YOLO v9‐S Net: YOLO V9 Squeeze SegNet for Object Detection Using Vehicle Image

ABSTRACT Object detection plays a vital role in autonomous driving vehicular systems and intelligent transportation for better environment perception by understanding and analyzing the scenes. Accurate and real‐time object detection is critical for autonomous driving and intelligent transportation systems to perceive and understand complex traffic environments. However, existing object detection techniques often suffer from high computational costs and longer processing times, limiting their efficiency in real‐world settings. This creates a need for a more computationally efficient and precise detection method that can robustly identify objects from vehicle images. Thus, this article presented a You Only Live Once v9 Squeeze M‐SegNet (YOLO v9‐S Net) for the detection of objects from vehicle images. To accurately detect objects, the input vehicle images are initially denoised using an adaptive weighted median filter. The enhancement of the denoised vehicle image is performed using Contrast Limited Adaptive Histogram Equalization (CLAHE) to increase the image quality. Following this, the segmentation of objects is executed using fast fuzzy clustering, and the objects are accurately detected from the segmented object using the YOLO v9‐S Net model. The results obtained from the experiment demonstrate that the YOLO v9‐S Net approach attained high detection performance with F1‐score, recall, and precision of 92.81%, 92.58%, and 93.04%.