Search

This paper focuses on how to represent the testing face images for multi-feature face recognition. The choice of feature is critical for face recognition. The different features of the sample contribute differently to face recognition. The joint similar and specific learning (JSSL) has been effectively applied in multi-feature face recognition. In the JSSL, although the representation coefficient is divided into the similar coefficient and the specific coefficient, there is the disadvantage that the training images cannot represent the testing images well, because there are probable expressions, illuminations and disguises in the testing images. We think that the intra-class variations of one person can be linearly represented by those of other people. In order to solve well the disadvantage of JSSL, in the paper, we extend JSSL and propose the extended joint similar and specific learning (EJSSL) for multi-feature face recognition. EJSSL constructs the intra-class variant dictionary to represent the probable variation between the training images and the testing images. EJSSL uses the training images and the intra-class variant dictionary to effectively represent the testing images. The proposed EJSSL method is perfectly experimented on some available face databases, and its performance is superior to many current face recognition methods.

The infrared image is formed by the thermal infrared rays of the targets and environment. As the targets radiate more heat, we can see they have larger grey-scale, locating in the topend of histogram. According to this characteristic of infrared images, a new method for infrared image segmentation based on Otsu and genetic algorithm is proposed in this paper. First, covert the infrared image to grayscale image; Second, detect the image edge, and get the best threshold for image segmentation by Otsu and GA (Otsu-GA); Finally, Segment the image by the threshold, and contrast the results of two peaks and iteration. Experimental results show that the method effectively improved the quality of image segmentation.

A new color image segmentation algorithm based on histogram, FCM clustering, and region merging is proposed in this paper. First, the RGB space is transformed to HSV space, and the image is divided into non-singular points and singular points in accordance with the saturation. Second, characteristics of the image pixel are mapped to the one-dimensional histogram, we can determine the number of the cluster and the initial cluster center thought peaks selection algorithm, non-singular points and singular points are separately clustering by FCM,. Finally, we merger regions by image spatial information to eliminate the scattered small area after clustering, which overcomes the over-segmentation problem in FCM, and increases the ability of anti-noise. Experimental results show that this method not only can make the partition consistent with the human visual psychology, but also overcome the singularity of HSV space, and significantly reduce computational complexity and greatly improve the speed of the algorithm, realize automatically dividing images without manual intervention.

VFD (Vacuum Fluorescent Display) dot-matrix display module GU160X32-800B can display the character, Chinese and the figure. Its displaying dot-matrix has 160*32 dots and has some advantages, such as convenient application, higher brightness and being regulated. In this paper, the basic theory of GU160X32-800B is introduced and the methods of designing its hardware and programming its software.

In the design of time calibrator, time calibrator can be achieved on a higher stability, because of the highest stability of constant-temperature crystal oscillator highest stability, but as time goes on, the stability and the frequency stability of constant-temperature crystal oscillator will gradually decline, it will affect the measurement accuracy. In light of the above issues, we purpose a new method to obtain higher stability for it long time work. First, generate the frequency signal by voltage-controlled constant-temperature crystal oscillator; Second, get the signal 1S from the time GPS module; Third, comparison the above tow signals, thus, a error signal is generated; Finally, converter the error signal by digital-to-analog module, and then, use it to control the voltage-controlled constant-temperature crystal oscillator to achieve the purpose of automatic tame.

In order to sharpen image details and reducing noise, based on the multi-analysis wavelet threshold denoising method, a Labeling-based block-matching and wavelet transform filtering method combine hard and soft threshold denoising approaches (BWHS) is proposed in this paper. First, we estimate the noise variance of image. Second compute the matching blocks, and construct the 3D data array of those similar blocks, the high and low frequency sub-bands denoised by the best soft threshold, hard threshold that result from the iterative calculation of noise variance respectively, Finally, sharpen image details using DC coefficients of LL frequency sub-bands. Simulation results show that the algorithm can preserve and sharpen image details and effectively attenuate noise. Moreover, it has better performance than the traditional soft threshold, hard threshold, median and mean denoising methods.

We propose a connected component labeling algorithm using line labeling and region growing method (LRGM) in this paper. First, we analyse the basic characteristic of current labeling algorithms, and set the scan order of LRGM from left to right, top to bottom, to assign a label to all connected components. Second, we eliminate label conflict by region growing method, because a large number of K label arises of which many are equivalent. Finally, we optimize the search and judgment criterion of LRGM, to make the new algorithm is independent of connected components shape, and the search time is much less than it before costed. Experimenting on various types of document images (pictures, newspapers, etc.), we find that our method outperforms the other sequential methods issued in publication. It greatly increases the run efficient, and it is very useful for real-time and large images processing.

In this paper we devote to study some feature selection of an information system in which redundant or insignificant attributes in data sets can be eliminated. An approach of importance gain function is suggested to evaluate the global average information gain associated with a subset of features. A heuristic algorithm on iterative criterion of feature selection on the significance of attributes is proposed to get the least reduction of attribute set in knowledge discovery. The feasibility of feature selection proposed here is validated by some of examples.