Iris segmentation is a critical step in the entire iris recognition procedure. Most of the state-of-the-art iris segmentation algorithms are based on edge information. However, a large number of noisy edge points detected by a normal edge-based detector in an image with specular reflection or other obstacles will mislead the pupillary boundary and limbus boundary localization. Under non-ideal conditions, existing segmentation technique which based on local operations cannot find true iris boundary. As a result, system failure will occur. To solve this problem, a new algorithm which is based on feature extractions, classification using vgg16 is introduced. The execution has performed on the MATLAB software and performance results carried out in terms of accuracy, precision, sensitivity. Proposed scheme has been tested on three well-known iris databases CASIA, MMU and UBIRIS-V2, and is shown promising results with best accuracy rate of 90.91%.