Comparative Analysis of Deep Learning Models and ResNet101-SVM Ensemble for Effective Garbage Classification

Modern waste management systems must include effective classification in order to preserve resources and maintain a sustainable ecosystem. This study compares deep learning models and concludes that ResNet-101 performs better at classifying rubbish. For garbage classification, a hybrid model called ResNet101-SVM ensemble is created in an effort to increase performance metrics. In this context, the performance of deep learning models, which are renowned for their capacity to automatically extract features and categorise objects, is evaluated. Furthermore, the ensemble ResNet101-SVM, including a Support Vector Machine (SVM) for classification and a deep convolutional neural network (ResNet101) for feature extraction, is assessed. The outcomes demonstrate the advantages and disadvantages of each strategy. Depending on variables like deployment needs, computing capacity, and data availability, deep learning models or the ResNet101-SVM ensemble may be selected. This study adds to the continuing efforts to improve trash categorization techniques, which will ultimately support more effective waste management and a more sustainable, clean environment.