Energy flow: image correspondence approximation for motion analysis

Abstract

Abstract. We propose a correspondence approximation approach between temporally adjacent frames for motion analysis. First, energy map is established to represent image spatial features on multiple scales using Gaussian convolution. On this basis, energy flow at each layer is estimated using Gauss-Seidel iteration according to the energy invariance constraint. More specifically, at the core of energy invariance constraint is "energy conservation law" assuming that the spatial energy distribution of an image does not change significantly with time. Finally, energy flow field at different layers is reconstructed by considering different smoothness degrees. Due to the multiresolution origin and energy-based implementation, our algorithm is able to quickly address correspondence searching issues in spite of background noise or illumination variation. We apply our correspondence approximation method to motion analysis, and experimental results demonstrate its applicability. © The Authors. Published by SPIE under a Creative Commons Attribution 3.0 Unported License. Distribution or reproduction of this work in whole or in part requires full attribution of the original publication, including its DOI.

Switch Matrix Algorithm for Series Lithium Battery Pack Equilibrium Based on Derived Acceleration Information Gauss‐Seidel

“…Algorithm Example. According to the component parameters in the experiment, the energy transfer matrix of the switch array can be described by (16), and the final battery energy difference is shown in (17). The information is collected and transmitted to the single-chip microcomputer which solves the conduction time matrix of the equalization module switch group, thereby controlling the drive circuit to send the corresponding PWM signal to control the switch tube.…”

“…The Gauss-Seidel method is an iterative method in numerical linear algebra that can be used to find the approximate solution for a group of linear equations [14]. This method is also widely used in engineering practice, such as modulated filter bank [15], employ cloud-based computation offloading [16], motion analysis [17], etc. In [18], the algorithm based on outlier detection was proposed to solve the problem that present cell-balancing algorithms cannot identify the unbalanced cells in the battery pack.…”

Discussion on Robot Embedded Positioning System Based on Image Correspondence Algorithm

“…The given distance function D: I(d) 2 →R, and two images f, sI(d), geometric transformation group G:{g：R d R d } and mapping function group ψ, the image correspondence process is to find the transformation function g to shorten the distance between f and s (image corrected and yet to be matched), as shown below (1).…”

“…(*) denotes complex conjugate, (2) the inverse Fourier transform is a delta function with translation parameters as the centre. The inverse Fourier transform (IFFT) of equation (2) produces phase correlation spectrum, and the corresponding point to the maximum value is the parameters of translation for estimation.…”

“…The typical application of image correspondence in computer vision is the binocular vision stereo correspondence, monocular image sequence optical flow, but also can be used for remote sensing, image correspondence in medical image processing, video compression, image super resolution image motion estimation and other fields, it can also be applied to many other areas: scene-together, virtual scene reconstruction, information fusion, pattern recognition, image correspondence, etc. The usual matching problem can also be extended to threeview and be generalized to the broader N-view example [2][3][4].…”