| 查看: 807 | 回复: 1 | ||
| 本帖产生 1 个 翻译EPI ，点击这里进行查看 | ||
[求助]
论文翻译。。英译中。翻译成中文！！！
|
||
|
1)In order to help the humanoid robots build the environment map，a method of extracting the planes was proposed．The pair of original images was acquired by two CCD cameras，and through stereo matching，the depth image could be obtained．The median filter was adopted to reduce the noise 1eve1 of the depth image．The noise variance of image was estimated by averaging the estimates of the noise variance at each pixel for the sake of setting the thr eshold．Pixels of each line were scanned and split into two parts at the point furthest away along the z—axis from the chord between its tw o end points iteratively．The seed regions were selected from the groups of three line segments．Queue storing seed regions were ordered by the optimality criterion，and then the plane region was expanded around the seed region．Th e process of region growing would continue until the queue of seed region was empty．After trimming，the border of the planes became smooth．In the end，the edges of extracted planes were got．2、Two types of model：pipe and stairs in the experiment were adopted．Two planes in pipe model and 632 J．Cent．South Univ．Techno1．(201 1、18：627—632 six planes in stairs model were extracted exactly．The algorithm satisfied the demands of human oid robot’s navigation in terms of speed and precision．The plane would be flagged with level if the normal direction of plane was vertical to the robot’s reference surface，and then this information could be used for building the surrounding map. [ Last edited by 123wy123 on 2015-6-18 at 13:19 ] |
回复此楼» 猜你喜欢
考博自荐
已经有4人回复
-
售SCI一区T0P文章，我:8.O.5.5.1.O.5.4,科目齐全,可+急
已经有4人回复
-
售SCI一区T0P文章，我:8.O.5.5.1.O.5.4,科目齐全,可+急
已经有6人回复
-
售SCI一区T0P文章，我:8.O.5.5.1.O.5.4,科目齐全,可+急
已经有5人回复
-
售SCI一区T0P文章，我:8.O.5.5.1.O.5.4,科目齐全,可+急
已经有6人回复
-
申请博士
已经有3人回复
-
裁员滚滚，退居二线
已经有4人回复
-
2026博士还有哪些学校有名额
已经有6人回复
-
这年头没有找到涵评专家，还有中面上的可能吗
已经有13人回复
-
售SCI一区T0P文章，我:8.O.5.5.1.O.5.4,科目齐全,可+急
已经有4人回复
baoshanqiu
至尊木虫 (著名写手)
- 翻译EPI: 204
- 应助: 103 (高中生)
- 金币: 15293.5
- 红花: 41
- 帖子: 1611
- 在线: 431.6小时
- 虫号: 3856202
- 注册: 2015-05-08
- 性别: GG
- 专业: 胃肠道免疫相关疾病
【答案】应助回帖
★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★
123wy123: 金币+25, 翻译EPI+1, 非常感谢！！！谢谢 2015-06-22 11:51:18
123wy123: 金币+25, 翻译EPI+1, 非常感谢！！！谢谢 2015-06-22 11:51:18
| 1）为了帮助人形机器人构建环境映像，设计了一种提取平面的方法。由两台电荷耦合（CCD）摄像机获得成对的原始图像，经过立体匹配可得到深度图像。采用中滤波器降低深度图像的噪音水平。图像的噪音离散是估算设定阈值时每一个像素噪音离散的平均值。交替扫描每一条线的像素并在两个端点之间带沿z轴上最远点将像素分成两部分。种子区域从三线段组选取。排列存放种子区域按最优化标准排序，平面区域就在种子区域周围扩张。区域生长过程持续到排列的种子区域用完。经过修饰，平面的边界变得光滑。最后得到提取平面的边缘。2、两种模型：试验中采用了管道和楼梯模型。两个平面用管道模型 632 J．Cent．South Univ．Techno1．(201 1、18：627—632）。六个平面用楼梯模型提取。在速度和精度上算法符合人形机器人导航的要求。如果平面的正常方向垂直于机器人的参考表面，则以等级标记该平面，这个信息就可用于构建周围映像。 |
2楼2015-06-18 23:14:40
20