Automatic Detection Method for Daytime Traffic Flow Based on Background Difference and Edge Extraction

Real-time monitoring and analysis of traffic flow have become critical as urban congestion intensifies.To address the current challenges of low accuracy and insufficient detection efficiency in daytime traffic monitoring, this paper introduces an advanced method utilizing background difference and edge extraction techniques.Specifically, it employs the Gaussian Mixture Model (GMM) for dynamic background modeling and Contrast Limited Adaptive Histogram Equalization (CLAHE) to enhance input images.Furthermore, the Faster Regional Convolutional Neural Network (Faster R-CNN) and an optimized Canny edge detection algorithm are utilized for effective target detection and tracking.Experimental evaluations indicate significant improvements over existing benchmarks, achieving accuracy, recall, and F1 scores of 0.99, 0.97, and 0.98, respectively.Practical application tests demonstrate an average vehicle detection accuracy of 99.49%, with robustness assessments confirming the model's reliability under diverse datasets and environmental conditions.This research provides substantial improvements in automatic traffic flow detection, supporting more efficient urban traffic management.