Skip to main content
Journal of Cloud Computing

Advances, Systems and Applications

Fig. 6 | Journal of Cloud Computing

Fig. 6

From: LAE-GAN: a novel cloud-based Low-light Attention Enhancement Generative Adversarial Network for unpaired text images

Fig. 6

Visual comparison of existing methods on the SID dataset, revealing that Retinex, Cycle-GAN and IAT introduce distortion in the enhanced images. Zero-DCE and EnlightenGAN show suboptimal enhancement results with noticeable noise artifacts. On the other hand, LAE-GAN excels in noise reduction and texture detail restoration, delivering the most favorable outcome in terms of image quality

Back to article page