This research proposes a technique for intelligent medication recognition in vending machines. Because the name of the drug is typically the most visible character recognition in the picture, the identification of the medicinal name in this thesis may wind up being the most prominent character recognition in the image. We choose the most linked region to acquire the medical name after finding the text area using SVM and associated components. Second, we separate the text into two pieces (fragment and link), which are then concatenated based on the parameters stated. This is known as a "fragment link." Character recognition software, such as optical character recognition, may also be used (OCR). According to studies, this technology accurately identifies between medicines.