An In-Depth Investigation Into The Effects Of Thresholding Techniques On Grayscale Images

Abstract

This communication presents an extensive survey of diverse image thresholding techniques applied to a standardized dataset of five 256x256 grayscale images. The evaluated methods include mean thresholding, histogram thresholding, edge thresholding, variable thresholding, and P-tile thresholding. Each technique's algorithm parameters are meticulously optimized to cater to the specific characteristics inherent in each image. This survey emphasizes the critical role of selecting appropriate thresholding techniques tailored to specific image characteristics and application requirements. Histogram thresholding emerges as a preferred method due to its consistent ability to achieve superior BPR results. Continued research efforts are crucial for advancing thresholding methodologies and expanding their applicability across diverse fields, such as study of medical images, remote sensing, and industrial quality control.