Dissertation > Excellent graduate degree dissertation topics show

The Research of Image Retrieval Technology Based on Visual Attention Model

Author: ZhouJing
Tutor: XiaoNanFeng
School: South China University of Technology
Course: Applied Computer Technology
Keywords: visual attention CBIR ITTI model region of interest
CLC: TP391.41
Type: Master's thesis
Year: 2012
Downloads: 206
Quote: 1
Read: Download Dissertation

Abstract


The rapid development of information technology has prompted a variety of imageinformation growing; people have to process and analyze these massive data with the help ofthe computer system. However, on one hand, the image data is increased faster than computerprocessing power; on the other hand, people only care about a very small part of the entiredata set. Therefore, how to quickly find and extract the concerned information from the dataset is an important problem. Visual attention mechanism is to solve the problem of imageretrieval technology.Visual attention mechanism overcome the" semantic gap" in a certain extent, at the sametime the image retrieval system based on the region of interest to a certain extent overcomethe impact of image background on the retrieval results. But there’s some problem in theexists visual attention model, for example, feature extraction is not full, feature fusion is toocomplex, and it’s hard to be introduced to image retrieval system etc. In view of the aboveproblems, this paper presents an image retrieval system based on improved itti visual attentionmodel. The system includes the following three parts:(1) The bottom visual feature extraction. This system will introduce texture feature to theimproved itti visual attention model, which makes the image texture feature reflected, toimprove the retrieval accuracy.(2) Image segmentation. The improved genetic algorithm based on optimal histogramthreshold image segmentation algorithm is proposed to find the optimal threshold value “Hs”,according to “Hs” to segment the multidimensional saliency maps. The improved geneticalgorithm can quickly find the candidate thresholds, followed by the Otsu algorithm to findthe optimal threshold. The results of image segmentation are more accurately.(3) Extraction of region of interest. We let the saliency points of multidimensionalsaliency maps as the seeds, respectively, we use which to do regional growth for the saliencymap after the segmentation. Then we get the interest region of various features maps. Throughthe interest region merging criterion, finally we get the region of interest. In this paper, we present a content-based image retrieval systems based on visualattention mechanism. We use efficient algorithm in image feature extraction, imagesegmentation, region growing parts, finally get the image region of interest, which is used asimage feature to compare image similarity. The experimental results show that, the system hasbetter precision and recall.

Related Dissertations

  1. Application of Q-Learning in the Content-Based Image Retrieval Technology,TP391.41
  2. Diffusion Tensor Imaging Study of Idiopathic Epilepsy at 3.0T High Field Magnetic Resonance Scaner,R445.2
  3. The Effect of Fixation on Auditory Sensory Gating,R741
  4. The Research of Medical Image Fusion Algorithm Based on Level Set Regions Segmentation,TP391.41
  5. Study on Extraction Methods of Image Interesting Regions,TP391.41
  6. Research of Shape Based Image Retrieval in Han Dynasty Stone,TP391.41
  7. Video Watermarking Research Approaches Based on Human Visual Attention,TP309.7
  8. Research for CBIR Based on Multi-intelligent Algorithms and Image Fusion,TP391.41
  9. Modulation of Self-face Processing on Visual Spatial Attention,B842
  10. Acceleration of X-ray Computed Tomography Reconstruction for Rice Tiller,TP391.41
  11. Multi-Class Object Recognition in High Resolution Remote Sensing Images,TP751
  12. A Method for Image Quality Assessment Based on Visual Attention and Structure Distortion,TP391.41
  13. Research of Visual Attention Model for Complex Object,TP391.41
  14. Research and Implement of Content-based Image Retrieval System,TP391.41
  15. Improved ROI Coding Method Based on SPIHT,TN919.81
  16. Research and Application of Visual Attention Model,TP391.41
  17. Research and Application of Compute Model on Visual Attention Mechanism,TP391.41
  18. Computational Model of Visual Attention Mechanism and Its Application Research,TP391.41
  19. Characteristic style of clothing based on cognitive research and implementation of self- classification,TS941.11
  20. Low-quality document image binarization research,TP391.41
  21. Research on Salient Region Detection Algorithms Based on Visual Attention,TP391.41

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