Applications Of Interval Algorithm In Computer Vision

Computer vision is a comprehensive subject that plays a very important role in solving various practical problems.Thus it is of great significance to explore and research in the field of computer vision.As two important research fields in computer vision,camera calibration and target tracking have high requirements for the model accuracy and computational accuracy of their algorithms.In this paper,the Rump interval algorithm is used to study the error range of the five-point algorithm and eight-point algorithm for camera calibration and the target tracking algorithm,so as to realize the error control of the calculation results of the five-point algorithm and eight-point algorithm for camera calibration and the target tracking algorithm.The main research contents are as follows:(1)Design a credible verification algorithm for computer vision camera calibration.Based on the five-point algorithm and eight-point algorithm for camera calibration,the Rump interval operation is used to calculate the credible error bound of the solution of the linear equation of the essential matrix and the basic matrix.The algorithm is guaranteed to be within the error bound,there are exact solutions of the linear equations of the essential matrix and the fundamental matrix.And using the credible error bound,the credible error bound of the essential matrix and the fundamental matrix can be determined,and within this error bound,the precise essential matrix and fundamental matrix of the five-point algorithm and eight-point algorithm for camera calibration can be obtained.Finally,according to the calculated essential matrix and fundamental matrix,the three-dimensional reconstruction of the space point is realized by using the two-dimensional plane coordinate information.(2)Design a credible estimation algorithm for computer vision target tracking.Based on the target tracking model of the spectral filter,the target tracking model is improved by using the Rump interval algorithm.The linear equation for solving the coefficients of the tracking model is converted into the form of interval operation,the confidence interval of the tracking model coefficient of the spectrum filter is obtained,the interval coefficient is applied to the target tracking algorithm to estimate the coordinate range of the tracked target and realize the credible estimation of computer vision target tracking.When the tracked target is lost,color recognition is used to preprocess the background of each frame to estimate the region range of the target,and then we use the target tracking algorithm to recapture the target.