Scholarlink

<p>To solve the problem of identification difficulty caused by the appearance similarity and the fuzziness and uncertainty in feature extraction of organic soybeans and conventional soybeans, an organic soybeans identification method of a granular support vector machine (GSVM) was proposed. Firstly, the information granulation process was enhanced to characterize the fuzziness and uncertainty of soybean external features effectively. Secondly, the granular fusion rules are defined to complete the transformation of soybean data. Finally, the granular computing rules of the support vector machine were designed to improve the identification accuracy and stability of the algorithm. This method was applied to the discrimination test of organic soybeans, and the results showed that the discrimination accuracy of the GSVM was 94.34%, which was improved by 3.18%, 12.64%, 11.7%, 2.92%, and 9.29% compared with a decision tree, naive Bayes, <i>K</i>-nearest neighbor, linear discriminant analysis, and traditional support vector machine. Compared with the conventional support vector machine, the precision, recall, F1 value, and AUC value of the proposed method are improved by 8.28%, 9.98%, 9.11%, and 6.25%, respectively. The GSVM realizes the precise identification of organic soybeans, which provides a new idea for organic foods and safety detection.​.</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

Identification of organic and conventional soybeans from external features using granular support vector machine

  • 期刊论文

  • 发表日期：

    2026年2月25日

查看全文
  • Xingyu Fu,
  • Jian Lin,
  • Yuli Chen,
  • Junzhe Zhang,
  • Ying Lin,
  • Tao Wu,
  • Xinmiao Wang

摘要

To solve the problem of identification difficulty caused by the appearance similarity and the fuzziness and uncertainty in feature extraction of organic soybeans and conventional soybeans, an organic soybeans identification method of a granular support vector machine (GSVM) was proposed. Firstly, the information granulation process was enhanced to characterize the fuzziness and uncertainty of soybean external features effectively. Secondly, the granular fusion rules are defined to complete the transformation of soybean data. Finally, the granular computing rules of the support vector machine were designed to improve the identification accuracy and stability of the algorithm. This method was applied to the discrimination test of organic soybeans, and the results showed that the discrimination accuracy of the GSVM was 94.34%, which was improved by 3.18%, 12.64%, 11.7%, 2.92%, and 9.29% compared with a decision tree, naive Bayes, K-nearest neighbor, linear discriminant analysis, and traditional support vector machine. Compared with the conventional support vector machine, the precision, recall, F1 value, and AUC value of the proposed method are improved by 8.28%, 9.98%, 9.11%, and 6.25%, respectively. The GSVM realizes the precise identification of organic soybeans, which provides a new idea for organic foods and safety detection.​.