Dolphin Inspired Optimization for Feature Extraction in Augmented Reality Tracking

Feature extraction has the prominent role in Augmented Reality (AR) tracking. AR tracking monitor the position and orientation to overlay the 3D model in real-world environment. This approach of AR tracking, encouraged to propose the optimum feature extraction model by embedding the dolphin grouping system. We implemented dolphin grouping algorithm to extract the features effectively without compromising the accuracy. In addition, to prove the stability of the proposed model, we have included the affine transformation images such as rotation, blur image and light variation for the analysis. The Dolphin model obtained the average precision of 0.92 and recall score of 0.84. Whereas, the computation time of dolphin model is identified as 2ms which is faster than the other algorithm. The comparative result analysis reveals that accuracy and the efficiency of the proposed model surpasses the existing descriptors.

@article{S2025,
  title     = {Dolphin Inspired Optimization for Feature Extraction in Augmented Reality Tracking},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {16},
  number    = {1},
  year      = {2025},
  publisher = {The Science and Information Organization},
  author    = {Indhumathi S and Christopher Clement J},
  doi       = {10.14569/IJACSA.2025.01601107},
  url       = {https://doi.org/10.14569/IJACSA.2025.01601107}
}