-5171ffc0182b6af927f-3fe7 20221014050531275 beie:beie director@blueeyesintelligence.org WEB-FORM International Journal of Innovative Technology and Exploring Engineering IJITEE 22783075 10.35940/ijitee https://www.ijitee.org/ 11 30 2022 11 12 Department of Computer Science & Engineering, Vellore Institute of Technology, Chennai (Tamil Nadu), India. Suhas Reddy P Bhargavi Rao Department of Computer Science & Engineering, Vellore Institute of Technology, Chennai (Tamil Nadu), India. Jayanth Anala Department of Computer Science & Engineering, Vellore Institute of Technology, Chennai (Tamil Nadu), India. Megha Dangayach Department of Computer Science & Engineering, Vellore Institute of Technology, Chennai (Tamil Nadu), India. The image Recognition system is a vital problem in the field of computer vision because it must be precise, successful in the desired goal, strong, healthy, and self-loading. The following are the most critical essential phases in image alignment/registration: feature matching, feature detection, derivation of transformation function based on related features in pictures, and reconstruction of images based on generated transformation function. In many applications, the goal of computer vision is to create an ideal and accurate image, which is dependent on optimal feature matching and detection. This paper's inquiry summarizes the similarity among five alternative approaches for robust features/interest points (or landmarks) detector and picture identification. This research also focuses on the extraction of unique features from photos that may be utilized to conduct effective matching of diverse perspectives of the images/objects/scenes. 11 30 2022 1 4 CC BY-NC-ND 4.0 10.35940/BEIESP.CrossMarkPolicy www.ijitee.org true 10.35940/ijitee.L9308.11111222 https://www.ijitee.org/portfolio-item/l930811111222/ Associate Professor, Department of Computer Science & Engineering, Jabalpur Engineering College, Jabalpur (M.P), India. Dr. A. Hemlata S. Saranya Student, Department of Computer Science & Engineering, Jabalpur Engineering College, Jabalpur (M.P), India. Dr. Mahesh Motwani Professor, Department of Computer Science & Engineering, University Institute of Technology, RGPV, Bhopal (M.P), India. The competitive examinations conducted in multiple shifts have different question papers for different shifts. The assigned difficulty index of the entire question papers are kept same for the entire shift, however the calculated difficulty index may vary. The major challenge faced in the conduction of the multiple shift examination is the generation of the common merit list. This research paper concentrates on the study of various merit list generation techniques used for multi shift examinations. It also proposes a new merit list generation method taking in to consideration the difficulty index of the question paper. This paper also compares various merit list generation techniques such as actual score method, normalized score method, percentile score method and the proposed difficulty index based score method. The merit list generated by percentile score method gives equal number of selection from each shift and does not consider the difficulty index of the question paper. Whereas in normalization method, the score is normalized by considering the mean, standard deviation of the score of each shift and of all shifts. This method also equalises the selection count. However, the proposed techniques take in to account the difficulty index of the question paper as well, which may vary the selection count in each shift. This assures that the deserving and the eligible candidate does not get affected due to the difficulty level variation of question paper. 11 30 2022 5 11 CC BY-NC-ND 4.0 10.35940/BEIESP.CrossMarkPolicy www.ijitee.org true 10.35940/ijitee.K9309.11111222 https://www.ijitee.org/portfolio-item/k930910111122/