Reinforcement Learning Based Dark Image Enhancement Through Color Feature Balancing
Bibliographic record
Abstract
Image enhancement refers to the process of manipulating an image to improve the visual information it contains. Images can degrade due to various factors, such as operator incompetence or low-quality picture- recording devices. The degraded images exhibit color imbalance, unwanted noise, and hue saturation disparity, especially in dark images. Furthermore, any photograph taken in low-light conditions typically falls short of attaining the desired level of visibility and essential details of the image. Most image enhancement systems depend on a predetermined set of instructions. Deep learning approaches instruct and execute certain actions during training without knowledge of the properties of individual images. Insufficient knowledge occasionally leads to the distortion of images. Conversely, an agent that utilizes Reinforcement Learning (RL) can make decisions about which actions to perform based on the input image. In order to address the aforementioned challenges, we present a novel approach in this study that use Reinforcement Learning to enhance dark images. This method accurately emulates the sequential process employed by humans during retouching. Our reinforcement learning-based agent, similar to a human expert, aims to evaluate the present state of the image prior to making any decision. Upon conducting a comprehensive analysis, we have determined that our agent, which utilizes reinforcement learning (RL), has demonstrated exceptional efficacy in maintaining the authenticity and color equilibrium of dark images. In general, the experimental results demonstrate the outstanding efficacy of the proposed approach.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".