Deep learning algorithm based tablet drug screening system

At present, the world's major pharmaceutical companies often identify defective tablets by visual inspection, but because of fatigue staff and human eye's recognition limit tiny parts, so there are missed inefficient inspection phenomenon.[1] How to effectively monitor ensure more qualified rate in activities, which is current problems encountered companies. efficiently detect maintain a high detection testing process, currently difficulties faced In order improve service quality company, adoption machine sorting technology replace manual has become trend. [2]Therefore, after systematic research, we obtained method for monitoring surface defects tablet products based on convolutional neural network. Using this method, developed screening technique defects, significantly improved accuracy efficiency defect detection.