Mianshu Chen

A Dynamic Blood Information Management System Based on RFID

As a patient and recipient of a blood transfusion, it is important for him or her to receive the safest blood possible. Information about the donated blood should be under the track to guarantee the quality of the blood source. In this paper, we present a RFID-based blood information management system that aims at ensuring the quality of the blood and increasing the efficiency of operation management. In this system, the fingerprint sensor is adopted to enable the process of identifying blood donor more reliable and credible and RFID tag is used to make the management more convenient. In addition, GPRS is applied in this system so that real-time data can be transmitted between the bloodmobile and blood center through wireless Internet

Image retrieval based on multi-feature similarity score fusion using genetic algorithm

This paper proposes an image retrieval method based on multi-feature similarity score fusion using genetic algorithm. Single feature describes image content only from one point of view, which has a certain one-sided. Fusing multi-feature similarity score is expected to improve the systems retrieval performance. In this paper, the retrieval results from color feature and texture feature are analyzed, and the method of fusing multi-feature similarity score is described. For the purpose of assigning the fusion weights of multi-feature similarity scores reasonably, the genetic algorithm is applied. For comparison, other three methods are implemented. They are image retrieval based on color feature, texture feature and fusion of color-texture feature similarity score with equal weights. The experimental results show that the proposed method is superior to other methods.

Multi-dimensional vector matrix theory and its application in colour image coding

Abstract Based on matrix theory, the concept of two-dimensional matrix is extended to fit multi-dimensional matrix. In order to reduce the space redundancy, time redundancy and colour redundancy better, we propose a multi-dimensional vector matrix model. Some related properties are discussed, and Walsh orthogonal square matrix is given in multi-dimensional vector matrix space, then its application in colour image compression and coding is introduced. By means of multi-dimensional vector matrix model, colour image data can be expressed and processed in a unified mathematical model, and better compression results are obtained. On the one hand, it inherits high efficiency of classical matrix transform in the aspect of wiping off redundancy of colour space; on the other hand, it overcomes the restriction of traditional two-dimensional matrix multiplication. By multi-dimensional Walsh orthogonal transform and traditional discrete cosine transform, favourable image compression effect can be obtained.

Hardware Design for Mpeg-7 Compact Color Descriptor Based on Sub-Block

In this paper we propose a subblock-based compact color descriptor and present a hardware design for this descriptor, which contains four main sections: partition, quantification, histogram computation, and Haar transform. It can be used for video content filtering, image matching, sketch-based matching and browsing application. Several design solutions of the descriptor extraction are given

6D vector orthogonal transformation and its application in multiview video coding

Abstract Multiview video (MVV) is multiple video sequences that integrated different viewpoints data of the same three-dimensional (3D) scene. Each viewpoint data are taken from the ordinary video camera. Thus, the data are very large for the MVV. So compression is necessary in order to store and transmit effectively. Based on the theory of multi-dimensional vector matrix (MDVM),2 we propose a six-dimensional (6D) vector orthogonal transform nuclear matrix, and prove its orthogonality and energy concentration. We apply the theory to multiview video coding (MVC). This transformation is based on discrete cosine transform (DCT), which has the optimal performance for video data. We represent MVV data with a multi-dimensional (MD) mathematical model. The chosen MVV is earlier eight frames in YUV format from two viewpoints. We divide the Y, U and V components into cubes respectively, and combine the two views data into one cube, on which the transformation is conducted. Good results are obtained in terms of energy concentration. This paper provides a new method for handling MVV, and prepare for the next quantisation and coding.

Multi-dimension Multimedia Retrieval Model Implementation Based on XML Database

Derived from the thought of data warehouse, data cube and xml, this paper presents a new database structure model which organizes the multimedia data in a multi-dimensional data cube based on XML Database. In this data cube of XML, clustered data are stored in instance table. A leading data corresponding are stored in dimension table. The relational model is helpful to construct data model, but it lacks flexibility, now the new data model can complement the defect of relational model. When querying, a leading data is gained from dimension table of XML¿then receiving the multimedia data through XQuery.

Multimedia database retrieval based on data cube

Derived from the thought of data warehouse and data cube, this paper presents a new database structure model which organizes the multimedia data in a multi-dimensional data cube. Clustered data are stored in instance table in the data cube. A corresponding leading data are stored in dimension table. When querying, a leading data is gained from dimension table firstly, followed by a k-nearest neighbor query from the corresponding category, the retrieval results are returned. Because of not in need of searching all the records in database, the time cost is low.

Video Compression Based on Adaptive Multidimensional Vector Matrix DCT

The traditional multi-dimensional vector matrix discrete cosine transform compression adopts video frame segmentation of fixed block size. In fact, the texture complexities of image in different regions are usually very different, so transformation adaptively with proper block size, not only further improves the compression performance, but also gets better improvement in subject evaluation. A three-dimensional variable block size video images segmentation method is proposed by using gradient edge detection operator as image activity criterion, and implement variable block size multi-dimensional vector matrix transform coding, experimental results show that the proposed method gets better performance than H. 264, and the quality of the reconstructed images is improved in subject evaluation. It is very efficient for low Bit Rate video compression.

Quantisation and scan order of multidimensional ICT coefficients for multidimensional vector matrix codec

Abstract This paper proposes a technique for generating the quantisation cube and an improved zigzag scanning method suitable to multidimensional vector matrix DCT integer transform (MD-VMICT) codec after studying the statistical properties of the DC and AC coefficients. An e-exponential function is used to quantisation and it is verified to two-exponential function in order to easily carrying out by shift operator. After determining the proper parameter by experiments, the proposed quantisation and scan order are tested on various standard test video sequences. The experiments show the wide adaptability. Also the comparisons are carried out with the literature and MPEG-4, whose experiment results show superiority than the literature and MPEG-4. The comparisons between MD-VMICT and H.264/AVC show potential advantages at low bit rate with high activities sequences.

Image Semantic Classification Based on Integration of Feature Subspaces

This paper proposes an image semantic classification algorithm based on feature subspaces. It is implemented by SVM and AdaBoost algorithm. In every feature subspace, a SVM is trained. According to the error rate of every SVM, the integrating weight of feature subspace is determined, with which different subspace features are concatenated into a feature vector. Then AdaBoost algorithm is employed to train classifier, whose weak classifiers are SVMs. Test and comparison are conducted between the proposed algorithm and the linear kernel SVM classifier, the RBF-SVM classifier as well as the AdaBoost classifier. The proposed algorithm is turned out to be the best one. In addition, toward the feature subspace algorithm proposed, this paper has also given the comparative experiment, its contrast group is respectively color feature space, textural feature space and equal-weight color and texture space. The experiment shows that this method increases the classifier precision effectively.