发明名称 |
DISPARITY ESTIMATION METHOD OF STEREOSCOPIC IMAGE |
摘要 |
A disparity estimation method of stereoscopic image is provided. A matching cost computation is executed for a first image and a second image, and one extreme value is selected from cost values corresponding to estimating disparities for each pixel to obtain a matching point corresponding to each pixel. And a matching disparity corresponding to each matching point is adjusted based on edge detection according to a scan order. |
申请公布号 |
US2014254918(A1) |
申请公布日期 |
2014.09.11 |
申请号 |
US201313950271 |
申请日期 |
2013.07.24 |
申请人 |
Acer Incorporated |
发明人 |
Cheng Chao-Chieh;Chang Tian-Sheuan |
分类号 |
G06K9/00 |
主分类号 |
G06K9/00 |
代理机构 |
|
代理人 |
|
主权项 |
1. A disparity estimation method of stereoscopic images for an electronic apparatus, comprising:
obtaining a plurality of cost values of each of a plurality of pixels of a first image corresponding to a second image within a disparity search range, wherein a size of the first image is identical to that of the second image; selecting an extreme value from the cost values corresponding to each of the pixels and using the extreme value as a matching cost value to obtain a matching point corresponding to each of the pixels, wherein each of the matching points has a corresponding matching disparity; and adjusting the matching disparity corresponding to each of the matching points according to a scan order, and the step comprising:
reading disparity data of the matching point corresponding to each of the pixels individually according to the scan order and executing a first computation through a first energy function based on edge detection according to the read disparity data of a current pixel and the read disparity data of pixels adjacent to the current pixel; andexecuting a second computation through a second energy function based on the edge detection according to the disparity data of a currently processed pixel and the disparity data of the pixels adjacent to the currently processed pixel. |
地址 |
New Taipei City TW |