| Abstrakt: |
Over the years, Agriculture has been a mainstay of life for Indians and about half the working population of Tamil Nadu. Spinach is an integral part of everyone's meal and its nutrient content is higher than other veggies. The nutrients are unique for varied varieties so there is a dire need to classify them and thus to predict them. Furthermore, exactitude prediction leads to easy detection of spinach leaves. In this work, we selected 5 varieties of spinach leaves populated under a huge dataset. We implemented the same employing a Deep Hybrid approach which is a fusion of conventional Machine Learning with state-of-the-art Deep Learning using Orange toolkit. Out of the plethora of these AI Domaine approaches, four classifiers, such as Support Vector Machine (SVM), k-Nearest Neighbour(kNN), Random Forest (RF), and Neural Network (NN) were chosen and implemented. Existing methods using these algorithms have achieved promising results, with individual accuracies of 98.80% (RF), 98.20% (KNN), 99.9% (NN), and 99.60% (SVM). However, the IDHLM aims to surpass these individual performances by integrating them into a cohesive framework. This approach leverages each algorithm's complementary strengths to achieve even higher classification accuracy. The abstract concludes by highlighting the potential of the IDHLM for achieving pristine accuracy in spinach leaf classification. [ABSTRACT FROM AUTHOR] |
