1 / 16

AUTOMATIC IMAGE ORIENTATION DETECTION

This project aims to develop a system for automatic detection of image orientation through color analysis. The primary goal is to categorize images into four distinct groups based on defined criteria. Key sub-goals include addressing color images using RGB and HSV representations, while considering efficiency metrics such as database size and runtime. The method utilizes feature vectors composed of mean and variance of color channels and edge density measurements within peripheral blocks. Future enhancements will focus on machine learning techniques, rejection criteria, and object recognition capabilities.

hop
Télécharger la présentation

AUTOMATIC IMAGE ORIENTATION DETECTION

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. AUTOMATIC IMAGE ORIENTATION DETECTION

  2. Goal Main Goal: • For any given picture detect its orientation. Sub Goals: • How to deal with color images • Define criteria for images to separate them to 4 groups: • Efficiency: DB size, vector size, runtime.

  3. What is Color?

  4. What is Color?

  5. Color representation - RGB

  6. Color difference - RGB

  7. Color representation - HSV

  8. Classify function in MatLab

  9. Peripheral blocks

  10. Edge ratio

  11. Feature Vector Vector size: N=4 • Image resolution = 800X600 • NXN blocks • 4N-4 peripheral blocks • For each block: • Mean of H,S,V • Var of H,S,V • Edge density Block size : 16 peripheral blocks: 12 Vector size: 12*(3+3+1)+4 = 88

  12. Results

  13. Results

  14. Results

  15. Future work • Improving the feature vector • Testing new method of “machine learning” • Add a rejection criteria • Add classifier of indoor/outdoor • Add an object recognition algorithm

  16. Thank you!

More Related