An improved estimation of motion vectors of feature points is proposed for tracking moving objects of dynamic image sequence. Feature points are firstly extracted by the improved minimum intensity change （MIC） algor...An improved estimation of motion vectors of feature points is proposed for tracking moving objects of dynamic image sequence. Feature points are firstly extracted by the improved minimum intensity change （MIC） algorithm. The matching points of these feature points are then determined by adaptive rood pattern searching. Based on the random sample consensus （RANSAC） method, the background motion is finally compensated by the parameters of an affine transform of the background motion. With reasonable morphological filtering, the moving objects are completely extracted from the background, and then tracked accurately. Experimental results show that the improved method is successful on the motion background compensation and offers great promise in tracking moving objects of the dynamic image sequence.展开更多
Speedometer identification has been researched for many years.The common approaches to that problem are usually based on image subtraction,which does not adapt to image offsets caused by camera vibration.To cope with ...Speedometer identification has been researched for many years.The common approaches to that problem are usually based on image subtraction,which does not adapt to image offsets caused by camera vibration.To cope with the rapidity,robust and accurate requirements of this kind of work in dynamic scene,a fast speedometer identification algorithm is proposed,it utilizes phase correlation method based on regional entire template translation to estimate the offset between images.In order to effectively reduce unnecessary computation and false detection rate,an improved linear Hough transform method with two optimization strategies is presented for pointer line detection.Based on VC＋＋ 6.0 software platform with OpenCV library,the algorithm performance under experiments has shown that it celerity and precision.展开更多
In this paper the idea of Intelligent Scissors is adopted for contourtracking in dynamic image sequence. Tracking contour of human can therefore be converted to trackingseed points in images by making use of the prope...In this paper the idea of Intelligent Scissors is adopted for contourtracking in dynamic image sequence. Tracking contour of human can therefore be converted to trackingseed points in images by making use of the properties of the optimal path (Intelligent Edge). Themain advantage of the approach is that it can handle correctly occlusions that occur frequently whenhuman is moving. Non-Uniform Rational B-Spline (NURBS) is used to represent parametrically thecontour that one wants to track. In order to track robustly the contour in images, similarity andcompatibility measurements of the edge are computed as the weighting functions of optimal estimator.To reduce dramatically the computational load, an efficient method for extracting the regioninterested is proposed. Experiments show that the approach works robustly for sequences withfrequent occlusions.展开更多
文摘An improved estimation of motion vectors of feature points is proposed for tracking moving objects of dynamic image sequence. Feature points are firstly extracted by the improved minimum intensity change （MIC） algorithm. The matching points of these feature points are then determined by adaptive rood pattern searching. Based on the random sample consensus （RANSAC） method, the background motion is finally compensated by the parameters of an affine transform of the background motion. With reasonable morphological filtering, the moving objects are completely extracted from the background, and then tracked accurately. Experimental results show that the improved method is successful on the motion background compensation and offers great promise in tracking moving objects of the dynamic image sequence.
基金Supported by the National Natural Science Foundation of China (61004139)Beijing Municipal Natural Science Foundation(4101001)2008 Yangtze Fund Scholar and Innovative Research Team Development Schemes of Ministry of Education
文摘Speedometer identification has been researched for many years.The common approaches to that problem are usually based on image subtraction,which does not adapt to image offsets caused by camera vibration.To cope with the rapidity,robust and accurate requirements of this kind of work in dynamic scene,a fast speedometer identification algorithm is proposed,it utilizes phase correlation method based on regional entire template translation to estimate the offset between images.In order to effectively reduce unnecessary computation and false detection rate,an improved linear Hough transform method with two optimization strategies is presented for pointer line detection.Based on VC＋＋ 6.0 software platform with OpenCV library,the algorithm performance under experiments has shown that it celerity and precision.
文摘In this paper the idea of Intelligent Scissors is adopted for contourtracking in dynamic image sequence. Tracking contour of human can therefore be converted to trackingseed points in images by making use of the properties of the optimal path (Intelligent Edge). Themain advantage of the approach is that it can handle correctly occlusions that occur frequently whenhuman is moving. Non-Uniform Rational B-Spline (NURBS) is used to represent parametrically thecontour that one wants to track. In order to track robustly the contour in images, similarity andcompatibility measurements of the edge are computed as the weighting functions of optimal estimator.To reduce dramatically the computational load, an efficient method for extracting the regioninterested is proposed. Experiments show that the approach works robustly for sequences withfrequent occlusions.
