Automatic text imprint analysis from pill images
Siroratt Suntronsuk, Sukanya Ratanotayanon · 2017
Pill identification is a serious concern for pharmacists due to similarity of pill appearances. Pill imprints usually contain important information that can be used to add or search for pill information on existing pill databases. However, current techniques for extracting imprints often give results as vectors which cannot be used with existing databases. Thus, this paper proposed an approach for automatically extracting text on imprints. This approach relied on a set of rules to locate imprint locations and a noise elimination technique to remove problematic pixels in binary images obtained by OTSU's thresholding. Our approach can achieve an F-measure of 0.77 for printed imprints and 0.57 overall.