An Efficient Illumination Invariant Tiger Detection Framework for Wildlife Surveillance
Gaurav Pendharkar, A. Ancy Micheal, Jason Misquitta, Ranjeesh Kaippada
TL;DR
The work investigates the existence and nontriviality of $T$-periodic solutions for convex Hamiltonian systems in the subquadratic at infinity setting. It leverages a variational dual action framework, introducing spectral gap conditions involving $\gamma$ and $\lambda$ to guarantee at least one nontrivial periodic orbit via minimization of the dual functional $\\psi(u)$. The results consolidate classical findings (Rabinowitz) and duality-based approaches (Clarke and Ekeland), and extend them with subharmonic counts (Michalek & Tarantello) under symmetry and pinching arguments. The analysis provides a rigorous foundation for periodic and subharmonic dynamics in autonomous Hamiltonian systems with $(A_{ abla\infty},B_{ abla\infty})$-subquadratic growth conditions, with explicit criteria for nontriviality and minimality of the solutions. This framework offers a powerful toolset for studying long-time behavior and multiplicity of orbits in convex Hamiltonian physics and related mathematical models.
Abstract
Tiger conservation necessitates the strategic deployment of multifaceted initiatives encompassing the preservation of ecological habitats, anti-poaching measures, and community involvement for sustainable growth in the tiger population. With the advent of artificial intelligence, tiger surveillance can be automated using object detection. In this paper, an accurate illumination invariant framework is proposed based on EnlightenGAN and YOLOv8 for tiger detection. The fine-tuned YOLOv8 model achieves a mAP score of 61% without illumination enhancement. The illumination enhancement improves the mAP by 0.7%. The approaches elevate the state-of-the-art performance on the ATRW dataset by approximately 6% to 7%.
