Ravi S. Hegde

Broadband Optical Response in Ternary Tree Fractal Plasmonic Nanoantenna

The ability to precisely tailor lineshapes, operational bandwidth, and localized electromagnetic field enhancements (“hot spots”) in nanostructures is currently of interest in advancing the performance of plasmonics-based chemical and biological sensing techniques. Fractal geometries are an intriguing alternative in the design of plasmonic nanostructures as they offer tunable multiband response spanning the visible and infrared spectral regions. A numerical study of the optical behavior of ternary tree fractal plasmonic nanoantenna is presented. Self-similar features are seen to emerge in the extinction spectra with the increase in fractal order N of the tree structure. Plasmon oscillations occurring at different length scales are shown to correspond to the multiple peaks and are compared with the spatial maps of electric field enhancement at the surface of the nanoantenna. The multiple peaks are shown to be independently tunable by structural variation. The robustness of the spectral response and polarization dependence arising due to various asymmetries is discussed.

Multilayered Aluminum Plasmonic Metasurfaces for Ultraviolet Bandpass Filtering

A multilayered aluminum-based metasurface is proposed toward realization of compact bandpass transmission filters for the ultraviolet spectral region. A systematic numerical study is presented that can assist in precisely tailoring the filter characteristics. The filter performance characteristics are benchmarked against previously reported filters. The proposed structures are CMOS process compatible and are suitable as replacements of bulky Woods filters; consequently, they are a promising step toward cost-effective UV photodetectors and multispectral imagers.

Highly Sensitive Refractive Index Sensors with Plasmonic Nanoantennas−Utilization of Optimal Spectral Detuning of Fano Resonances

We analyze and optimize the performance of coupled plasmonic nanoantennas for refractive index sensing. The investigated structure supports a sub- and super-radiant mode that originates from the weak coupling of a dipolar and quadrupolar mode, resulting in a Fano-type spectral line shape. In our study, we vary the near-field coupling of the two modes and particularly examine the influence of the spectral detuning between them on the sensing performance. Surprisingly, the case of matched resonance frequencies does not provide the best sensor. Instead, we find that the right amount of coupling strength and spectral detuning allows for achieving the ideal combination of narrow line width and sufficient excitation strength of the subradiant mode, and therefore results in optimized sensor performance. Our findings are confirmed by experimental results and first-order perturbation theory. The latter is based on the resonant state expansion and provides direct access to resonance frequency shifts and line width changes as well as the excitation strength of the modes. Based on these parameters, we define a figure of merit that can be easily calculated for different sensing geometries and agrees well with the numerical and experimental results.

Dataset augmentation with synthetic images improves semantic segmentation

Although Deep Convolutional Neural Networks trained with strong pixel-level annotations have significantly pushed the performance in semantic segmentation, annotation efforts required for the creation of training data remains a roadblock for further improvements. We show that augmentation of the weakly annotated training dataset with synthetic images minimizes both the annotation efforts and also the cost of capturing images with sufficient variety. Evaluation on the PASCAL 2012 validation dataset shows an increase in mean IOU from 52.80% to 55.47% by adding just 100 synthetic images per object class. Our approach is thus a promising solution to the problems of annotation and dataset collection.

Transfer learning by finetuning pretrained CNNs entirely with synthetic images

We show that finetuning pretrained CNNs entirely on synthetic images is an effective strategy to achieve transfer learning. We apply this strategy for detecting packaged food products clustered in refrigerator scenes. A CNN pretrained on the COCO dataset and fine-tuned with our 4000 synthetic images achieves mean average precision (mAP @ 0.5-IOU) of 52.59 on a test set of real images (150 distinct products as objects of interest and 25 distractor objects) in comparison to a value of 24.15 achieved without such finetuning. The synthetic images were rendered with freely available 3D models with variations in parameters like color, texture and viewpoint without a high emphasis on photorealism. We analyze factors like training data set size, cue variances, 3D model dictionary size and network architecture for their influence on the transfer learning performance. Additionally, training strategies like fine-tuning with selected layers and early stopping which affect transfer learning from synthetic scenes to real scenes were explored. This approach is promising in scenarios where limited training data is available.

Rapid Design of Wide-Area Heterogeneous Electromagnetic Metasurfaces Beyond the Unit-Cell Approximation

We propose a novel numerical approach for the optimal design of wide-area heterogeneous electromagnetic metasurfaces beyond the conventionally used unit-cell approximation. The proposed method exploits the combination of Rigorous Coupled Wave Analysis (RCWA) and global optimization techniques (two evolutionary algorithms namely the Genetic Algorithm (GA) and a modified form of the Artificial Bee Colony (ABC with memetic search phase method) are considered). As a specific example, we consider the design of beam deflectors using all-dielectric nanoantennae for operation in the visible wavelength region; beam deflectors can serve as building blocks for other more complicated devices like metalenses. Compared to previous reports using local optimization approaches our approach improves device efficiency; transmission efficiency is especially improved for wide deflection angle beam deflectors. The ABC method with memetic search phase is also an improvement over the more commonly used GA as it reaches similar efficiency levels with upto 35% reduction in computation time. The method described here is of interest for the rapid design of a wide variety of electromagnetic metasurfaces irrespective of their operational wavelength.

Fractal Plasmonic Nanoantennae

The ability to precisely tailor lineshapes, operational bandwidth and localized electromagnetic field enhancements (“hot spots”) in nanostructures is currently of interest in advancing the performance of plasmonics based chemical and biological sensing techniques as well as in plasmonics based energy harvesting applications. Fractal geometries are an intriguing alternative in the design of plasmonic nanostructures as they offer tunable multi-band response spanning the visible and infrared spectral regions. This chapter reviews the recent developments concerning the incorporation of fractal geometries into plasmonic nanostructures. The scope is restricted to the review of fractal shaped antenna elements as opposed to fractal based array placement methods. Beginning with a brief overview of fractals and fractal based radio-frequency antenna engineering, the review focuses on two canonical geometries: the Sierpinski carpet and the fractal tree. Fractal geometries are promising for improving the performance of plasmonics based optical applications like ultrasensing and energy harvesting.

Bilayered nanoantenna design improves the performance of silicon metasurfaces in the visible-wavelength region

Abstract. High-transmissivity all-dielectric metasurfaces have recently attracted attention toward the realization of ultracompact optical devices and systems. Silicon-based metasurfaces, in particular, are highly promising considering the possibility of monolithic integration with complementary metal–oxide–semiconductor very large scale integration circuits. Realization of silicon-based metasurfaces operational in the visible wavelengths, however, remains a challenge. A numerical study of bilayered truncated-cone shaped nanoantenna elements is presented. Metasurfaces based on the proposed stepped conical geometry can be designed for operation in the 700- to 800-nm wavelength window and can achieve full-cycle phase response (0 to 2π) with an improved transmittance in comparison with the previously reported cylindrical geometry. A systematic parameter study of the influence of various geometrical parameters on the achievable amplitude and phase coverage is reported.

Fractal tree plasmonic nanoantenna array for multispectral surface-enhanced spectroscopy

Plasmonic nanoantenna are pushing the sensitivity and detection limits of chemical and biochemical assay technologies. The challenge is to precisely tailor the spectral lineshapes and operational bandwidth and to engineer the localized electromagnetic field enhancements (“hot spots”) optimally for a particular spectroscopic task. Here we present a numerical study of the optical behavior of ternary tree fractal plasmonic gold nanoantenna. Spectral features spanning the visible to mid IR range are seen with the observation of characteristic self-similar features as the fractal order increases. The geometrical regularity of the ternary tree is perturbed and its influence on the optical properties is studied. The nanoantenna is seen to exhibit reduced polarization sensitivity even in highly assymetric structures. The intriguing optical properties of ternary tree fractal nanoantenna are particularly promising for the optimal design of substrates for performing multispectral surface-enhanced spectroscopy.