Abstract
Background: A physical object, which is actually in 3D form, is captured by a sensor/camera (in case of computer vision) and seen by a human eye (in case of a human vision). Whensomeone is observing something, many other things are also involved there which make it more challengingto recognize. After capturing such a thing by a camera or sensor, a digital image is formedwhich is nothing other than a bunch of pixels. It is becoming important to know that how a computerunderstands images.
Objective: This paper is for highlighting novel techniques on 3D object recognition system with localshape descriptors and depth data analysis.
Methods: The proposed work is applied to RGBD and COIL-100 datasets and this is of four-fold aspreprocessing, feature generation, dimensionality reduction, and classification. The first stage of preprocessingis smoothing by 2D median filtering on the depth (Z-value) and registration by orientationcorrection on 3D object data. The next stage is of feature generation and having two phases of shapemap generation with shape index map and SIFT/SURF descriptors. The dimensionality reduction is thethird stage of this proposed work where linear discriminant analysis and principal component analysisare used. The final stage is fused on classification.
Results: Here, calculation of the discriminative subspace for the training set, testing of object data andclassification is done by comparing target and query data with different aspects for finding propermatching tasks.
Conclusion: This concludes with new proposed approach of 3D Object Recognition. The local shapedescriptors are used for 3D object recognition system to implement and test. This system is achieves89.2% accuracy for Columbia object image library-100 images by using local shape descriptors.
Keywords:RGBD, SIFT, SURF, COIL-100, principal component analysis, depth map, 3D classification.
Graphical Abstract
Recent Patents on Computer Science
Title:3D Object Recognition System Based On Local Shape Descriptors and Depth Data Analysis
Volume: 12Issue: 1
Author(s):Chiranji Lal Chowdhary*
Affiliation:
- Department Computer Applications and Creative Media, School of Information Technology and Engineering, Vellore Institute of Technology, Vellore,India
Keywords:RGBD, SIFT, SURF, COIL-100, principal component analysis, depth map, 3D classification.
Abstract: Background: A physical object, which is actually in 3D form, is captured by a sensor/camera (in case of computer vision) and seen by a human eye (in case of a human vision). Whensomeone is observing something, many other things are also involved there which make it more challengingto recognize. After capturing such a thing by a camera or sensor, a digital image is formedwhich is nothing other than a bunch of pixels. It is becoming important to know that how a computerunderstands images.
Objective: This paper is for highlighting novel techniques on 3D object recognition system with localshape descriptors and depth data analysis.
Methods: The proposed work is applied to RGBD and COIL-100 datasets and this is of four-fold aspreprocessing, feature generation, dimensionality reduction, and classification. The first stage of preprocessingis smoothing by 2D median filtering on the depth (Z-value) and registration by orientationcorrection on 3D object data. The next stage is of feature generation and having two phases of shapemap generation with shape index map and SIFT/SURF descriptors. The dimensionality reduction is thethird stage of this proposed work where linear discriminant analysis and principal component analysisare used. The final stage is fused on classification.
Results: Here, calculation of the discriminative subspace for the training set, testing of object data andclassification is done by comparing target and query data with different aspects for finding propermatching tasks.
Conclusion: This concludes with new proposed approach of 3D Object Recognition. The local shapedescriptors are used for 3D object recognition system to implement and test. This system is achieves89.2% accuracy for Columbia object image library-100 images by using local shape descriptors.
Export Options
About this article
Cite this article as:
Chowdhary Lal Chiranji*, 3D Object Recognition System Based On Local Shape Descriptors and Depth Data Analysis, Recent Patents on Computer Science 2019; 12 (1) .https://dx.doi.org/10.2174/2213275911666180821092033
| DOI https://dx.doi.org/10.2174/2213275911666180821092033 | Print ISSN 2213-2759 |
| Publisher Name Bentham Science Publisher | Online ISSN 1874-4796 |
Related Journals
International Journal of Sensors, Wireless Communications and Control
Recent Advances in Computer Science and Communications
Current Computer Science
Journal of Intelligent Systems in Current Computer Engineering
Wireless and Communication Letters
Current Machine Learning
Current Science, Engineering and Technology
Related Books
Computational Intelligence and its Applications
Demystifying Emerging Trends in Machine Learning
Blockchain-Enabled Internet of Things Applications in Healthcare: Current Practices and Future Directions
Data Management and Security in Blockchain Systems
Artificial Intelligence Development in Sensors and Computer Vision for Health Care and Automation Application
Practical Digital Forensics: A Guide for Windows and Linux Users
The Future of Computing: Ubiquitous Applications and Technologies
Metal Complexes: Environmental and Biomedical Applications
Artificial Intelligence: A Multidisciplinary Approach towards Teaching and Learning
Smart Home and Industrial IoT Devices: Critical Perspectives on Cyberthreats, Frameworks and Protocols
19
3[8]ページ先頭