Biologically-inspired object recognition system with features from complex wavelets

In this paper, a novel cortex-inspired feed-forward hierarchical object recognition system based on complex wavelets is proposed and tested. Complex wavelets contain three key properties for object representation: shift invariance, which enables the extraction of stable local features; good directional selectivity, which simplifies the determination of image orientations; and limited redundancy, which allows for efficient signal analysis using the multi-resolution decomposition offered by complex wavelets. In this paper, we propose a complete cortex-inspired object recognition system based on complex wavelets. We find that the implementation of the HMAX model for object recognition in [1, 2] is rather over-complete and includes too much redundant information and processing. We have optimized the structure of the model to make it more efficient. Specifically, we have used the Caltech 5 standard dataset to compare with Serre's model in [2] (which employs Gabor filter bands). Results demonstrate that the complex wavelet model achieves a speed improvement of about 4 times over the Serre model and gives comparable recognition performance. © 2011 IEEE.

Growth window and possible mechanism of millimeter-thick single-walled carbon nanotube forests.

Our group recently reproduced the water-assisted growth method, so-called "SuperGrowth", of millimeter-thick single-walled carbon nanotube (SWNT) forests by using C2H4/H2/H2O/Ar reactant gas and Fe/Al2O3, catalyst. In this current work, a parametric study was carried out on both reaction and catalyst conditions. Results revealed that a thin Fe catalyst layer (about 0.5 nm) yielded rapid growth of SWNTs only when supported on Al2O3, and that Al2O3 support enhanced the activity of Fe, Co, and Ni catalysts. The growth window for the rapid SWNT growth was narrow, however. Optimum amount of added H2O increased the SWNT growth rate but further addition of H2O degraded both the SWNT growth rate and quality. Addition of H2 was also essential for rapid SWNT growth, but again, further addition decreased both the SWNT growth rate and quality. Because Al2O3 catalyzes hydrocarbon reforming, Al2O3 support possibly enhances the SWNT growth rate by supplying the carbon source to the catalyst nanoparticles. The origin of the narrow window for rapid SWNT growth is also discussed.

Hydrothermally-grown ZnO nanowire tips for scanning tunnelling microscopy.

The probe tip is pivotal in determining the resolution and nature of features observed in the Scanning Tunnelling Microscope (STM). We have augmented a conventional Pt/Ir metallic tip with a hydrothermally grown ZnO nanowire (NW). Atomic resolution imaging of graphite is attained. Current-voltage (IV) characteristics demonstrate an asymmetry stemming from the unintentional n-type doping of the ZnO NW, whereas the expected Schottky barrier at the ZnO-Pt/Ir interface is shown to have negligible effect. Moreover the photoconductivity of the system is investigated, paving the way towards a photodetector capable of atomic resolution.

Co-occurrence flow for pedestrian detection

The last few years have seen considerable progress in pedestrian detection. Recent work has established a combination of oriented gradients and optic flow as effective features although the detection rates are still unsatisfactory for practical use. This paper introduces a new type of motion feature, the co-occurrence flow (CoF). The advance is to capture relative movements of different parts of the entire body, unlike existing motion features which extract internal motion in a local fashion. Through evaluations on the TUD-Brussels pedestrian dataset, we show that our motion feature based on co-occurrence flow contributes to boost the performance of existing methods. © 2011 IEEE.