In order to enhance the precision and effectiveness of automated image analysis systems, this study offers a thorough investigation into transfer learning-based image categorization. The goals of this study are to examine the usefulness of pre-trained models for picture classification tasks, to look at feature extraction methods, and to compare and contrast various machine learning algorithms. The suggested approach begins with the collection of a large number of photos from a variety of sources, followed by data preparation to get them ready for model training, and finally the creation of a transfer learning pipeline that use InceptionV3 based feature extraction as well as Random Forest based classification. The study results show that the suggested model is better, as it has an outstanding precision of 95.93% on test’s data.