Automated License Plate Recognition (ALPR) is vital in various Intelligent Transportation Systems applications. ALPR faces challenges in unconstrained scenarios such as blurred images due to fast-moving vehicles and adverse weather conditions. To address this, we introduce an Attention-based image resolution Generative Adversarial Network (AIRGAN), a super-resolution model incorporating a self-attention mechanism to enhance image resolution. The images are then fed into the Automatic Warped Planar Object Detection (AWPOD) network. Later, the images are cropped to pass to the character segmentation model for isolating individual characters and performed using the MobileLPRnet model. Experiments shows the effectiveness in improving accuracy, even in adverse weather conditions and vehicle movement. The proposed system holds promise for enhancing the reliability and efficiency of ALPR systems in various transportation applications.