A Global Optimization Method for Specular Highlight Removal From a Single Image

Wenyao Xia,Terry M Peters,Stephen E Pautler,Elvis C S Chen

doi:10.1109/access.2019.2939229

Wenyao Xia, Terry M Peters + Show 2 more

Open Access

https://doi.org/10.1109/access.2019.2939229

Copy DOI

Journal: IEEE Access	Publication Date: Jan 1, 2019
Citations: 65	License type: CC BY 4.0

Affiliation: Robarts Clinical Trials, Western University

Abstract

The presence of specular highlight is a critical issue for both natural and medical images such as those produced by laparoscopes, which can lead to erroneous visual tracking, stereo reconstruction, and image segmentation. Specular highlight removal from a single image is necessary for image analysis and applications. Due to the differences between natural and medical image scenes, existing literature to address this issue has only been effective on natural images or medical images with textureless regions. To overcome this limitation, we propose a global optimization method for specular highlight removal from a single image based on a dichromatic reflection model. In addition to introducing modified illumination chromaticity, the proposed method consists of two novel steps: one for estimating diffuse chromaticity by correcting hue and saturation on highlighted regions, and the other for estimating diffuse and specular reflection coefficients using convex optimization with double regularization. The estimated diffuse chromaticity is proven to approximate the true diffuse chromaticity and the proposed optimization algorithm is guaranteed to find the optimal diffuse coefficients. Experimental results show that the proposed method can effectively remove specular highlights from both natural images and endoscopic images with texture detail preservation. To further demonstrate the efficacy of our proposed method, an application of stereo reconstruction using a public dataset illustrates that our highlight removal method can enhance surface reconstruction accuracy from 1.10mm RMSD to 0.69mm RMSD.

Highlights

Digital images captured under discrete source illumination often contain specular reflections, which conceal useful image features such as colors and textures [1], [2]
PROPOSED METHOD To reduce the limitation of current highlight removal methods we propose a global optimization method for specular highlight removal from a single image based on the dichromatic model
For specular highlight removal testing, we compare our method with five dichromatic reflection model-based methods: Shen’s [10], Yang’s [11], Suo’s [8], Ren’s [12], and Souza’s [45] methods, and two inpaintingbased methods: Saint-Pierre’s [29] and Bernal’s methods [34]

Summary

Introduction

Digital images captured under discrete source illumination often contain specular reflections, which conceal useful image features such as colors and textures [1], [2]. Aside from natural images, the presence of specular highlights following the advancement of minimally invasive surgeries becomes an important issue for endoscopic images. Due to the proximity of the camera light source and organ surfaces, endoscopic images often suffer from strong specular. Regions with specular highlights may contain vital information relating to the organ such as color and textures. It is desirable to remove specular reflections while preserving the original color and texture details of the organ surface for medical imaging applications. Numerous highlight removal methods have been presented in the literature. They can be categorized into multi-and single-image methods. Wang et al [22] proposed a energy minimization with respect to the local weighting coefficient for highlight removal from multiple images, based

Methods

Results

Conclusion