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.
The paper presents a method to reconstruct a controllable 3D head and face model for a specific person. The 3D grid-muscle head and face model of the certain person is created after selecting key feature points from a front face photo and a left side face photo of the person, deforming the grid points and muscle vectors of a basic model. Then after texture mapping, we reconstruct the special person's...
The vehicle license plate (VLP) location is the key technique of the vehicle license plate recognition system. The region of license plate has rich edge information and fixed color feature. The location method fusing gray edge and color feature is more accurate and easier than the method only in gray space or in color space. The experiments on 200 Chinese automobile RGB real color images taken from...
This paper presents a novel approach of feature selection based on analysis of covariance matrix of training patterns, a correlation-based feature selection method is put forward. An objective measure is proposed and defined. It is shown that for a given set of features, a subset of features that has the highest sum of the correlation coefficients has the tendency to be reduced, if it meets the requirement...
Canonical Correlation Analysis (CCA) is a well-known method for feature extraction and dimension reduction. CCA can simultaneously deal with two sets of data. It makes CCA can be used for feature level fusion. But it suffers the Small Sample Size (SSS) problem. In this paper, a new optimization criterion is presented for overcoming the SSS problem. The optimization problem can be solved analytically...
With the popularity of the network and development of multimedia technology, the traditional information retrieval techniques do not meet the users?? demand. Recently, the content-based image retrieval has become the hot topic and the techniques of content-based image retrieval have been achieved great development. In this paper, the basic components of content-based image retrieval system are introduced...
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...
As complex and varied concrete structures and their disease characteristics, it is difficult to extract a stable identification feature. Interpretation of the ground penetrating radar(GPR) scanned image is mostly based on experts experiences. Thus we designed an efficient concrete bridge disease identification system based on sample database(CBDI). The CBDI is based on principal component analysis...
Ant colony optimization (ACO) algorithms have been applied successfully to combinatorial optimization problems. Rough set theory offers a viable approach for feature selection from data sets. In this paper, the basic concepts of rough set theory and ant colony optimization are introduced, and the role of the basic constructs of rough set approach in feature selection, namely attribute reduction is...
The classification for similar features classes is quite difficult task in many existing pattern-recognition systems. When the amount of samples is insufficient, neural networking training is hard. The dimension reduction, classification, clustering etc serial steps in recognition process takes such much time that the practical recognizing application is ease to meet the real time requirement. The...
This paper presents a method for expressing color features of images through color hue histograms, according to the distribution characteristics of image color space. In this method, the spatial features of colors are described in terms of three independent components of the HSV color space, with rotation, scaling and translation invariant features. Tests have proved that favorable retrieval effects...
Gender classification is one of the most challenging problems in the field of pattern recognition. The pixel-based gray image recognition method is quite sensitive to illumination variation and has high dimensions for computation. PCA-based image feature recognition algorithm can reduce the image dimension, but it is only on the basis of optimal entropy to choose face features which neglects the different...
In this paper segmentation on both color and spatial space is implemented to compare their influence to the performance of content-based image retrieval (CBIR) system. Firstly, images are converted from RGB color space to HSV space. And then cumulative histograms on H component are extracted from four segmented blocks as one color feature of the image. At the same time, the HSV color space is further...
Currently, surgery is the most effective and common way to treat cataract, one of the leading causes for blindness worldwide. Of all surgical methods, phacoemulsifieation is the most popular. During the operation, surgeons have to evaluate the hardness degree of the cataractous lens by themselves. To make the evaluation intelligent, a machine-aided classification method for cataractous lens is proposed...
The new method of active fortify organizing a supervised area based on color and geometric feature is proposed in this paper. By the basis of color feature in the real supervised scene, in the first instance extracting several regions of interest (ROI) with noise, then matching with geometric shape by Fourier descriptors in the database, sequentially achieving automatic organizing a supervised area...
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...
This paper presents a new method based on bootstrap technique for modulation recognition under the low SNR and multi-path conditions. The bootstrap method is performed when only few samples are available at the receiver. A new neural network is also proposed to select effective features and classifies signals. Extensive simulations show that the new method has preferable performance in low SNR and...
The current practice of recognition spectra manually is no longer applicable to a large extent. This work is particularly focused on helping astronomers finding their interesting celestial objects. In this paper an efficient hierarchical clustering data mining method based on principal component analysis (PCA) is proposed. Massive stellar spectral data are clustered by improved hierarchical clustering...
Since face detection use a single classification is very time-consuming and inefficient. To solve this problem, we propose to establish two classifiers which are based on facial features and human eye features using the Adboost algorithm. First, use the facial features classifier detected out a rough location of the face. Then give out the accurate location of the face accord to the distance and the...
The objective of this study was to develop a computerized empty glass bottle inspection method in an attempt to replace manual inspection. The inspection system structure based on machine vision was illustrated in the article. Morphologic methods and wavelet transform were used to extract features of the bottle body and the finish from images. The fuzzy support vector machine neural network was adopted...
According to the global and local features of Chinese manual alphabet images, Fourier descriptor and other multi-features is introduced for the vision-based multi-features classifier of Chinese sign language recognition. At first, extracting features of letter images is done, then classification method of SVMs for recognition is brought into use. Experimentation with 30 groups of the Chinese manual...
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.