India is the second largest producer of tomatoes in the world. The various diseases significantly reduce both the quality and yield of tomato plants. Early and automatic detection of tomato plant diseases allow farmers to quickly address any issues with tomato crops, which ultimately increase the production of tomato. However, the available machine learning and deep learning technology for tomato plant disease diagnosis often fall short in providing the level of accuracy and efficiency required for modern agricultural demands. To address this gap, this research study proposed a capsule networks-based method for tomato disease detection that emphasized the role of dynamic routing. The proposed TPDD-CapsNet: Tomato Plant Disease Detection Using Capsule Networks employing dynamic routing mechanisms. With the data obtained from the Kaggle Tomato Leaf Disease. The proposed TPDD-CapsNet model significantly performs better over the CNN, VGG16, ResNet50. The classification accuracy of TPDD-CapsNet is 95% for the nine diseases and one healthy class.