Authors : Israa Hadi Ali and Qasim Jaleel
Abstract: One of the main problems with video processing is the extraction of feature. Interest points are one of the feature types in the video. There are a number of algorithms that depend on the interest points in the stitching process. The aim of this study is to compare three methods to extract interest points for video hidden target recovery. Hidden target makes the target non-clear for a reason such as security cases or to save the privacy of the target in the video. The proposed method is implemented in two stages. Firstly, determining the hidden region in the target and create a large window around the desired target to track and discover the target parts that appear. Secondly, the stage is the stitching stage of the parts using one of the best three methods namely Harris corner detection, SURF and FREAK. After selecting the best image resulting from those methods and comparing the results. Experiments have demonstrated the efficiency and robustness of the Harris algorithm compared to the SURF and the FREAK in the proposed method.
Israa Hadi Ali and Qasim Jaleel, 2018. Determining the Best Method to Extract Interest Point for Video Hidden Object Recovery. Journal of Engineering and Applied Sciences, 13: 2413-2419.