VLSI Implementation of Iris Localization Using Circular Hough Transform

Deep learning-based models such as the interactive variant of U-Net and deep multi-task attention networks, while powerful, are computationally intensive, require large-scale annotated datasets and involve complex parameter tuning—factors that limit their practicality for fundamental tasks like edge detection. In contrast, classical methods such as Canny edge detection and the Circular Hough Transform offer efficient, robust and interpretable alternatives. Canny edge detection provides noise resilience, sub-pixel accuracy and thin edge localization, making it well-suited for object recognition and segmentation. The Circular Hough Transform is highly effective in detecting circular patterns with varying radii and orientations, finding broad applications in medical imaging, iris recognition, industrial inspection and robotics. These classical techniques continue to serve as essential tools for feature extraction, shape analysis and object tracking across diverse domains, especially where computational efficiency and reliability are paramount.