Journals / Technological Applied Sciences / 2016 / Cilt: 11 Sayı: 4

A METRIC BASED NOVEL CLASSIFICATION APPROACH TO KINECT POINT CLOUD DATA

KINECT NOKTA BULUTU VERİSİNE METRİK TABANLI YENİ BİR SINIFLANDIRMA YAKLAŞIMI

Pages
153–170
Publication Date
2016-10-13
DOI
—
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The results obtained metric classification algorithm and mean shift algorithm on automatic classification of simple planimetric object shapes with the surface of the point cloud show that proposed method is an efficient process.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Object classification in 3D point cloud data is an emerging topic attracting increasing research interest.
📊
Key Findings & Data:This paper proposes a novel method for the efficient detection of the real objects with respect to 3D point cloud data.
💡
Conclusion & Impact:The results obtained metric classification algorithm and mean shift algorithm on automatic classification of simple planimetric object shapes with the surface of the point cloud show that proposed method is an efficient process.
🔬Design:Research Article
📍Scope:Comprehensive Sample
⚙️Variables:Sınıflandırma · Nokta Bulutu · Mean-Shift

Abstract

Object classification in 3D point cloud data is an emerging topic attracting increasing research interest. Object detection is one of the most important challenges in computer vision. This paper proposes a novel method for the efficient detection of the real objects with respect to 3D point cloud data. The real object detection is performed using a technique based on mean shift clustering algorithm. The efficiency of the method is verified comparative official 3D data and real 3D point cloud data. We embed presented approach in a framework that combination of extracts shape and point cloud data metric information to improve the outcome of the classification stage. For this aim, classification of 3D point cloud data allows robust segmentation and feature descriptions into different objects by significantly reducing the error. Performed mean shift classification algorithm on the raw data and metric data classification with mean shift algorithm implementation are automatically compared to for evaluation the accuracy of the classification of metric classification algorithm. The results obtained metric classification algorithm and mean shift algorithm on automatic classification of simple planimetric object shapes with the surface of the point cloud show that proposed method is an efficient process.

Özet

3B nokta bulutu verileri ile nesne sınıflandırma araştırma alanında gelişmekte çalışmalarında en önemli sorunlardan biridir. Bu yazıda 3B nokta bulutu verilerine göre gerçek nesnelerin etkin şekilde tespiti için yeni bir metriksel yöntem önermekteyiz. Gerçek nesnelerin algılanması ortalama kayma (mean-shift) sınıflandırma algoritmasına dayalı bir yöntemle yapılmaktadır. Yöntemin verimliliği karşılaştırmalı resmi 3B verileri ve gerçek 3B nokta bulutu verileri ile doğrulanmaktadır. Nokta bulutu verileri ve metrik bilgilerin kombinasyonu bir yazılım çerçevesinde uygulanarak sınıflandırma aşamasının sonuçlarını iyileştirilmektedir. Bu amaçla, 3B nokta bulutu verilerinin sınıflandırılması önemli ölçüde hata azaltılarak farklı nesneler içine sağlam segmentasyon ve özellik çıkarımları değerlendirilmiştir. Ham veri üzerine uygulanan ortalama kayma algoritması ile metrik sınıflandırma gerçekleştirilmiş veri üzerine ortalama kayma algoritması uygulaması otomatik olarak karşılaştırılarak metrik sınıflandırma algoritmasının doğruluğu değerlendirilmektedir. Sonuçlar metrik sınıflandırma algoritmasının basit düzlemsel şekle sahip farklı nesnelerin otomatik olarak sınıflandırılmasına verimli bir sürece sahip olduğu göstermektedir

🔖

Cite This Article

Özbay, E., Çınar, A. (2016). A METRIC BASED NOVEL CLASSIFICATION APPROACH TO KINECT POINT CLOUD DATA. Technological Applied Sciences, Cilt: 11 Sayı: 4, 153–170. https://ajindex.com/articles/1007724

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (94)

  1. Eskandari, A.R. and Kouchaki, Z., (2012). Regular Shapes
  2. Detection in Satellite Images, Malaysian Journal of Computer
  3. Science, vol:25, no:1, pp:56-66.
  4. Lai, K. and Fox, D., (2010). Object Recognition in 3D Point
  5. Clouds Using Web Data and Domain Adaptation, The International
  6. Journal of Robotics Research, vol:29, no:8, pp:1019-1037.
  7. Schnabel, R., Wahl, R., Klein, R., (2007). Efficient RANSAC for
  8. Point-Cloud Shape Detection, Computer Graphics Forum,
  9. vol:26, no: 2, pp:214-226.
  10. Hao, W., Wang, Y., Ning, X., Zhao, M., Zhang, J., Shi, Z.,
  11. and Zhang, X., (2013). Automatic Building Extraction from
  12. Terrestrial Laser Scanning Data, Advances in Electrical and
  13. Computer Engineering, vol:13, no:3, pp:11-16.
  14. Zhang, Z., (2012). Microsoft Kinect Sensor and Its Effect,
  15. MultiMedia, IEEE, vol:19, no:2, pp:4-12.
  16. Zhang, M., Zhang, Z., Chang, Y., Esche, S. K., Chassapis, C.,
  17. . Kinect-based Universal Range Sensor and its Application
  18. in Educational Laboratories, International Journal of Online
  19. Engineering (IJOE), vol:11, no:2, pp:26-35.
  20. Khoshelham, K. and Elberink, S.O., (2012). Accuracy and
  21. Resolution of Kinect Depth Data for Indoor Mapping Applications,
  22. Sensors, Vol:12, No:2, pp:1437-1454.
  23. Ozbay, E. and Cinar, A., (2013). 3D Reconstruction Technique
  24. With Kinect and Point Cloud Computing, in Proc. 3rd World
  25. Conference on Information Technology, vol:3, pp:1748-1754.
  26. Ozbay, E., Cinar, A., (2013). A Novel Approach to Smoothing on
  27. d Structured Adaptive Mesh of The Kinect-Based Models, in Proc.
  28. th International Conference on Advanced Information
  29. Technologies and Applications, pp:13-22.
  30. Atmosukarto, I. and Shapiro, L.G., (2008). A Learning Approach
  31. to 3D Object Representation for Classification, in Proc.
  32. Structural, Syntactic, and Statistical Pattern Recognition,
  33. Pp:267-276.
  34. Wei, X., Phung, S.L., and Bouzerdoum, A., (2014). Object
  35. segmentation and classification using 3-D range camera, Journal
  36. of Visual Communication and Image Representation, Vol:25, No:1,
  37. pp:74-85.
  38. Berger, K., Ruhl, K., Schroeder, Y., Bruemmer, C., Scholz, A.,
  39. and Magnor, M.A., (2011). Markerless Motion Capture using
  40. multiple Color-Depth Sensors, In VMV pp:317-324.
  41. Zalevsky, Z., Shpunt, A., Malzels, A., and Garcia, J., (2013).
  42. Method and system for object reconstruction” Patent No,400,494, Washington Patent and Trademark Office.
  43. Smisek, J., Jancosek, M., Pajdla, T., (2013). 3D with Kinect,
  44. In Consumer Depth Cameras for Computer Vision Springer London,
  45. Pp:3-25.
  46. Rusu, R.B. and Cousins, S., (2011). 3D is here: Point Cloud
  47. Library (PCL), in Proc. 2011 IEEE International Conference on
  48. Robotics and Automation (ICRA), pp. 1-4.
  49. Khoshelham, K., (2011). Accuracy Analysis Of Kinect Depth Data.
  50. ISPRS workshop laser scanning, vol:38, no:5, in Proc.
  51. ISPRS Workshop Laser Scanning.
  52. Tombari, F. and Stefano, L.D., (2011). 3d Data Segmentation by
  53. Local Classification and Markov Random Fields, in Proc.
  54. International Conference on 3D Imaging, Modeling, Processing,
  55. Visualization and Transmission, pp:212-219.
  56. Weinberger, K.Q., Blitzer, J., and Saul, L.K., (2005). Distance
  57. Metric Learning for Large Margin Nearest Neighbor
  58. Classification. The Journal of Machine Learning Research,
  59. vol:10, pp:207-244.
  60. Altman, N.S., (1992). An Introduction to Kernel tnd Nearest-
  61. Neighbor Nonparametric Regression, The American Statistician,
  62. vol:46, no:3, pp:175-185.
  63. Cheng, Y., (1995). Mean Shift, Mode Seeking, and Clustering,
  64. Pattern Analysis and Machine Intelligence, IEEE Transactions,
  65. Vol:17, no:8, pp:790-799.
  66. Jaskowiak, P.A. and Campello, R.J.G.B., (2011). Comparing
  67. Correlation Coefficients as Dissimilarity Measures for Cancer
  68. Classification in Gene Expression Data, in Proc. Brazilian
  69. Symposium on Bioinformatics, pp. 1-8.
  70. Rottensteiner, F., Sohn, G., Jung, J., Gerke, M., Baillard, C.,
  71. Benitez, S., and Breitkopf, U., (2012). The ISPRS Benchmark on
  72. Urban Object Classification and 3d Building Reconstruction,
  73. ISPRS Ann. Photogramm. Remote Sens. Spat. Inf. Sci, Vol:1, No:3,
  74. pp:293-298.
  75. Klasing, K., Wollherr, D., and Buss, M., (2008). A Clustering
  76. Method for Efficient Segmentation of 3d Laser Data, in Proc.
  77. IEEE International Conference on Robotics and Automation (ICRA),, pp:4043-4048.
  78. Kim, T.K., Kee, S.C., and Kim, S.R., (2001). Real-time
  79. Normalization and Feature Extraction of 3D Face Data Using
  80. Curvature Characteristics, in Proc. 10th IEEE International
  81. Workshop on Robot and Human Interactive Communication,
  82. Pp:74-79.
  83. Hwang, J.P., Baek, J., Choi, B., and Kim, E., (2015). A Novel
  84. Part-Based Approach to Mean-Shift Algorithm for Visual Tracking,
  85. International Journal of Control, Automation and Systems,
  86. Vol:13, No:2, pp:443-453.
  87. Talu, M.F., Türkoğlu, İ., and Cebeci, M., (2010). Real-time
  88. Kernel Based Object Tracking Using Mean Shift, in Proc. 18th
  89. IEEE Signal Processing and Communications Applications (SIU),
  90. Pp:328-331.
  91. Yao, W., Hinz, S., and Stilla, U., (2009). Object Extraction
  92. Based on 3D-segmentation of Lidar Data by Combining Mean Shift
  93. with Normalized Cuts: Two Examples from Urban Areas, in Proc.
  94. IEEE Urban Remote Sensing Event, pp:1-6.