A STUDY ON DEEP LEARNING BASED IMAGE CLASSIFICATION TECHNIQUE USING RANDOM FOREST ALGORITHM

Abstract— one essential technique for handling image classification is a convolutional neural network (CNN). It is capable of effectively representing features and has made significant progress in the field of computer vision. However, CNNs often require extensive training time. On the other hand, High classification accuracy and quick training speed are two benefits of random forests (RF). To tackle image classification challenges. This study suggests a hybrid approach that extracts characteristics using CNN, which are then input into RF for classification. Deep neural networks are used in this study to classify images, also well-known as deep learning, through the framework of Keras. Python's integration with Keras makes it the programming language of choice. The study specifically examines input data. DNN was chosen as the optimal method for the training process because it achieved a high level of precision. The findings are examined using the image classification accuracy. Keywords: Deep learning (DL), Convolutional neural network (CNN), Artificial neural network (ANN), Keras