Detecting tilted and skewed vehicle license plates is a significant challenge, often leading to reduced detection accuracy. Traditional detection models struggle with such variations, affecting the efficiency of automated license plate monitoring systems. This research devised a novel skewed license plate detection model that effectively addresses the issue of tilted and skewed license plates. The proposed model calculates the four coordinates and the centroid coordinates of each plate. To differentiate the angular distortion of the tilted plates, the inverse of theta is calculated to convert the slope into an angle. Based on estimated angles the license plates are categorized as straight horizontal (𝜃 ≈ 0°), clockwise skewed (𝜃 > 0°) and anticlockwise skewed (𝜃 < 0°). The tilted coordinates then transform the quadrilateral plate into a rectangular shape using the homography matrix. Applied to a dataset of 1100 vehicle images with varied skewed and complex weather conditioned images, the model demonstrates enhanced feature extraction and localization precision. Experimental results show 99% detection accuracy, with notable gains in precision, recall, and mAP, outperforming existing methods. The proposed model supports SDG 11: Sustainable Cities and Communities through enhanced intelligent traffic systems.