The management of medical waste has a significant impact on both human health and environmental sustainability. Medical waste segregation techniques currently in use mostly rely on manual sorting, which poses a risk to workers' safety and increases the likelihood of environmental contamination. This study presents a novel automated medical waste classification system that combines brute force methods with convolutional neural networks (CNN) to classify waste into hazardous and non-hazardous categories accurately. Our system utilizes a large dataset that encompasses six distinct categories of medical waste: syringes, eyeglasses, gloves, masks, medical caps, and medications. This allows for accurate and scalable classification, in contrast to traditional methods that are frequently constrained by limited classifications or lack real-time capabilities. One significant innovation in this study is the smooth integration of OpenGL simulations to illustrate the waste sorting procedure, which improves the system's usefulness even more, masks, gloves, syringes, eyeglasses, medical caps, and medicines—using a lightweight sequential CNN architecture suitable for low-resource deployment. The suggested approach outperformed traditional techniques with an astounding accuracy of 99.66% for six distinct types. The entire system outperformed conventional classification techniques, achieving a segregation accuracy of 86.2%, a precision of 0.85, a recall of 0.83, and an F1-score of 0.87 when combined with the brute-force methodology for final binary segregation. Additionally, the classification system effectively classifies garbage according to predetermined characteristics by employing a brute force algorithm, providing a dependable first sorting solution. The results are visually represented through an OpenGL-based bin-sorting simulation, where hazardous objects are dropped into red bins and non-hazardous ones into green bins, demonstrating the decision pipeline from perception to actuation.