基于深度学习的药片识别技术研究进展

打开文本图片集

中图分类号：TP18 文献标志码：A 文章编号：2095-2945（2025）34-0005-07

Abstract：Deeplearninghasbeenwidelyappliedinthe'healthcarefield，promotingtheintellgentdevelopmentofhealthcare services.Curently，artificialinteligencemainlyfocusesondiseasepredictioninclinicalenvironments，withrelativelylitle researchonthemedicationsthemselves.However，ensuring medicationsafetyandavoidingmedicationerorsiscrucialfor guaranteeing thecorectnessofthemedicationsthatptientstake.Thispapersystematicallyoutlinestherelevantresearchand developmenthistoryofpilldetectionandrecognitiontechnology，focusingonthemethodsbasedondeeplearning.Indep learning，theapplicationresearchofonvolutionalneuralnetworksisparticularlyimportant.Comparedwithtraditionalmachine learningmethods，deplearningshowshigherflexiblityandacuracyinfeatureextraction.Inadition，thispaperalsointroduces publicdatasetsrelatedtotabletidentification，andanalyesandprospectstheproblemsfacedbycurrnttabletidentification technology.

Keywords： deep learning; feature extraction; convolutional neural network；target detection； pillrecognition

国家食品药品监督管理总局经调查显示，我国每年有250万人因为用药错误而损害健康，其中约有20万人死亡，是每年全国交通事故死亡人数的2倍，医疗失误同时也是第三大最常见的死亡原因，其中最常见且可预防的医疗失误类型是用药失误。（剩余9097字）