欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
为提高机器人在复杂背景下利用模板匹配对工件的识别率,提出基于二次阈值分割的方向倒脚匹配（orientation chamfer matching）方法。利用大津法对复杂背景的目标图像进行第一次图像阈值分割,使用迭代阈值法对目标图像进行第二次图像阈值分割,运用方向倒脚匹配的方法对工件目标进行识别。对选取的40张复杂背景下的工件图片进行测试与分析,工件的平均识别率达到72.75%,误检测降低到0.75%。实验结果表明,该改进算法与传统方向倒脚匹配相比性能有较大提升,在复杂背景下对工件识别有效可行。
英文摘要：
To enhance the robots identification rate of work-piece in the complex background,an algorithm of orientation chamfer matching based on the twin-threshold segmentation was proposed.First segmentation was carried out to process the target image in the complex background using Otsu method.And second image segmentation was carried out to process the target image using iterative threshold method with the purpose of segmenting the targets form the complex background.Orientation chamfer matching was used to recognize the targets.40 work-piece images were used to test this algorithm,the work-piece＇s average recognition rate reached 72.75% and the false detection reduced to 0.75%.Results of test show that method proposed is better than the traditional algorithm,and it is feasible and effective on recognizing work-pieces in the complex background.
同期刊论文项目
同项目期刊论文
Fault-tolerant design of a limited universalfruit-picking end-effector based on visoin positioning e
期刊信息
位置：