Real-Time Object Detection in Overexposed Images Based on YOLOv5

In the context of rapid advancements in the fields of the Internet of Things (IoT) and artificial intelligence (AI), edge computing technology and its devices have undergone significant development and wide application, leading to the increasingly common deployment of lightweight neural networks on these devices. Despite this, research on real-time object detection tasks on edge computing devices under conditions of poor image quality remains relatively scarce. This is primarily because real-time object detection requires consideration of the computational power limitations of edge computing devices, making it challenging to satisfy both detection quality and speed simultaneously. In response to this challenge, this study proposes a YOLOv5 object detection method based on the Jetson Orin Nano edge computing device, aimed at addressing overexposure issues to ensure that detection accuracy is maintained while effectively solving overexposure correction problems.