TreEnhance: A tree search method for low-light image enhancement. (April 2023)

• Record Type: Journal Article
• Authors: Cotogni, Marco; Cusano, Claudio
• Is Part Of: Pattern recognition. Volume 136(2023)
• Publication Date: 2023-04
• Subjects: Low-light image enhancement -- Deep reinforcement learning -- Automatic image retouching -- Image processing -- Tree search; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2022.109249
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: This is the first work presenting the application of tree-search theory to image enhancement problems TreEnhance acts as a fully explainable neural architecture for image enhancement. TreEnhance provides along with the output image the sequence of enhancing operators selected by the algorithm. TreEnhance presents two different inference strategies for enhance low quality image based on the data, time an memory constraints. Abstract: In this paper we present TreEnhance, an automatic method for low-light image enhancement capable of improving the quality of digital images. The method combines tree search theory, and in particular the Monte Carlo Tree Search (MCTS) algorithm, with deep reinforcement learning. Given as input a low-light image, TreEnhance produces as output its enhanced version together with the sequence of image editing operations used to obtain it. During the training phase, the method repeatedly alternates two main phases: a generation phase, where a modified version of MCTS explores the space of image editing operations and selects the most promising sequence, and an optimization phase, where the parameters of a neural network, implementing the enhancement policy, are updated. Two different inference solutions are proposed for the enhancement of new images: one is based on MCTS and is more accurate but more time and memory consuming; the other directly applies the learned policy and is faster but slightly less precise. As a further contribution, we propose a guided search strategy that "reverses" the enhancement procedure that a photo editor applied to a given input image. Unlike other methods from the state of the art, TreEnhance does not pose any constraint on the image resolution and can be used in a variety of scenarios with minimal tuning. We tested the method on two datasets: the Low-Light dataset and the Adobe Five-K dataset obtaining good results from both a qualitative and a quantitative point of view.