Scholarlink

<p>Herbal medicine fruits play a vital role in traditional medicine systems, where accurate identification is essential for prescription verification and quality control. However, distinguishing among numerous herbal medicine fruits, some of which share similar appearances but have distinct therapeutic effects, presents a significant challenge. Traditional manual identification methods are labor-intensive and inefficient for handling large datasets. Deep learning techniques offer a promising solution by enhancing accuracy, improving efficiency, and reducing costs associated with fruit identification. This research paper introduces a novel deep learning-based ensemble model to classify herbal medicine fruits. The model integrates three efficient transfer learning architectures—MobileNet-V2, VGG-16, and ResNet—leveraging their distinct strengths. Two diverse datasets are employed to evaluate the model's performance comprehensively. Experimental results demonstrate the proposed model's superior classification accuracy, achieving 99.17% on dataset-1 and 97.75% on dataset-2. This research showcases the effectiveness of deep learning in automating herbal medicine fruit classification, underscoring its potential for enhancing traditional medicine quality assurance practices.</p>

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

An ensemble-based feature fusion approach for robust classification of herbal medicine fruits

  • 期刊论文

  • 发表日期：

    2025年7月10日

查看全文
  • S. Ida Evangeline,
  • S. Darwin

摘要

Herbal medicine fruits play a vital role in traditional medicine systems, where accurate identification is essential for prescription verification and quality control. However, distinguishing among numerous herbal medicine fruits, some of which share similar appearances but have distinct therapeutic effects, presents a significant challenge. Traditional manual identification methods are labor-intensive and inefficient for handling large datasets. Deep learning techniques offer a promising solution by enhancing accuracy, improving efficiency, and reducing costs associated with fruit identification. This research paper introduces a novel deep learning-based ensemble model to classify herbal medicine fruits. The model integrates three efficient transfer learning architectures—MobileNet-V2, VGG-16, and ResNet—leveraging their distinct strengths. Two diverse datasets are employed to evaluate the model's performance comprehensively. Experimental results demonstrate the proposed model's superior classification accuracy, achieving 99.17% on dataset-1 and 97.75% on dataset-2. This research showcases the effectiveness of deep learning in automating herbal medicine fruit classification, underscoring its potential for enhancing traditional medicine quality assurance practices.