Based on Sparse Coding, a Novel Key-Point Detector PROJECT TITLE : A Novel Key-Point Detector Based on Sparse Coding ABSTRACT: Harris corner, MSER, SIFT and SURF are among the most common hand-crafted key-point detectors for detecting corners, blobs, or junctions in an image. Detectors' inflexibility might be attributed to the pre-designed character of these detectors un many circumstances. Non-uniform lighting also has a significant impact on these detectors' performance. There are some earlier efforts that have dealt with one of the two issues, but there is currently no effective approach for solving both at the same time. Here, we present a unique Sparse Coding Key-point detector (SCK) based on affine intensity change that is completely invariant to any specific structure. To find a crucial point in a picture, the detector uses a complexity measurement derived from the block surrounding the spot in question. For comparison and selection when the maximum number of key-points is limited, a strength measure is provided. It is shown in this work that the suggested detector has desired properties. On three publicly available datasets, the suggested detector demonstrates considerable performance in terms of repeatability and matching score. Did you like this research project? To get this research project Guidelines, Training and Code... Click Here facebook twitter google+ linkedin stumble pinterest Multiple Description Image Coding Using Polyphase Down-Sampling A Novel Fractional-Order Variational Model for Images with Extremely Low Light Based on Retinex