Spherical-Orthogonal-Symmetric Haar Wavelet for Driver's Visual Detection

This work proposes a spherical-orthogonal-symmetric Haar wavelet to decompose and reconstruct spherical iris signals to obtain stronger geometric features of iris surface. It compares its feature extraction abilities of spherical harmonics, semi-orthogonal and nearly orthogonal spherical Haar wavelet. The developed spherical-orthogonal-symmetric Haar wavelet with a convolutional neural network is also proposed for drivers' iris recognition. It can effectively capture the local fine features of iris spherical surface, and has stronger ability of iris recognition than semi-orthogonal or nearly orthogonal spherical Haar wavelet bases.

Automated summary

This work proposes a new spherical-orthogonal-symmetric Haar wavelet for decomposing and reconstructing spherical iris signals, aiming to obtain stronger geometric features of iris surface. The proposed wavelet is compared with spherical harmonics, semi-orthogonal, and nearly orthogonal spherical Haar wavelets in terms of feature extraction abilities. The results show that the developed spherical-orthogonal-symmetric Haar wavelet, when combined with a convolutional neural network, has a stronger iris recognition ability compared to the other wavelet bases.