A novel hybrid methodology for character detection, segmentation and recognition from English Language Vehicle Number Plates is proposed by combining Convolutional Neural Networks (CNNs) with tailored image processing techniques. CNNs are employed as the primary tool for character detection, leveraging their capacity for automatic feature learning and spatial hierarchies recognition. These networks excel at discerning characters amidst complex backgrounds and lighting conditions, adapting well to diverse real-world scenarios. Additionally, the approach incorporates tailored preprocessing techniques optimized for license plate characteristics, enhancing accuracy and robustness. By synergistically combining CNNs with preprocessing, the proposed methodology effectively addresses the unique challenges posed by Indian license plates, such as variations in font styles, non-uniform lighting conditions, occlusions, and background clutter. The proposed work detects and segments the characters with high precision, offering significant advancements in Indian license plate recognition for applications including automated toll collection, law enforcement, and vehicle tracking systems.