The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
Pixel point gradient features represent much of the intrinsic structures of an image and can be used to the description of machine vision object. By ICA technique, pixel gradient data can be projected from a high-dimensional space to a lower-dimensional space, which reduce the redundancy with no image segment based on threshold. A method of visual object matching based on gradient ICA feature is provided...
A feature point detection algorithm is presented based on the scale-space theory. The algorithm overcome the drawback that a typical single-scale Harris detector usually leads to either missing significant corner points or detecting false corner points due to noise and position displacement. Original matching is solved by similarity of the image gradient module and argument, and the crude matching...
Some target tracking occasions often requires to tracking a kind of target, such as human face, automobile and so on. A specific target tracking algorithm based on support vector machine (SVM) and AdaBoost is proposed. Moreover, the characteristic data of SVM is a critical factor to success to detecting target. The method selects part of Harr wavelet characters by AdaBoost as input data of SVM training...
An important algorithm SIFT, which has been successfully applied in image matching, is employed in face recognition. Firstly, the main region of a face is detected from background images by AdaBoost. Secondly, face features are extracted by using SIFT. Then, face recognition is conducted by the comparing real extracted features with training sets. Experiment shows that, in the ORAL face DB, this scheme...
A novel approach for scene matching real-time image geometric rectification is proposed. Based on the analysis of the reason caused geometric distortion in the process of real-time image capturing, two coordinates system are given, the real-time image distortion classification are analyzed, and the distortion mathematic models are established. Then, the geometric rectification of the real-time image...
Fingerprint feature detection is one of the key techniques of automatic fingerprint identification. There are some problems of rotation and shift in the present fingerprint feature detection. This paper has given an algorithm of bifurcation detection based on neural network template matching. Correlative matching is to calculate the correlative value between template and target image according to...
Video stabilization is a technology to remove the dithering between frames in a video sequence by motion estimation, motion filter and motion compensation. In this paper, we focus on the motion estimation (ME) step which is known as the most important part for a successful video stabilization system because the latter processing heavily relies on it. However, traditional ME methods mainly hypothesize...
It is very important to detect singular points (core and delta) accurately and reliably for classification and matching of fingerprint. In this paper, an improved method for singularity detection in fingerprint images, which based on continuous orientation field, is proposed to improve accuracy of the position and reliability of the singularity. Firstly, the blocks which may contain singularities...
Fractal image encoding with full search typically requires a very long runtime, which is essentially spent on searching for the best-matched block to an input range block in a large domain pool. This paper thus proposed an effective method to improve the drawback, which is mainly based on gray value moment features of normalized block and related inequality is presented by the authors. During the...
3D reconstruction has a long history, few workable reconstruction systems are available although many magnificent theories and algorithms are reported in literature. In this paper, a 3D reconstruction system for indoor scenes with a rotating platform is reported, including its key components such as system calibration, feature matching and 3D reconstruction. Experimental results show that this system...
In this paper, we present a novel method to reconstruct the large scale scenes from multiple calibrated images. It first generates a quasi-dense 3D point cloud of the scene by matching key points across images. Then it builds a tetrahedral decomposition of space by computing the 3D Delaunay triangulation of the 3D point set. Finally, a triangular mesh of the scene is extracted by labeling Delaunay...
Among the solutions for stereo matching which is regarded as one fundamental topic in computer vision, methods based on regions can overcome the drawbacks of methods using points and lines as the features since regions can more exactly reflect the inherent properties of images processed. In this paper, we propose an efficient method to extract the local invariant region, in which the affine moment...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.