To address issues such as low detection accuracy and limited real-time performance in road scene detection, a novel road scene detection algorithm based on CMS-YOLO is proposed in this paper. In this algorithm, an efficient backbone called the cross-stage partial DWNeck is devised. By using large-scale depthwise separable convolutions and residual structures, it enables the acquisition of more comprehensive feature information, thereby increasing both the receptive field and the richness of extracted features. Meanwhile, a feature pyramid called the multi-scale fusion feature pyramid network is designed to strengthen the fusion of shallow and deep-level information, effectively preventing the loss of feature information in the transmission process. Besides, a new decoupled head structure called the special decoupled head is introduced, which effectively addresses the conflict between classification and regression tasks through a three-layer joint output structure. Finally, experiments were conducted on two publicly available datasets, namely Udacity Self-Driving and BDD100K. Experimental results indicate that the CMS-YOLO algorithm achieved an impressive detection accuracy of 90.3% and 59.1% in mAP@0.5, demonstrating a remarkable improvement of 4.2% and 7.2% over YOLOv5 respectively. Moreover, in real-world scenarios, the algorithm achieves an impressive real-time detection speed of 34.5 frames per second. These results demonstrate that CMS-YOLO not only meets but also surpasses the requirements for detection accuracy and real-time performance for object detection in autonomous driving scenarios.

Contemporary algorithms for enhancing images in low-light conditions prioritize improving brightness and contrast but often neglect improving image details. This study introduces the Swin Transformer-based Light-enhancing Generative Adversarial Network (SwinLightGAN), a novel generative adversarial network (GAN) that effectively enhances image details under low-light conditions. The network integrates a generator model based on a Residual Jumping U-shaped Network (U-Net) architecture for precise local detail extraction with an illumination network enhanced by Shifted Window Transformer (Swin Transformer) technology that captures multi-scale spatial features and global contexts. This combination produces high-quality images that resemble those taken in normal lighting conditions, retaining intricate details. Through adversarial training that employs discriminators operating at multiple scales and a blend of loss functions, SwinLightGAN ensures a seamless distinction between generated and authentic images, ensuring superior enhancement quality. Extensive experimental analysis on multiple unpaired datasets demonstrates SwinLightGAN's outstanding performance. The system achieves Naturalness Image Quality Evaluator (NIQE) scores ranging from 5.193 to 5.397, Blind/Referenceless Image Spatial Quality Evaluator (BRISQUE) scores from 28.879 to 32.040, and Patch-based Image Quality Evaluator (PIQE) scores from 38.280 to 44.479, highlighting its efficacy in delivering high-quality enhancements across diverse metrics.

Aiming at the problems of color bias and noise enhancement on the output image by the traditional low-light image enhancement algorithm, we propose a deep multi-scale pyramid hybrid network (DMPH-Net) algorithm that fuses attention mechanism and multi-scale pyramid is proposed in this paper. The algorithm uses DecomNet to decompose reflectance and light components, and uses multi-scale illumination attention module to fuse light and reflectance for the decomposed low-light reflectance to improve the realism and details of reflectance; by using five-layer feature pyramid and kernel selection in PRID-net module to achieve the fusion of contextual information between different scale feature layers, while effectively removing the enhanced of noise, and the added color loss effectively suppresses the color bias of the output image; using multi-scale cascading and channel attention mechanisms to adjust the illumination and fuse the illumination ratios, effectively enhancing the brightness, texture, and other feature information in the image. The DMPH-Net algorithm is experimentally validated on LOL and no-reference LIME, MEF, and NPE datasets, and the objective evaluation metrics PSNR $$\uparrow $$↑ , SSIM $$\uparrow $$↑ , LIPIPS $$\downarrow $$↓ , and NIQE $$\downarrow $$↓ are 23.3772, 0.8442, 0.1386, and 3.5966 on LOL dataset. The objective evaluation metrics NIQE on the reference-free datasets LIME, NPE, and MEF are 3.0735, 3.1711, and 2.9464, respectively. The experiments show that the DMPH-Net algorithm maintains high image details and textures in image enhancement and denoising, effectively enhances low-light images, and reduces noise and color bias of images. Compared with RUAS, UnRetinex-Net, and other enhancement algorithms, it improves in objective evaluation metrics PSNR, SSIM, LIPIPS, and NIQE.

This paper investigates the delay-limited throughput of a satellite-terrestrial relay network with hybrid automatic repeat request (HARQ) protocol in which a satellite communicates with a user with the aid of an amplify-and-forward (AF) terrestrial relay. Particularly, the proposed analysis includes both the channel state information (CSI)-assisted and the fixed gain AF relaying protocols for two representative HARQ schemes, namely, chase combining (CC) and incremental redundancy (IR). Numerical results are provided to evaluate the performance of different HARQ schemes and relay protocols, and also indicate the impact of key parameters, such as the HARQ round number and target information rate, on the achievable performance.