Motion Estimation in Flotation Froth Images Based on Edge Detection and Mutual Information
MetadataShow full item record
Motion Estimation is an important research field with many applications including remote sensing, surveillance, navigation, process control, and image compression. One of the popular methods used for motion estimation is intensity based block motion estimation. A common drawback with intensity based block motion estimation is that the velocity of blocks located at the boundaries of moving objects is not estimated accurately. This is mainly because blocks do not fully surround moving objects. In this paper, a novel method for the estimation of motion using edge matching is presented. The Canny edge detector is used to create the edges. The image is divided into nonoverlapping rectangular blocks. The best match to the current block is searched for in the previous of frame of the video within a search area for the location of the current block. Mutual information (MI) with a bin size of two is used as the matching criterion. Experimental results from test image sequences show that the proposed edge-based motion estimation technique improves the motion estimation accuracy in terms of the peak signal-to-noise ratios of reconstructed frames.
Showing items related by title, author, creator and subject.
Truong, Huy S. (1999)Still image and image sequence compression plays an important role in the development of digital television. Although various still image and image sequence compression algorithms have already been developed, it is very ...
Arandjelovic, O.; Pham, DucSon; Venkatesh, S. (2016)Our aim is to estimate the perspective-effected geometric distortion of a scene from a video feed. In contrast to most related previous work, in this task we are constrained to use low-level spatiotemporally local motion ...
Arandjelovic, O.; Pham, DucSon; Venkatesh, S. (2015)Our aim is to estimate the perspective-effected geometric distortion of a scene from a video feed. In contrast to all previous work we wish to achieve this using from low-level, spatio-temporally local motion features ...