Research On Medicine Label Recognition And Sorting Based On Machine Vision

Intravenous infusion is an important means of clinical drug treatment,which resulted in a huge infusion dispensing workload,in order to reduce the workload of health care workers to improve the quality of infusion dispensing,a class of medical personnel can replace the preparation of automatic dispensing of pharmaceutical preparations Robots are gradually being introduced into major medical institutions.In this paper,machine vision technology is introduced to study the identification of drug labels and the sorting of drugs.In this paper,the identification of drug labels and the sorting of drugs are carried out.Drug tag identification is divided into four key steps: First,the vial outline picking,two,vial image plane correction,three,the state quasi-word number positioning and extraction,four,tag character recognition.The Canny operator was used to detect the edge of the vial,and the vial image was separated by vertical template detection and image projection.According to the principle of small hole imaging,a mathematical model of plane expansion of cylindrical bottle is established,which effectively unfolds the vial image in the camera field of view into a plane,and eliminates the distortion of the drug label character.The SIFT image matching algorithm is used to complete the fast positioning of the Chinese characters on the label image.Combining with the search and merging algorithm of the character stroke and concatenation domain,the character extraction of the quasi-word number is realized.This paper proposes a candidate character extraction algorithm based on statistical features and two character recognition algorithms based on BP neural network and exclusive-OR enhancement algorithm.The tag characters are used to identify the one-vote veto system to ensure the accuracy of the drug recognition results.Identify the feasibility of the algorithm.Design and simulation analysis of sorting mechanism for sorting operation of wrong medicine.According to the characteristics of the movement of the vials on the conveyor belt and the actual drug sorting requirements,the two-degree-of-freedom cylindrical coordinate arm robots for the picking of the vials and the moving of the grips were designed and the corresponding pneumatic system loops were designed.ADAMS is used to simulate the overall motion of the sorting mechanism,which proves that the structure is reasonable and can perform the sorting action correctly.The key performance parameters such as the closing speed and clamping force of the handle are simulated and analyzed.The results show that the design of the grasping structure can meet the demand and lay the foundation for the next optimization design.