Liveness detection is the most important field in the field of biometrics. The development of biometric-based liveness detection algorithms provides a key role in the success of security applications. Liveness detection helps the system to identify whether the user is real or fake. Proposed Algorithm Provides novel approach of liveness detection based on the fusion of discrete cosine transform (DCT) and Zernike moments. Conjunctival vasculature which is a region of interest is a low-frequency image area. This makes DCT transform based feature extraction important for low-frequency feature analysis. Off angle iris database generation gets affected by the lightening effect. To overcome this drawback Zernike moments-based feature extraction is proposed. A combination of DCT and Zernike makes an innovative feature for an off- angle iris. Statistical features are used to test the accuracy of the system. It can be shown that the fusion approach performs better as compared with individual methods. The algorithm is tested by using Extreme Learning Machine (ELM). Equal error rate (EER) is calculated to test the robustness of the liveness detection system.