Abstract
The semiconductor material InP plays a key role in optoelectronic devices,
high-speed devices, and fiber optic communications systems. The major problems with
these materials are the high lattice mismatch and variance in thermal expansion
coefficient between InP and Si. This mismatch produces high dislocation density at the
interface and the propagation of the threading dislocations away from the interface into
the device layer is a major concern in optoelectronic applications. Image processing
algorithms play a pivotal role in the medical field, archaeology, and remote sensing.
This work proposes an image processing method to analyze the SEM images of the InP
heteroepitaxy layer to determine the etch pits to confirm whether the substrate is
suitable for optoelectronic applications. In this work, a variant of an anisotropic
diffusion filter for noise reduction on SEM images and Fuzzy C means clustering
method for image segmentation was employed for analysis.