People affected by many diseases in their lives and this effect with the productivity of the individual in society and the lives of the entire person and most of these diseases can be cured by medicines. The problem in pharmaceutical industries, in the actual situation, existing a number of defects incorporated in tablets inadequate fines to granules ratio, inadequate moisture content and poor machine settings can be some of the reasons for those visual defects such as faults in a cover of tablet pill. In addition, the production of medicines and pharmaceutical factories are expanded so it is difficult to control the quality of the tablets after packaging. The aim of this paper is to upgrade this manual tablet-sorting machine into an automated system with the aim of improving the speed and accuracy of the sorting process. The hole in the cover plastic package is common defects that can be found in tablets. Therefore, this defect was considered for the purpose of this paper. Damage on the Cover plastic package can be produced during the production of the packaging and can be produced before packaging therefore; will design system inspection the cover plastic before packaging for less damage money as much as possible, also inspection the cover plastic after packaging because the damage on the cover plastic packaging causes damage in pills and capsules. The proposed system algorithm includes preprocessing and feature extraction using opponent local binary patterns &opponent coordinated clusters representation (it’s the contribution research) and finally the classification of tablets for damage or undamaged issue using an artificial neural network algorithm (ANN) and specifically feed forward back propagation learning. The neural network training with feature extraction from the data after it has been tested. The experimental results are acceptable, the performances of the check pill cover plastic system indicated the total accuracy of 94.4% for testing. Also, the approach using opponent features provides better recognition accuracy than other approaches. The system is evaluated using sensitivity, specificity, and accuracy.

 Rihab Hazim Qasim and Muzhir Shaban Al-Ani

CCR; Hole in the covered plastic; LBP; Packaging; Texture