Optimized Plant Species Classification through MobileNet-Enhanced Hybrid Models

Medicinal plants are used to cure various diseases such as asthma, cancer, heart disease etc. Plant identification plays a pivotal role in various domains such as agriculture, ecology, and environmental conservation. Manual identification of plant species is tedious and time consuming as there exist inter class similarity and intra class variations among the plant species that may lead to human error. Hence automatic identification is required based on machine learning and deep learning models. In this study, we propose two novel hybrid models for accurate plant species classification, leveraging the combination of deep learning architectures. The first hybrid model integrates InceptionNetV3 and MobileNet, achieving an accuracy of 88.11 %, while the second hybrid model combines AlexNet and MobileNet, outperforming the first model with an accuracy of 93.86 %. Both hybrid models employ deep learning classifiers. This research underscores the efficacy of hybrid models in enhancing plant species classification accuracy and contributes towards advancing the plant identification. The experiments are carried out using self-built datasets that are captured in unconstrained environment