Quality Detection and Optimization Model of LED Large-Screen Dynamic Display Based on Edge Calculation and Image Processing

In this paper, a model for detecting and optimizing the dynamic display quality of LED large screen based on edge calculation and image processing is proposed to solve the problem that the dynamic display quality of LED large screen is restricted by data transmission delay, image distortion and uneven brightness in practical application. This model combines the low delay advantage of edge computing and the image enhancement ability of image processing technology. By collecting, preprocessing and analyzing LED large-screen display data in real time, the image is optimized by using multi-scale enhancement architecture and dynamic gamut mapping algorithm. The system uses the improved YOLOv5s framework for anomaly detection, and realizes dynamic parameter adjustment through reinforcement learning (RL) framework to continuously optimize the image processing effect. The experimental results show that the model is significantly superior to the traditional algorithm in image structure similarity, video quality evaluation and color accuracy, and the processing delay is reduced by 37.5%, which meets the real-time requirements. In addition, the model is excellent in reducing cloud dependence and improving the efficiency of parameter adjustment, which provides an efficient and practical solution for the detection and optimization of LED large-screen dynamic display quality.