| 查看: 942 | 回复: 4 | ||
| 本帖产生 1 个 翻译EPI ，点击这里进行查看 | ||
[求助]
求助翻译一段摘要
|
||
| 针对细胞图像序列模糊，传统的特征提取方法鲁棒性不强，伪匹配点对较多，图像匹配耗时过长，融合效果不佳等问题，提出了一种强鲁棒性、快速和精确的图像拼接算法。该算法首先用基于尺度不变(SIFT)算法提取细胞图像特征点；接着采用改进的BBF(Best-Bin-First)算法对特征集进行初始的双向匹配；然后采用随机抽样一致性(RANSAC)算法对匹配点对进行进一步提纯并估算出单应性矩阵；最后根据细胞图像序列之间的单应性矩阵关系，将其投影到统一标准的平面坐标系下，用具有塔型结构的多分辨率融合算法对图像进行无缝融合，完成全景图拼接。实验结果证明：该算法提取到的提取到的特征点分布均匀且数量适中，误配情况明显减少，能够有效地实现显微全景图的无缝拼接。 |
回复此楼» 猜你喜欢
课题组招2027级博士 上海工程技术大学 激光智能制造方向
已经有4人回复
-
现代”学阀”该如何界定
已经有8人回复
-
上海工程技术大学 激光智能制造课题组 2027级博士研究生招生
已经有3人回复
-
实名举报：华南理工大学物理与光电学院副院长李志远婚内两次出轨女学生（博士）
已经有16人回复
-
上海工程技术大学激光智能制造课题组｜2027级博士研究生招生公告
已经有11人回复
-
上海工程技术大学激光智能制造课题组招收博士研究生
已经有10人回复
-
申博发邮件
已经有11人回复
-
各位大神，目前国内有哪些比较好用的逆合成软件？
已经有12人回复
-
我的奶奶
已经有3人回复
-
国社科系统bug了，是不是要放榜了？
已经有9人回复
sunfeng_upc
至尊木虫 (职业作家)
- 应助: 62 (初中生)
- 金币: 10391.5
- 红花: 8
- 帖子: 4935
- 在线: 169.5小时
- 虫号: 2977961
- 注册: 2014-02-19
- 性别: GG
- 专业: 微/纳机械系统
【答案】应助回帖
RXMCDM: 好像翻得不行啊！ 2014-09-03 17:47:19
| For the blurred cell image sequence , the traditional robust feature which is extracted is not evident, pseudo matching points more, image matching takes too long, the effect of poor integration and other issues, presents a strong robust, fast and accurate the image stitching algorithm. The algorithm is based on first with scale-invariant (SIFT) algorithm for image feature points extracted from the cells; then using the improved BBF (Best-Bin-First) algorithm features a two-way matching initial set; then use random sampling consistency (RANSAC) algorithm matching point for further purification and estimate the homography; Finally, according homography relationship between image sequences cells, down to its projection plane coordinate system of uniform standards, with a tower-type structure with multi-resolution seamless fusion algorithm for image fusion, complete panorama stitching. Experimental results show that: the algorithm to extract the feature points extracted to moderate amount evenly distributed and significantly reduce the mismatch situation, can effectively achieve seamless microscopic panorama. |
2楼2014-09-03 16:57:28
3楼2014-09-03 17:35:40
4楼2014-09-03 19:53:38
RXMCDM
版主 (文坛精英)
- 翻译EPI: 530
- 应助: 401 (硕士)
- 贵宾: 1.908
- 金币: 38947.6
- 散金: 4908
- 红花: 88
- 沙发: 4
- 帖子: 11453
- 在线: 1355.6小时
- 虫号: 2739168
- 注册: 2013-10-20
- 性别: GG
- 专业: 一般管理理论与研究方法论
- 管辖: 论文翻译
【答案】应助回帖
★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★
xiuzi0731: 金币+35, 翻译EPI+1, ★有帮助, 感谢版主 2014-09-04 08:15:39
xiuzi0731: 金币+35, 翻译EPI+1, ★有帮助, 感谢版主 2014-09-04 08:15:39
| To overcome the problem of cell image sequence blur, bad robust of the traditional extraction method,too much pseudo matching points,long time of the image matching, bad effect of integration,etc.;we presented a good robust ,quick ,and accurate image stitching algorithm. The presented algorithm is based on,firstly,scale-invariant (SIFT) algorithm to extract featured points of the cell image and improved BBF (Best-Bin-First) algorithm to initially set a two-way matching of the featured set. Then,the matching point was further purified and homography estimated by the use of random sampling consistency (RANSAC) algorithm .Finally,the homography relationship of the cell image sequences was projected to standardized plane coordinate system for seamless fusion of the image through a tower- typed structure fusion algorithm of multi-resolution to complete panorama stitching. Results showed that moderate amount extracted feature points were evenly distributed,mismatches were reduced,and the algorithm can realize seamless fusion of microscopic panorama effectively. |
5楼2014-09-03 23:42:23