BILATERAL EDGE DETECTORS

We propose to employ bilateral filters to solve the problem of edge detection. The proposed methodology presents an efficient and noise robust method for detecting edges. Classical bilateral filters smooth images without distorting edges. In this paper, we modify the bilateral filter to perform edge detection, which is the opposite of bilateral smoothing. The Gaussian domain kernel of the bilateral filter is replaced with an edge detection mask, and Gaussian range kernel is replaced with an inverted Gaussian kernel. The modified range kernel serves to emphasize dissimilar regions. The resulting approach effectively adapts the detection mask according as the pixel intensity differences. The results of the proposed algorithm are compared with those of standard edge detection masks. Comparisons of the bilateral edge detector with Canny edge detection algorithm, both after non-maximal suppression, are also provided. The results of our technique are observed to be better and noise-robust than those offered by methods employing masks alone, and are also comparable to the results from Canny edge detector, outperforming it in certain cases.

Fabrication of back contacts using laser writer and photolithography for inscribing textured solar cells

Semiconductor fabrication process begins with photolithography. Preparing a photo mask is the key process step in photolithography. The photo mask was fabricated by inscribing patterns directly onto a soda lime glass with the help of a laser beam, as it is easily controllable. Laser writer LW405-A was used for preparing the mask in this study. Exposure wavelength of 405 nm was used, with which 1.2 mu m feature size can be written in direct write-mode over the soda lime glass plate. The advantage of using the fabricated mask is that it can be used to design back contacts for thin film Photovoltaic (PV) solar cells. To investigate the process capability of LW405-A, same pattern with different line widths was written on soda lime glass samples at different writing speeds. The pattern was inscribed without proximity effect and stitching errors, which was characterized using optical microscope and field emission scanning electron microscope (FE-SEM). It was proven that writing speed of a mask-writer is decided according to the intended feature size and line width. As the writing speed increases, the edges of the patterns become rougher due to uneven scattering of the laser beam. From the fabricated mask, the solar cell can be developed embedding both the contacts at the bottom layer, to increase the absorption of solar radiation on the top surface effectively by increasing light absorption area.

Electrolithography- A New and Versatile Process for Nano Patterning

We report a new lithography technique based on electromigration driven material transport for drawing patterns at nanometer scales in ambient conditions. We use a thin metal film as a masking layer and a polymer layer beneath it as a pattern transfer layer. The desired pattern is drawn in the metal layer by etching the metal with a conducting scanning probe assisted by liquid electromigration. The pattern drawn on the metal layer is transferred to the polymer layer by etching the polymer with an appropriate solvent. Subsequently, the pattern is transferred to the desired material layer using a film deposition technique followed by conventional lift-off process. Using this simple technique, we have achieved pattern resolutions of 9 nm on the polymer and 40 nm on transferring the pattern to another material. Based on the ease of use and process costs, this technique promises to be competitive to e-beam lithography that employs high energy and ultra-high vacuum, or the industrial standard ultra-violet light photolithography that employs extremely expensive implements to reach nano-scale resolutions. We also demonstrate direct mask writing using this technique and explain the fundamentals behind the workings of the developed method.

SPEECH ENHANCEMENT USING DISCRIMINATIVE RANDOM FIELDS

Speech enhancement in stationary noise is addressed using the ideal channel selection framework. In order to estimate the binary mask, we propose to classify each time-frequency (T-F) bin of the noisy signal as speech or noise using Discriminative Random Fields (DRF). The DRF function contains two terms - an enhancement function and a smoothing term. On each T-F bin, we propose to use an enhancement function based on likelihood ratio test for speech presence, while Ising model is used as smoothing function for spectro-temporal continuity in the estimated binary mask. The effect of the smoothing function over successive iterations is found to reduce musical noise as opposed to using only enhancement function. The binary mask is inferred from the noisy signal using Iterated Conditional Modes (ICM) algorithm. Sentences from NOIZEUS corpus are evaluated from 0 dB to 15 dB Signal to Noise Ratio (SNR) in 4 kinds of additive noise settings: additive white Gaussian noise, car noise, street noise and pink noise. The reconstructed speech using the proposed technique is evaluated in terms of average segmental SNR, Perceptual Evaluation of Speech Quality (PESQ) and Mean opinion Score (MOS).