Locally Adaptive Block Thresholding Method with Continuity Constraint

TL;DR

This paper introduces a locally adaptive block thresholding algorithm that maintains image continuity by using neighboring thresholds, allowing for enhanced local detail preservation in image processing.

Contribution

The proposed method adaptively determines thresholds for sub-images based on neighboring thresholds, ensuring continuity and improved local detail extraction.

Findings

Range of acceptable thresholds is high across various images.

The method effectively preserves local details while maintaining image continuity.

Threshold adaptation improves image segmentation quality.

Abstract

We present an algorithm that enables one to perform locally adaptive block thresholding, while maintaining image continuity. Images are divided into sub-images based some standard image attributes and thresholding technique is employed over the sub-images. The present algorithm makes use of the thresholds of neighboring sub-images to calculate a range of values. The image continuity is taken care by choosing the threshold of the sub-image under consideration to lie within the above range. After examining the average range values for various sub-image sizes of a variety of images, it was found that the range of acceptable threshold values is substantially high, justifying our assumption of exploiting the freedom of range for bringing out local details.