A novel dense descriptor based on structure tensor voting for multi-modal image matching

• DOI: 10.1016/j.cja.2020.02.002
• Publication Date: 2020-03-12T20:08:20Z
• Authors: Jiazhen LU; Maoqing HU; Jing DONG; Songlai HAN; Ang SU
• Keywords: Similarity measure

Automatic and robust matching of multi-modal images can be challenging owing to the nonlinear intensity differences caused by radiometric variations among these images. To address this problem, a novel dense feature descriptor improved similarity measure are proposed for enhancing performance. The is built on voting scheme structure tensor that effectively capture geometric structural properties It not only illumination contrast invariant but also against degradation significant noise. Further, adapt reversal orientation inversion between enable development practical template-matching algorithm We verify with broad range including optical, infrared, Synthetic Aperture Radar (SAR), digital surface model, map data. experimental results confirm its superiority state-of-the-art methods.