Estimation of tiger population counts plays a pivotal role in wildlife conservation and ecological management. Traditional tiger census methods, such as pugmark tracking and/or traditional camera-trap image assessment, are time-consuming and prone to human-errors. This paper incorporates an Automated Tiger Census System using Deep Pattern Recognition of Stripe Biometrics, where distinct tiger identification is done on the basis of distinct stripe patterns. The proposed system comprises image preprocessing, deep feature extraction, pattern identification, and capture-recapture population assessment. Theoretical models of pattern identification and match assessment have been formulated, and the end-to-end algorithm has been outlined. Experimental assessments have been done on the camera-trap image data, and high accuracy and efficacy have been achieved under practical forest settings. The system is scalable, non-invasive, and viable at all the levels for tiger census.