from the ultraviolet radiations to the visible light spectrum.
