A Review of Available Standard Dataset for Automated Medicine Pill Recognition Technology

In this paper, we’ll review available standard datasets for automated medicine pill recognition technology. A process whereupon a dataset people discover or provide highly creative technique and tools for sorting and ranking consumer photos of prescription pills. To the need for healthcare providers and the general public to quickly identify unknown prescription medications all of these databases or datasets were created in response. Most of the prescription pills come in blister, alu-alu containers, or Ziploc packs. After unpacking the medical drugs, it is a very difficult tasks for a patient as well as professional medical staff to differentiate between them due to the lack of variability in size, shape, and color though it is a daily task for all of them. Often it is the case that label of the pack is somehow damaged, and for those how to have to take more than one medication at a time it is difficult to differentiate and find the right one between one and another from where the label is me deface or several pills are mistakenly come down from required containers. This error prone nature of the task as many of the pills have almost similar features, increases the medication error which sometimes appears as fatal for patients. The complications are even worse for elderly people with low vision and visually challenged people are at risk because they are not able to identify the color of the text printed on the pill.