Zhou Hui, Li Jun, Mao Yaming, Liu Lu, Lu Yiyang
School of Automation, Nanjing University of Science and Technology, Nanjing 210094, China.
J Imaging. 2025 Jul 28;11(8):253. doi: 10.3390/jimaging11080253.
Imaging technologies are widely used in surveillance, medical diagnostics, and other critical applications. However, under low-light conditions, captured images often suffer from insufficient brightness, blurred details, and excessive noise, degrading quality and hindering downstream tasks. Conventional low-light image enhancement (LLIE) methods not only require annotated data but also often involve heavy models with high computational costs, making them unsuitable for real-time processing. To tackle these challenges, a lightweight and unsupervised LLIE method utilizing a dual-stage frequency-domain calibration network (DFCNet) is proposed. In the first stage, the input image undergoes the preliminary feature modulation (PFM) module to guide the illumination estimation (IE) module in generating a more accurate illumination map. The final enhanced image is obtained by dividing the input by the estimated illumination map. The second stage is used only during training. It applies a frequency-domain residual calibration (FRC) module to the first-stage output, generating a calibration term that is added to the original input to darken dark regions and brighten bright areas. This updated input is then fed back to the PFM and IE modules for parameter optimization. Extensive experiments on benchmark datasets demonstrate that DFCNet achieves superior performance across multiple image quality metrics while delivering visually clearer and more natural results.
成像技术广泛应用于监测、医学诊断和其他关键应用中。然而，在低光照条件下，捕获的图像往往存在亮度不足、细节模糊和噪声过大的问题，从而降低了图像质量并阻碍了下游任务。传统的低光照图像增强（LLIE）方法不仅需要标注数据，而且通常涉及计算成本高的重型模型，因此不适用于实时处理。为了应对这些挑战，提出了一种利用双阶段频域校准网络（DFCNet）的轻量级无监督LLIE方法。在第一阶段，输入图像经过初步特征调制（PFM）模块，以引导光照估计（IE）模块生成更准确的光照图。通过将输入图像除以估计的光照图来获得最终增强图像。第二阶段仅在训练期间使用。它将频域残差校准（FRC）模块应用于第一阶段的输出，生成一个校准项，该校准项被添加到原始输入中，以使暗区域变暗，亮区域变亮。然后，这个更新后的输入被反馈到PFM和IE模块进行参数优化。在基准数据集上进行的大量实验表明，DFCNet在多个图像质量指标上都取得了优异的性能，同时提供了视觉上更清晰、更自然的结果。
