Chapter 10: Deep Retinal Image Non-Uniform Illumination Removal

Retinal images have been widely used by clinicians for early diagnosis of ocular diseases. However, the quality of retinal images is often clinically unsatisfactory due to eye lesions and imperfect imaging processes. The non-uniform or poor illumination on retinal images hinders the pathological information and further impairs the diagnosis of ophthalmologists. To solve these issues, we propose a deep learning-based retinal image non-uniform illumination removal called NuI-Go, which combines the powerful capabilities of convolutional neural networks (CNNs) with the characteristics of retinal images with non-uniform illumination. Concretely, the proposed NuI-Go consists of three Recursive non-local encoder–decoder residual blocks (NEDRBs) for progressively enhancing the degraded retinal images. Each NEDRB contains a feature encoder module that captures the hierarchical feature representations, a non-local context module that models the context information, and a feature decoder module that recovers the details and spatial dimension. Extensive experimental results demonstrate that the proposed method performs favorably against the state-of-the-art methods on both synthetic dataset and real retinal images. Besides, we further demonstrate the advantages of the proposed method for improving the performance of retinal vessel segmentation.