An end-to-end network is proposed for low-light images natural colorization using a deep fully convolutional architecture. The network consists of a downsampling sub-network and an upsampling sub-network. The downsampling component extracts the high-level features of the input images, while the upsampling component transforms the high-level features to color. A skip connection is used to transmit low layer information to the deep layer so as to improve the colorization accuracy. Gamma correction and random noise augmentation are used to improve the network adaptability to low-light images. The trained model can naturally colorize low-light images without any reference image or artificial scribbles.
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.