Journal of Computer Applications ›› 2021, Vol. 41 ›› Issue (7): 2062-2069.DOI: 10.11772/j.issn.1001-9081.2020091501

Special Issue: 多媒体计算与计算机仿真

• Multimedia computing and computer simulation • Previous Articles     Next Articles

Adaptive binary simplification method for 3D feature descriptor

LIU Shuangyuan1,2, ZHENG Wangli3, LIN Yunhan1,2   

  1. 1. School of Computer Science and Technology, Wuhan University of Science and Technology, Wuhan Hubei 430065, China;
    2. Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System(Wuhan University of Science and Technology), Wuhan Hubei 430065, China;
    3. State Grid Electric Power Research Institute Company Limited, Nanjing Jiangsu 211106, China
  • Received:2020-09-27 Revised:2020-12-22 Online:2021-07-10 Published:2021-01-20
  • Supported by:
    This work is partially supported by the Hubei Natural Science Foundation Youth Project (2020CFB116), the Scientific Research Program for Young and Middle-Aged Talents of Department of Education of Hubei Province (Q20191108).

面向三维特征描述子的自适应二进制简化方法

刘双元1,2, 郑王里3, 林云汉1,2   

  1. 1. 武汉科技大学 计算机科学与技术学院, 武汉 430081;
    2. 智能信息处理与实时工业系统湖北省重点实验室(武汉科技大学), 武汉 430081;
    3. 国网电力科学研究院有限公司, 南京 211106
  • 通讯作者: 林云汉
  • 作者简介:刘双元(1995-),男,山西忻州人,硕士研究生,主要研究方向:机器人感知、三维视觉;郑王里(1985-),男,福建仙游人,工程师,硕士,主要研究方向:电力通信设备检测;林云汉(1985-),男,福建仙游人,讲师,博士,机器人感知、机器人推理与控制。
  • 基金资助:
    湖北省自然科学基金青年项目(2020CFB116);湖北省教育厅科学研究计划中青年人才项目(Q20191108)。

Abstract: In the study of 3-Dimensional (3D) local feature descriptor, it is difficult to strike a balance among accuracy, matching time and memory consumption. To solve this problem, an adaptive binary simplification method for 3D feature descriptor was proposed based on the standard deviation principle in statistical theory. First, different binary feature descriptors were generated by changing the binarization unit length and the number of standard deviations in the simplification model, which were applied into the currently widely used Signature of Histogram of OrienTations (SHOT) descriptor, and the optimal combination of binarization unit length and the number of standard deviations was determined by experiments. Finally, the simplified descriptor under the optimal combination was named Standard Deviation feature descriptor for Signature of Histogram of OrienTations (SD-SHOT). Experimental results show that compared with the SHOT descriptor without simplification, SD-SHOT reduces the key point matching time to 1/15 times and the memory occupancy to 1/32 times of SHOT; compared with the existing mainstream simplification methods such as Binary Feature Descriptor for Signature of Histogram of OrienTations (B-SHOT), SD-SHOT has the optimal comprehensive performance. In addition, the validity of the proposed method is verified in the actual robot sorting scene consisting of five different categories of objects.

Key words: adaptive binary method, 3-Dimensional (3D) feature descriptor, 3-Dimensional (3D) object recognition, point cloud, object sorting

摘要: 在三维(3D)局部特征描述子研究中,准确度、匹配时间以及内存消耗存在此消彼长的问题。针对上述问题,基于统计理论中的标准差原理提出一种面向3D特征描述子的自适应二进制简化方法。首先,通过改变简化模型中二值化单元长度和标准差个数来生成不同的二进制特征描述子;然后,将它们应用到当前被广泛使用的基于签名的方向直方图(SHOT)描述子中,并通过实验确定最优的二值化单元长度和标准差个数的组合;最后,将最优组合下的简化描述子命名为SD-SHOT。实验结果表明,与未进行简化的SHOT描述子相比,SD-SHOT在关键点匹配时间上减少为原来的1/15,内存占有率降低为原来的1/32;与现有主流简化方法如B-SHOT等相比,SD-SHOT的性能达到了综合最优水平。此外,在由五种不同类别的物体构成的实际机器人分拣场景中验证了所提方法的有效性。

关键词: 自适应二进制方法, 三维特征描述子, 三维物体识别, 点云, 物体分拣

CLC Number: