Detection of the cyclone eye is the primary task to track the cyclone path. Cyclone path prediction helps alert the government and people in cyclone-prone areas so adequate preparations are made and the necessary precautions taken. We used Synthetic Aperture Radar (SAR) cyclone images for an analysis of Tropical Cyclones in India. India faces an average of 3 cyclone landfalls every year and is surrounded by more intensified cyclone-generated Ocean and Seas like the Indian Ocean and Bay-of-Bengal. This highlights the necessity of cyclone eye detection in the cyclone tracking system. Given the lack of SAR images for cyclones in India, designing cyclone eye detection and path prediction models is a challenge. Hence, we introduced an Object Translation-based Image Augmentation (OTIA) technique under the no-deep learning-model type of data augmentation. OTIA artificially expands the cyclone image dataset which helps the YOLO to build an efficient model that detects the eye of the cyclone. OTIA generates 8 different images from one cyclone image with a relocated cyclone structure that increases the number of images 80 times more than in the original dataset. The implementation results have shown that the YOLO model with a proposed image augmentation technique detects the cyclone eye with high prediction accuracy.