Dissertation > Excellent graduate degree dissertation topics show

Research on Key Techniques of Content-Based Image Retrieval

Author: ZhaoShan
Tutor: ZhouLiHua
School: Xi'an University of Electronic Science and Technology
Course: Applied Computer Technology
Keywords: Content-based image retrieval (CBIR) texture primitive keyblocks color vector angle salient points bit-plane distribution entropy image retrieval in DCT domain
CLC: TP391.41
Type: PhD thesis
Year: 2007
Downloads: 1411
Quote: 14
Read: Download Dissertation


With the development of the computer,multimedia and internet techniques,people can get more and more image information.How to rapidly and effectively search the desired images from large-scale image database becomes an active hot point in the area of retrieval research.Content-base image retrieval(CBIR)is a set of techniques to solve the problems based on automatically derived image features.In recent years,CBIR is a very active research direction and has been applied to many fields.In this dissertation,the exploratory research work has been done around the low-level feature extraction,which include color,shape,spatial features and so on.The main contributions of this dissertation are summarized as follows:1.Several key techniques and algorithms of CBIR are deeply analyzed and discussed,such as,color space,the low-level feature descriptions including color,shape,and texture,the similarity measure between the features and the evaluation methods of image retrieval algorithms.2.Two image retrieval approaches based on block truncation coding(BTC)are proposed.(1)Firstly,combing the human visual feature with the principles of BTC,the texture primitive of images is defined.According to the distribution of these primitives,a texture primitive co-occurrence matrix is introduced and a few of significant values are extracted to describe the texture feature.At the same time,primitive histogram is presented as the shape feature of images. Finally,the two features are integrated into the image retrieval.(2)In order to effectively use existing text information retrieval(TIR)methods in content-based image retrieval,the keyblocks of image is defined according to the BTC.The image can be considered as a text and the TIR method is adopted into the image retrieval.Finally,the effect of the frequency of different kinds of keyblocks on the image retrieval is taken into account,a weighted function is proposed.3.A novel image retrieval algorithm based on color vector angle is presented.At first,the properties of color vector angle of being insensitive to light but sensitive to Hue and Saturation is discussed.Then the image is divided into two parts according to the features of the color vector angle,thus the spatial distribution information is introduced.Then,two spatial descriptors,directional color vector angle histogram and spatial union-distribution entropy are defined to extract the features of the two domains.At the same time,three color channels are considered in the algorithm to avoid the mistaken retrieval induced by the color qualitation.Finally,the image retrieval including these two descriptors is recommended.4.A new image retrieval method based on bit-plane distribution entropy is reported.Based on the analysis of problems of the histogram in image retrieval, the bit-plane distribution entropy is adopted to describe the image feature combining the bit-plane with the entropy.The spatial information of the image is extracted by the ideas of bit-planes,the entropy is recommended in order to reduce feature dimension and three color channels are done to avoid the mistaken retrieval by the color qualitation.Meantime,the effect of the changes in image intensity values on bit-planes are taken into account,the gray-coded of bit-plane is adopted.Finally,the Mahalanobis Distance is used to measure the similarity because of the correlation between the concerned vectors after designing the correlation-weighted matrix.5.An image retrieval algorithm based on image salient points is introduced.Firstly, by deep analysis of the methods on interest points in image retrieval,a newly robust and self-adaptive extraction algorithm of salient points is given in this paper based on an improved block difference of inverse probabilities model (BDIP).Firstly,the BDIP image is extracted to express the original image by using some significant data.Then the salient points are proposed according to the distribution of pixels in the BDIP image.Based on which,the color feature and the whole shape feature are considered to image retrieveal.6.An image retrieval approach based on texture and shape in DCT domain is given by intruducing the spatial information of DCT blocks and DCT coefficients into the feature extraction.Firstly,according to the complex of the DCT blocks and distribution of DCT coefficients,a weighted complex histogram is presented to extract the texture information.At the same time, DCT blocks are classified into different edges using five special coefficients in the DCT blocks and a spatial distribution map is constructed.Based on which, the edge-spatial distribution feature is extracted as the shape feature.Finally,the effect of the energy of the points in the map on the image retrieval is considered, a weighted function is introduced.

Related Dissertations

  1. Application of Q-Learning in the Content-Based Image Retrieval Technology,TP391.41
  2. The Research of Image Retrieval Technology Based on Visual Attention Model,TP391.41
  3. Research of Image Retrieval Based on Texture Features,TP391.41
  4. Image Similarity Matching Based on Colour and Texture,TP391.41
  5. Texture element extraction and texture synthesis method and its application,TP391.41
  6. Image Retrieval Using Salient Points and Keyblocks,TP391.3
  7. The Research on the Key Technologies of Content-Based Image Database Retrieval,TP391.3
  8. Research on Image Retrieval Technology Based on Texture Spectrum Feature,TP391.41
  9. Signature image retrieval technology based on multi-feature,TP391.41
  10. Research on Key Techniques of Content-Based Image Retrieval,TP391.41
  11. Properties of guidance contains feedback mechanism , content-based image classification,TP391.41
  12. Image Retrieval Based on Geometric Partitioning for Edge Structure,TP391.3
  13. Color-Based Image Retrieval Methods and System Implementation,TP391.3
  14. Research and Development of Product Design Patent Retrieval Analysis System,TP311.52
  15. Research and Realization of CBIR System Basted on Domain Color and Its Distribution,TP391.3
  16. Research and Implementation of Content-Based Image Retrieval Algorithm,TP391.3
  17. Research of Image Retrieval System Combining with Visual and Semantic Features,TP391.41
  18. Study on Several Issues of Content-based Image Retrieval,R318
  19. Research on Content-Based Image Retrieval,TP391.3
  20. Study on the Key Techniques for the Control and Security Mechanism of the Object-Oriented Multimedia Information System,TP37

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Image recognition device
© 2012 www.DissertationTopic.Net  Mobile