Past Projects

Any image can be described as a product of illumination and reflectance of the subject. Very often, high variations in the illumination lead to poor quality photos since the camera is unable to capture the entire range of intensities. The resulting image is over exposed or under exposed in some areas. Such an image can be enhanced with the help of Homomorphic Transform. In this method the illumination and reflectance are separated and filtered in frequency domain to achieve the desired result. As a final project for a graduate course on Digital Image Processing at University of Houston, I have developed Homomorphic and Laplace transform filters for image enhancement. The programing has been done in Python. The work has been completed in collaboration with five other classmates, each working on a unique set of filters. Our efforts led to a user-friendly GUI which incorporates all filters on a single platform. The code can be provided upon request.

Sum Frequency Generation Spectroscopy (SFG) provides an accurate and versatile tool for studying surface chemistry. At Surface Chemistry Lab, the spectroscopic system requires hardware interfacing with legacy devices, synchronous control of multiple systems, and automation. I developed the computer-hardware interface for spectroscopic components such as monochromator, time gated signal integrators, OPG/OPA (motor controls for crystals), and other instruments through communication protocols such as GPIB, Ethernet, USB, DAQ etc. Furthermore, to provide seamless automation of the spectroscopy, I developed a LabVIEW program that controls these instruments, collects the signal and performs basic spectral analysis.

The details and LabView files can be provided upon request.

The magnetic properties of a material sample, such as magnetic susceptibility, can be reliably and quantitatively measured in a Vibrating Sample Magnetometer (VSM). However, off the shelf versions are quite expensive, making them unattainable for scientific research in Pakistan. This project aims to model, construct and analyze a versatile VSM for scientific research. The instrument consists of a mechanism to vibrate the sample in a strong magnetic filed of an electromagnet, a homemade set of pick-up coils to detect the magnetic flux from the magnetized sample as it vibrates in the magnetic filed, and a set of data acquisition and amplification electrons that send the coil signal to the computer.

The mechanical vibrations are provided by a mechanical wave driver (SF-9324 by PASCO). It is driven by a sine wave input signal with a tunable frequency of 0.1 Hz to 5000 Hz. The amplitude of the oscillation from this driver were quantified with a linear variable differential transformer (LVDT). To reduce the cost of the instrument further, the transfer function for the mechanical oscillator has been computed for the whole range of input frequencies using Bode plots; hence removing the need to keep LVDT permanently.

The pick-up coil geometry has been simulated in MATLAB, selecting the optimal solution to maximize the desired signal detection. The coils are housed in a wooden block inside the magnetic filed and around the vibrating sample. The output of the coils is amplified with an op-amp and sent to computer where a homemade LabView program implements the automation of the experiment as well as the recording and processing of data. Tests on multiple samples confirm the performance of the instrument.

The project involved all aspects of research and development for instrumentation. The work comprises of LabView and Matlab programming, machining and construction of components, computation and simulation of of instrument response, and quantitative analysis of the performance. The details can be found here.

Diffraction is one of the remarkable consequences of the wave nature of light. This experiment is developed to teach the phenomenon through diffraction patterns for single slit and double slit arrangements, illustrating the relation between the shape of the diffraction pattern and that of the slit which creates it. The process has been further refined by the development and integration of computer vision and image analysis techniques. A key issue in automating quantitative analysis of diffraction patters is presented by the high dynamic range of intensities, which render standard cameras insufficient via intensity saturation. The homebuilt program captures the diffraction patters with multiple exposure levels of a standard camera to cover the entire range of intensities. It then combined these exposures to generate a HRD image by utilizing the relation between illumination and pixel intensity. Finally, a quantitative diffraction pattern is obtained by converting the pixel positions into a physical distance using geometric optics and camera calibration through computer vision. The code has been implemented in Matlab. This experiment further provides a Matlab GUI for simulating single-slit diffraction. The details can be found on LUMS University's course website here.

Polarization is one of the fundamental properties of light. This experiment provides a fundamental understanding of the polarization as well as the effects of different optical components such as polarizing beam splitter and quarter wave plate. The analysis is performed using Jones Calculus. The details can be found on LUMS University's course website here.

One can completely determine the polarization of light by simply using a polarizer and a quarter wave plate. This optical setup provides a method to generate as well as analyze different polarization of light using these two components and Fourier transform. The details can be found on LUMS University's course website here.

The output power and wavelength of a laser diode can be modulated by varying its current and temperature. This experiment has been developed to provide an understanding of how a laser diode’s optical power and wavelength can be varied by controlling its temperature and operating current. The temperature control has been implemented through integrating Peltier heaters in laser diode mount TCLDM09 with a homebuilt power circuit controlled by a homebuilt LabView program that operates on a Partial Integral Control logic. The results indicate that the wavelength can be varied from 784nm to 797 nm with a temperature variation of 20 to 55 degree C. The details can be found on LUMS University's course website here.