Photo-to-Sketch Transformation using Contour Accentuation for High-Quality Outputs
Philomina Princiya Mascarenhas, M S Sannidhan, Niveditha Niveditha, Dabis Camero, Jason Elroy Martis · 2025
Sketches play a vital role in forensic investigations and other fields but creating them is a time-consuming process that relies heavily on human effort. We have explored various image processing techniques to assist in sketch generation and research has shown that production of high-quality sketches from photographs poses a significant challenge. This article introduces a Contour Accentuation Technique (CAT), which is a method designed to improve composite sketch generation and enhance facial feature visibility in sketches. The proposed works in four key steps namely Grayscaling, Blending, Contrast Correction and Edge Enhancement. Also, the technique removes color information, helping the system focus on intensity variations by sharpening the edges and reducing noise. It also improves contrast thereby enhancing the visibility of facial contours by adjusting intensity levels. Experimental results show significant improvements: the NIQE score decreased from 3.39 to 2.76, SSIM increased from 0.82 to 0.88, and Absolute Error (AE) dropped from 89.26 to 71.41, confirming the effectiveness of CAT in improving sketch clarity and quality.