BIOSYNTHESIZED MgO-ZnO METAL OXIDE NANOCOMPOSITE IMAGE ANALYSIS USING MACHINE LEARNING ALGORITHMS
Parashuram Bannigidad1, Sagar Chingali2, Prabhuodeyara M Gurubasavaraj3
1,2Department of Computer Science, 3Department of Chemistry,Rani Channamma University, Belagavi-591156, Karnataka
ABSTRACT: Metal oxide nanocomposites are the composites which are formed by combining two or more metal oxides. When two metal oxides, like MgO (Magnesium Oxide) and ZnO (Zinc Oxide), are combined, they will have improved stability, superior optical qualities, and increased catalytic activity. Different oxide components can interact to create synergistic effects that improve efficiency in applications such as sensors, photo catalysis, and antibacterial activity. This paper suggests biosynthesis of MgO-ZnO metal oxide nanocomposite using 5.12g of Mg (NO3)2•6H2O (0.1M) (200mL) and 50ml rudanti fruit extract and using machine learning techniques for segmenting and classifying MgO-ZnO metal oxide nanocomposites based on their size and shape using machine learning approaches. We have used K-means segmentation and Random Forest classification machine learning techniques to extract the geometrical properties of these oxides. The size of each nanocomposite were extracted from the SEM image using the area of the pixels and categorized the segmented nanocomposites in various ranges such as: 0 nm–50 nm, 51 nm–100 nm, 101 nm–150 nm, 151 nm–200 nm, 200+ nm. The Random Forest classifier technique is used to categorize three different shapes namely circular, triangular, and elliptical. We have got 79% accuracy for classification of MgO-ZnO nanocomposites with f1-score (87%), precision (84%), and recall (91%).
KEYWORDS: Biosynthesized Metal Oxides, Image Analysis, MgO-ZnO, Nanocomposites, Machine Learning, K-means Clustering, Random Forest classifier.