Enforcing the use of helmets on every bike rider is mandatory nowadays because of the high accident rate and poor road conditions. There are laws regarding safety measures that ensure the use of a helmet. But for now, they involve manual intervention which is not so effective as of now because bike riders sometimes tend to escape without any penalty/fine after breaking the safety rules like wearing a helmet while riding. Automation is a better way to deal with this problem but automation in this area comes with its own challenges. To name a few, Low-quality image frames (low image resolution, pixel density, etc.), rain, dew and fog, and partly hidden faces. The robustness of detection methodology strongly depends on the strength of extracted features and also the ability to deal with the lower quality of extracted data. The first goal of this project is to boost the potency of helmet detection and then recognizing the license number plate recognition. This model consists of many essential steps developed using today’s most advanced amp; optimized CNN, GAN models amp; libraries. It is a classification-based model that uses a supervised learning approach to train CNN and Character Segmentation algorithm. The proposed helmet detection model can be used to detect helmets and recognizes license plates even in adverse conditions using character segmentation and CNN.