researching - setup enviroment and sift image
TRANSCRIPT
MULTI-CAMERA APPLICATION
SUCH AS MULTI-FOCUS OR MULTI-CUE FOR CAPTURE IMAGE
WEEKLY REPORT: NOV 8 2016
Professor: Pei-Jun LeeResearcher: Bui Trong An
OUTLINE
• Introduce topic• Issue• Progressing• In Progressing• To do
INTRODUCE RESEARCH TOPIC
Multi-cameras application, such as multi-focus or multi-cue for capture image.
ISSUE
• The limit of depth-of-focus if optical lenses in Charged Coupled Device (CCD) devices is possible to obtain an image that contains all of the relevant object in focus.
CCDs are sensors used in digital cameras and video cameras to record still and moving images. The CCD captures light and converts it to digital data that is recorded by the camera.
PROGRESSING Published: Information Fusion · May 2015
IN PROGRESS
• SIFT algorithm
• OpenCV for Mobile Environment
• Convert Original image to SIFT Image ( Step 1 in Progressing)
• Result ( SIFT Demo)
SIFT ALGORITHMThere are mainly four steps involved in SIFT algorithm:
* IDEA OF ALGORITHM
1. Scale-space Extrema Detection
2. Keypoint Localization
3. Orientation Assignment
4. Keypoint Descriptor
http://docs.opencv.org/3.1.0/da/df5/tutorial_py_sift_intro.html
IDEA
SCALE-SPACE EXTREMA DETECTION
Search over multiple scales and image locations.
Detail: SIFT algorithm uses Difference of Gaussians
KEYPOINT LOCALIZATION• Fit a model to detrmine location and scale.
• Select keypoints based on a measure of stability.
ORIENTATION ASSIGNMENT• Compute best orientation(s) for each key point region.
KEY POINT DESCRIPTION• Use local image gradients at selected scale and rotation to describe each keypoint region.
OPENCV FOR MOBILE APP (ANDROID)
FUNCTION CONVERT
SIFT DEMO
SIFT DEMO
TO DO
• RESEARCH P1 AND P2 IN A PROGRESSING
PROBLEM
• SIFT PUBLISHED MAY 2015
• Robust Multi-Focus Image Fusion Using Multi-Task Sparse Representation and Spatial Context (MAY 2016)
8.0 MP -> 3~5s
20.7 MP -> ~20s