02025nas a2200253   4500000000100000000000100001008004100002260001200043653001800055653003400073653003500107653002300142100002000165700001600185700001500201700001300216700001800229245011600247856005800363300000900421490001300430520131400443022001401757       9998                            d  c05/202410aDeep Learning10aDiabetic Plant Identification10aDiscriminant Subspace Ensemble10aInternet of things1 aN. Sasikaladevi1 aS. Pradeepa1 aA. Revathi1 aS. Vimal1 aGaurav Dhiman00aAnti-Diabetic Therapeutic Medicinal Plant Identification Using Deep Fused Discriminant Subspace Ensemble (D2SE)  uhttps://www.ijimai.org/journal/bibcite/reference/3453  a1-110 vIn press3 aAbout 422 million people worldwide have diabetes, the majority living in low-and middle-income countries, and 1.5 million deaths are directly attributed to diabetes each year. According to the Botanical Survey of India, India is home to more than 8,000 species of medicinal plants. The natural medications with antidiabetic activity are widely formulated because they have better compatibility with human body, are easily available and have less side effects. They may act as an alternative source of antidiabetic agents. The fused deep neural network (DNN) model with Discriminant Subspace Ensemble is designed to identify the diabetic plants from VNPlant200 data set. Here, the deep features are extracted using DenseNet201 and the matrix-based discriminant analysis is adopted to learn the discriminative feature subspace for classification. To further improve the performance of discriminative subspace, a nearest neighbors technique is used to produce a subspace ensemble for final diabetic therapeutic medicinal plant image classification. The developed model attained the highest accuracy of 97.5% which is better compared to other state of art algorithms. Finally, the model is integrated into a mobile app for robust classification of anti-diabetic therapeutic medicinal plant with real field images.  a1989-1660