LLNet: A deep autoencoder approach to natural low-light image enhancement. (January 2017)

• Record Type: Journal Article
• Authors: Lore, Kin Gwn; Akintayo, Adedotun; Sarkar, Soumik
• Is Part Of: Pattern recognition. Volume 61(2017:Jan.)
• Page Start: 650
• Page End: 662
• Publication Date: 2017-01
• Subjects: Image enhancement -- Natural low-light images -- Deep autoencoders; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2016.06.008
• Languages: English
• ISSNs: 0031-3203

Abstract

Abstract: In surveillance, monitoring and tactical reconnaissance, gathering visual information from a dynamic environment and accurately processing such data are essential to making informed decisions and ensuring the success of a mission. Camera sensors are often cost-limited to capture clear images or videos taken in a poorly-lit environment. Many applications aim to enhance brightness, contrast and reduce noise content from the images in an on-board real-time manner. We propose a deep autoencoder-based approach to identify signal features from low-light images and adaptively brighten images without over-amplifying/saturating the lighter parts in images with a high dynamic range. We show that a variant of the stacked-sparse denoising autoencoder can learn from synthetically darkened and noise-added training examples to adaptively enhance images taken from natural low-light environment and/or are hardware-degraded. Results show significant credibility of the approach both visually and by quantitative comparison with various techniques. Abstract : Highlights: Novel application of stacked sparse denoising autoencoder enhances low-light images. Simultaneous learning of contrast-enhancement and denoising (LLNet). Sequential learning of contrast-enhancement and denoising (Staged LLNet). Synthetically trained model evaluated on natural low-light images. Learned features visualized to gain insights about the model.