The detection and removal of highlights are fundamental issues in computer vision and image processing. However, existing methods for highlight removal have certain drawbacks. For example, many existing methods are not suitable for repairing large highlight areas, and the applicability of current methods is relatively limited, requiring re-adjustment of parameters in different usage scenarios. Therefore, this paper proposes an effective deep learning model that can address highlight issues in a wide range of scenarios, specifically designed for removing highlights in individual images. The proposed approach is to first detect the positions of highlights and generate a highlight mask. The training process consists of two main parts. Initially, the highlight removal network is trained. This network, based on a generative adversarial network, generates information about the portions of the image affected by highlights. However, there may still be color deviations in the original highlight positions after generation. To address this, the desthighlighted image and the highlight mask are fed into a color adjustment network for secondary training. This corrects the colors in the repaired area, ultimately producing the final image after highlight removal.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.