The Implementation of Object Recognition using Deformable Part Model (DPM) with Latent SVM on Lumen Robot Friend

Kurniawan A, Saputra R, Marzuki Marzuki, Febrianti M S, Prihatmanto A S

Abstract


Object recognition is part of image processing. It is used to recognize the surrounding objects based on their features, then they are processed further to obtain valid data information that can be used for other purposes. Currently, object recognition is mostly used by robot developers as one of the features in humanoid robot. One of the recent challenges occurring in humanoid robot is how the robot detects and localizes generic objects from categories such as human or car in static images using sensory visuals. It is a difficult problem since objects in such categories vary both in appearance and shape. For example, it is difficult to recognize an object that most of its shape is blocked by other objects. To solve the problem, researcher used Deformable Part Model and latent svm methods, where the data collection was performed through Library Research and Field Research approaches. The conclusion of this research is that recognition to objects using deformable part model provides a passable accuracy. After 3 experiments had been performed, system was able to recognize objects with highest reach by 88%.

Keywords


Object Detection, Deformable Part Model, Object Recognition, pyramid HOG, LSVM

Full Text:

PDF

References


P. F. Felzenszwalb, R. B. Girshick, D. McAllester, and D. Ramanan, “Object detection with

discriminatively trained part-based models,” Pattern Analysis and Machine

Intelligence, IEEE Transactions on, vol. 32, no. 9, pp. 1627–1645, 2010.

P. F. Felzenszwalb, R. B. Girshick, D. McAllester, and D. Ramanan, “Object detection with

discriminatively trained part-based models,” Pattern Analysis and Machine

Intelligence, IEEE Transactions on, vol. 32, no. 9, pp. 1627–1645, 2010.

N. Dalal, B. Triggs, "Histograms of oriented gradients for human detection", Computer

Vision and Pattern Recognition 2005. CVPR 2005. IEEE Computer Society Conference

on, vol. 1, pp. 886-893, 2005.

R. Ranjan, V. M. Patel, and R. Chellappa, “A Deep Pyramid Deformable Part Model for Face

Detection A Deep Pyramid Deformable Part Model for Face Detection”, International

Conference on Intelligent Robots and Systems (IROS), 2015 IEEE. pp. 4880-4887,

S. Mittal, T. Prasad, and S. Saurabh, “Pedestrian detection and tracking using deformable

part models and Kalman filtering,” Int. SoC Des. Conf., 2012 IEEE, pp. 324–327,

G. Jie, Z. Honggang, C. Daiwu, and Z. Nannan, “Object Detection Algorithm Based on

Deformable Part Models”, Proceedings of IC-NIDC2014, 2014 IEEE, pp. 90-94, 2014.

L. C. Leon and R. Hirata, “Vehicle detection using mixture of deformable parts models: Static

and dynamic camera,” in Brazilian Symposium of Computer Graphic and Image

Processing, 2012 IEEE, pp. 237-244, 2012.

Kurniawan, A., “Implementasi Pengenalan Objek Menggunakan Deformable Part

Model(DPM) Pada Lumen Robot Friend”, Informatics Final Project, Universitas

Bandar Lampung, 2017.

Syarif, P. Nhirun, and S. Sholata, Pengembangan Lumen Sebagai Robot Pemandu Pameran,

Studi Kasus Electrical Engineering Days 2015, B300 Engineering Documents, Bandung

Institute of Technology, 2015.

Cygane. B, 2013. ”Object Detection And Recognation In Digital Images”. Poland : University

of Science and Technology.

W. Pratt, 2007. “Digital Image Processing”, New York : PIKS Scientific Inside. Hoboken,

NJ : John Wiley & Sons, Inc.

Marzuki, A. Sukoco, and M. S. Febrianti, “Visual-based machine understanding framework

for decision making on social robot,” 2015 4th Int. Conf. Interact. Digit. Media, 2015

IEEE, pp. 1–6, 2015.


Refbacks

  • There are currently no refbacks.


International Conference on Engineering and Technology Development (ICETD)
Bandar Lampung University
ISSN: 2301-5690