Dissertation > Excellent graduate degree dissertation topics show

Contour and Boundary Detection via Perceptual Mechanisms of Primary Visual Cortex

Author: TangQiLing
Tutor: SangNong
School: Huazhong University of Science and Technology
Course: Pattern Recognition and Intelligent Systems
Keywords: Contour and boundary detection Context interaction Environmental suppression Spatial enhancement Non-classical receptive field Primary visual cortex Visual perception mechanism
CLC: TP391.41
Type: PhD thesis
Year: 2007
Downloads: 409
Quote: 5
Read: Download Dissertation

Abstract


Contours and boundaries define the outer shape of the target to determine the dividing line between the regions, they are humans and computers important feature of target recognition. However, from the chaos of the natural contours of the object in the scene to extract the boundary is a very difficult task. To do this, three main issues need to be addressed: 1. Exclude a large background texture generated by the local edge ingredients; 2 scene context information based on local ingredients will be organized into meaningful global features; 3 Some important structural lack of a clear physical definition (for example, the boundary of the texture), and some part of the target and background may be of the same strength, so that the boundary or the lack of response is very weak partial valid evidence. In response to these difficulties, the paper according to the primary visual cortex perception mechanisms established various contours and boundary detection model, and by combining images and natural images verify performance of the algorithm. In order to reduce the background texture edge ingredients and highlight the border region, this cycle using non-classical receptive field inhibition of the dynamic properties of texture presents a method for inhibiting. This method textures and borders to take a different way, which significantly reduces the background interference components meaningless, and selectively preserved isolated silhouettes and regional boundaries. How complex scenes with the same spatial structure will be organized into a significant component of the outline is another key aspect of the study. In this paper, concyclic rules and visual preference for low-curvature path defines a silhouette combined with local aggregation function that will work with the arrangement and orientation of the two properties with clever link. The interaction through the context of local ingredients will be integrated into a meaningful global features and highlights from the background. By spatially separated excitement and inhibition zone, this paper will enhance the environment and spatial suppression combined in a unified model, allowing two opposing perceived behavioral exist. Based on this model, this paper emphasizes the spatial enhancement and environmental boundary detection and suppression in silhouette play different roles, mainly in the surface and the inhibition of texture segmentation, which is mainly used to enhance the binding contours and graphics background isolated. Color images carry more than the gray-scale image information of the image and can help to produce better results. Therefore, to further expand the gray model to color image processing. Color model will involve more properties homogeneity suppression, can more effectively remove the texture edge; On the other hand, the color contours provide more information gathered, more conducive to the integration of the same attributes. Finally, two application projects - angiography image enhancement and synthetic aperture radar images of the road detection, indicating extensive use of the model.

Related Dissertations

  1. Design of Visual Experimental Platform Based on Optical Imaging and Microelectrode Array,TP391.41
  2. Visual activity of the primary visual cortex of SD rats Ⅱ / Ⅲ layer inside the neuron level lateral synaptic connections plasticity,Q42
  3. Electrical Activites Characteristics of Rat Primary Visual Cortex to Flash Stimulation,Q42
  4. The Intervention and Mechanism Research of BuShenHuoXue on Primary Visual Cortex Damage in Rat Model of Chronic Elevated Intraocular Pressure,R276.7
  5. Neuronal Spike Sorting in Primary Visual Cortex of Rats Based on Matched Wavelet Transform,Q42
  6. The Analysis for Response Signal of Rats Primary Visual Cortex V1 under the Visual Stimulation,TN911.6
  7. Spike-sorting Based on Genetic Algorithm-support Vector Machine and Characteristic Analysis,TP18
  8. The primary visual cortex of SD rats Ⅱ / Ⅲ layer excitatory neurons during the development of the internal lateral plasticity of synaptic connections long -term potentiation ( LTP ),Q42
  9. GABA Improve Response Modulation of Primary Visual Cortex in Chronic Morphine Exposed Cats,Q42
  10. The Hyper-Column Structure Research of Primary Visual Cortex of Bionic Brain,TP391.41
  11. Contour Detection Based on Primaty Visual Model,TP391.41
  12. Contour detection algorithm research based on the non-classical receptive field properties,R318
  13. A Discussion and Analysis on Teacher’s Talk in College English Integrated Course,H319
  14. The Retinal Distribution of Neurons with Different Integrate Field Properties in the Cat Primary Visual Cortex,Q43
  15. Endocardiac Boundary Extraction Technology and Application in Echocardiogram,R445.1
  16. The cat Contrast identify the neural mechanisms of perceptual learning research,Q42
  17. Expression of mGluR1 at Primary Visual Cortex of Monocular Deprived Amblyopia Rats and the Observation of Neuronic Ultrastructure,R777.44
  18. The Research on the Neural Basis of Functional Degradation of Visual Cortical Cells in Aged Rats,Q42
  19. Image shape to stimulate the endogenous signal optical imaging in cat visual cortex and its mechanism,Q43
  20. Image Representation Model Based on Primary Visual Mechanism,TN919.81
  21. The Neural Coding of Retinathe Computational Models of Two Types W Ganglion Cells in the Cat,TN919.81

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