Publication Type : Journal Article
Publisher : International Journal of Applied Engineering Research
Source : International Journal of Applied Engineering Research, 10 (5), pp. 13155-13168, 2015.
Url : https://www.scopus.com/inward/record.uri?eid=2-s2.0-84927728899&partnerID=40&md5=4da0ed178782ed8a7505b4afcf36761e
Keywords : Broken tablets, Correlation features, Entropy filtering, Histogram based approach, Pharmaceutical Industry, Tablet inspection
Campus : Mysuru
School : School of Arts and Sciences
Department : Computer Science
Year : 2015
Abstract : Proper medication always increases the chances of speedy recovery for any type of diseases or illness. In this context inspection of tablet production in pharmaceutical industries is considered to be one of the prime necessity. Tablet production will takes place at large scale, it is very unusual to perform the manual inspection on each and every tablet blister to identify whether it is defective or not. An automated inspection and verification systems that can perform the identification of defective tablets blisters is required. The numerous industrial applications of image processing assists in development of an automatic inspection and verification systems for very large scale production of various products in the industries. In this paper we mainly focus on development of an image processing application software for identification of broken and unfilled tablet pills in the tablet blister. The algorithm is developed by incorporating the constructs like entropy based filtering, quad tree based binarization approach, histogram processing to perform the segmentation and correlation features based object recognition for the classification of tablet pills into defective class or non-defective class. The experimental results are very encouraging and have attained an average accuracy of 96.33% for tablets of various geometrical properties and dimensions. © Research India Publications.
Cite this Research Publication : Shobha Rani, N., Nithusha, V.K., Roshna, T.P., "Automatic recognition and verification of defective tablet blisters using entropy based filtering and histogram processing," International Journal of Applied Engineering Research, 10 (5), pp. 13155-13168, 2015.