Scholarlink

New developments in deep learning have highly improved the way of image classification and helped areas like medical botany. In this study, identification of medicinal plants from leaf images was assessed using three CNN structures including ResNet50, Dense Net, and Alex Net. Evaluations were done using accuracy, precision, recall and F1-measure. The highest accuracy of 98% was obtained by using ResNet50 for feature acquisition with dense acquisition as the closest follower, with an accuracy of 96%. While being rather fast to process input data, Alex Net was a bit delayed because of its less complicated structure. The exercitation of CNNs showed the capacity to enhance the recognition of medicinal plant and benefited the further progress of natural medicine and reasonable use of resource.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Enhancing Medicinal Plant Prediction Through Deep Neural Network Algorithms

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • M. Swapna,
  • C. Sireesha,
  • Sanjana Gavada,
  • Barigela Sreeja

摘要

New developments in deep learning have highly improved the way of image classification and helped areas like medical botany. In this study, identification of medicinal plants from leaf images was assessed using three CNN structures including ResNet50, Dense Net, and Alex Net. Evaluations were done using accuracy, precision, recall and F1-measure. The highest accuracy of 98% was obtained by using ResNet50 for feature acquisition with dense acquisition as the closest follower, with an accuracy of 96%. While being rather fast to process input data, Alex Net was a bit delayed because of its less complicated structure. The exercitation of CNNs showed the capacity to enhance the recognition of medicinal plant and benefited the further progress of natural medicine and reasonable use of resource.