Method for automatic image merge
AssigneeMustek Systems Inc.
InventorTsai Jewel
Life of this patent
Estimating the term…
Abstract
A method for automatic merging two images having overlapped parts for a handy scanner. The method starts from defining a meaningful sample in a first image using fuzzy logic. Then, a two level searches based on A* algorithm are executed thereafter to find a match point in the second image. After the match point is found, the two images are stitched together using linear interpolation.
Claims18
A method for automatically merging a first image and a second image, wherein said first image and said second image are overlapped, comprising the steps of:
- a. selecting a meaningful sample from said second image;
- b. selecting an array of pixels as a parent region from said first image, said parent region having a plurality of child regions;
- c. storing said parent region and said child regions in a region queue;
- d. comparing each region in said region queue with said meaningful sample according to a first estimate function;
- e. returning a region with an estimate close to that of said meaningful sample;
- f. creating a node queue based on said region queue;
- g. comparing the pixels in each node of said node queue with said meaningful sample;
- h. defining and returning a node as a match point if said node is within a minimum error tolerance;
- and
- i. merging said first image with said second image based on said match point.
A method for automatically merging a first image and a second image, wherein said first image and said second image are overlapped, comprising the steps of:
- a. selecting a meaningful sample from said second image;
- b. selecting a parent region from said first image, said parent region having a plurality of child regions;
- c. comparing said parent region with said meaningful sample and returning a rough region having an estimate close to said meaningful sample according to a first A* estimate function;
- d. comparing said meaningful sample with said rough region and returning a match point according to a second A* estimate function;
- e. repeating said step (c) until a best node is found or all nodes in said parent region have been examined;
- f. repeating said step (d) until a best node is found or the search space for said second image is exhausted;
- and
- g. applying linear interpolation to merge said first image data with said second image data based on said match point.
Dependencies are inferred from the claim text.