A Fourier Descriptor of 2D Shapes Based on Multiscale Centroid Contour Distances Used in Object Recognition in Remote Sensing Images
<p>The upper half is a blurred remote sensing image and the lower half shows the shapes of the objects in the upper half.</p> "> Figure 2
<p>Two shapes that are in different classes in the MPEG-7 CE1 Part B shape database.</p> "> Figure 3
<p>The solid and dashed lines are the centroid contour distance (CCD) feature vectors of the left and right shapes, respectively, in <a href="#sensors-19-00486-f002" class="html-fig">Figure 2</a>. These two curves are similar globally.</p> "> Figure 4
<p>The solid and dashed lines are the FCCD feature vectors of the left and right shapes, respectively, in <a href="#sensors-19-00486-f002" class="html-fig">Figure 2</a>. These two curves are similar, and even a large part of them overlap.</p> "> Figure 5
<p>Six pairs of shapes, each pair of which shows two globally similar shapes with only a little detail difference. The first pair in the first row and the first column shows a pencil and a bone. The second pair in the first row and second column shows a spoon and a banjo. The third pair in the second row and the first column shows a car and a person. The fourth pair in the second row and the second column shows a bottle and a fish. The fifth pair in the third row and the first column shows an octopus and an eight-pointed star. The sixth pair in the third row and the second column shows a tree and a bell.</p> "> Figure 6
<p>This figure shows the difference between two CCD feature vectors of each pair of shapes in <a href="#sensors-19-00486-f005" class="html-fig">Figure 5</a>. The subfigure of each pair is arranged in the same order as in <a href="#sensors-19-00486-f005" class="html-fig">Figure 5</a>. It can be seen that two curves of each pair are so similar, and even overlap.</p> "> Figure 7
<p>This figure shows the difference between two FCCD feature vectors of each pair of shapes in <a href="#sensors-19-00486-f005" class="html-fig">Figure 5</a>. The subfigure of each pair is arranged in the same order as in <a href="#sensors-19-00486-f005" class="html-fig">Figure 5</a>. It can be seen that two curves of each pair are so similar, and even overlap.</p> "> Figure 8
<p>The difference between multiscale CCD (MSCCD) feature vectors of two shapes in <a href="#sensors-19-00486-f002" class="html-fig">Figure 2</a> at each scale (<math display="inline"><semantics> <mrow> <mi>h</mi> <mo>=</mo> <mn>0</mn> <mo>,</mo> <mn>1</mn> <mo>,</mo> <mo>…</mo> <mo>,</mo> <mn>5</mn> </mrow> </semantics></math>). It can be seen that the difference between the two features becomes larger as <math display="inline"><semantics> <mi>h</mi> </semantics></math> increases.</p> "> Figure 9
<p>The difference between Fourier descriptor based on multiscale centroid contour distance (FMSCCD) feature vectors of two shapes in <a href="#sensors-19-00486-f002" class="html-fig">Figure 2</a> at each scale (<math display="inline"><semantics> <mrow> <mi>h</mi> <mo>=</mo> <mn>0</mn> <mo>,</mo> <mn>1</mn> <mo>,</mo> <mo>…</mo> <mo>,</mo> <mn>5</mn> </mrow> </semantics></math>). It can be seen that the difference between the two features becomes larger as <math display="inline"><semantics> <mi>h</mi> </semantics></math> increases.</p> "> Figure 10
<p>The difference between MSCCD feature vectors of each pair of shapes in <a href="#sensors-19-00486-f005" class="html-fig">Figure 5</a> when <math display="inline"><semantics> <mrow> <mi>h</mi> <mo>=</mo> <mn>3</mn> </mrow> </semantics></math>. It can be seen that the difference between each pair of MSCCD features is larger than that in the corresponding subfigure in <a href="#sensors-19-00486-f006" class="html-fig">Figure 6</a>.</p> "> Figure 11
<p>The difference between FMSCCD feature vectors of each pair of shapes in <a href="#sensors-19-00486-f005" class="html-fig">Figure 5</a> when <math display="inline"><semantics> <mrow> <mi>h</mi> <mo>=</mo> <mn>3</mn> </mrow> </semantics></math>. It can be seen that the difference between each pair of FMSCCD features is larger than that in the corresponding subfigure in <a href="#sensors-19-00486-f006" class="html-fig">Figure 6</a>.</p> "> Figure 12
<p>Some examples (a pair of shapes in each class) in MPEG-7 CE1 Part B.</p> "> Figure 13
<p>Some examples (three shapes in each class) in the Swedish Plant Leaf database.</p> "> Figure 14
<p>All shapes in the Kimia 99 database.</p> ">
Abstract
:1. Introduction
2. Methods
3. Results
3.1. On MPEG-7 CE1 Part B
3.2. On Swedish Plant Leaf
3.3. On Kimia 99
4. Discussion
5. Conclusions
Author Contributions
Acknowledgments
Conflicts of Interest
References
- Alajlan, N.; El Rube, I.; Kamel, M.S.; Freeman, G. Shape retrieval using triangle-area representation and dynamic space warping. Pattern Recognit. 2007, 40, 1911–1920. [Google Scholar] [CrossRef]
- Zahn, C.T.; Roskies, R.Z. Fourier descriptors for plane closed curves. IEEE Trans. Comput. 1972, 100, 269–281. [Google Scholar] [CrossRef]
- Wang, B.; Gao, Y.; Sun, C.; Blumenstein, M.; La Salle, J. Can walking and measuring along chord bunches better describe leaf shapes? In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA, 21–26 July 2017; pp. 6119–6128. [Google Scholar]
- Felzenszwalb, P.F.; Schwartz, J.D. Hierarchical matching of deformable shapes. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Minneapolis, MN, USA, 17–22 June 2007; pp. 1–8. [Google Scholar]
- Hu, R.-X.; Jia, W.; Zhao, Y.; Gui, J. Perceptually motivated morphological strategies for shape retrieval. Pattern Recognit. 2012, 45, 3222–3230. [Google Scholar] [CrossRef]
- Premachandran, V.; Kakarala, R. Perceptually motivated shape context which uses shape interiors. Pattern Recognit. 2013, 46, 2092–2102. [Google Scholar] [CrossRef] [Green Version]
- Yang, X.; Bai, X.; Latecki, L.J.; Tu, Z. Improving shape retrieval by learning graph transduction. In Proceedings of the European Conference on Computer Vision, Marseille, France, 12–18 October 2008; pp. 788–801. [Google Scholar]
- Ling, H.; Jacobs, D.W. Shape classification using the inner-distance. IEEE Trans. Pattern Anal. Mach. Intell. 2007, 29, 286–299. [Google Scholar] [CrossRef]
- Ling, H.; Jacobs, D.W. Using the inner-distance for classification of articulated shapes. In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, San Diego, CA, USA, 20–25 June 2005; pp. 719–726. [Google Scholar]
- Belongie, S.; Malik, J.; Puzicha, J. Shape matching and object recognition using shape contexts. IEEE Trans. Pattern Anal. Mach. Intell. 2002, 24, 509–522. [Google Scholar] [CrossRef] [Green Version]
- Zhang, D.; Lu, G. Study and evaluation of different fourier methods for image retrieval. Image Vis. Comput. 2005, 23, 33–49. [Google Scholar] [CrossRef]
- Yang, H.S.; Lee, S.U.; Lee, K.M. Recognition of 2d object contours using starting-point-independent wavelet coefficient matching. J. Vis. Commun. Image Represent. 1998, 9, 171–181. [Google Scholar] [CrossRef]
- Bai, X.; Wang, B.; Yao, C.; Liu, W.; Tu, Z. Co-transduction for shape retrieval. IEEE Trans. Image Process. 2012, 21, 2747–2757. [Google Scholar]
- Yang, X.; Koknar-Tezel, S.; Latecki, L.J. Locally constrained diffusion process on locally densified distance spaces with applications to shape retrieval. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Miami, FL, USA, 20–25 June 2009; pp. 357–364. [Google Scholar]
- Kontschieder, P.; Donoser, M.; Bischof, H. Beyond pairwise shape similarity analysis. In Proceedings of the Asian Conference on Computer Vision, Xi’an, China, 23–27 September 2009; pp. 655–666. [Google Scholar]
- Hu, R.X.; Jia, W.; Ling, H.; Huang, D. Multiscale distance matrix for fast plant leaf recognition. IEEE Trans. Image Process. 2012, 21, 4667–4672. [Google Scholar]
- Kaothanthong, N.; Chun, J.; Tokuyama, T. Distance interior ratio: A new shape signature for 2d shape retrieval. Pattern Recognit. Lett. 2016, 78, 14–21. [Google Scholar] [CrossRef]
- Fotopoulou, F.; Economou, G. Multivariate angle scale descriptor of shape retrieval. In Proceedings of the SPAMEC, Cluj-Napoca, Romania, 26–28 August 2011; pp. 105–108. [Google Scholar]
- Xie, J.; Heng, P.A. Shape matching and modeling using skeletal context. Pattern Recognit. 2008, 41, 1756–1767. [Google Scholar] [CrossRef] [Green Version]
- Amor, B.B.; Su, J.; Srivastava, A. Action recognition using rate-invariant analysis of skeletal shape trajectories. IEEE Trans. Pattern Anal. Mach. Intell. 2016, 38, 1–13. [Google Scholar] [CrossRef] [PubMed]
- Xu, G.; Fang, W. Shape retrieval using deep autoencoder learning representation. In Proceedings of the International Computer Conference on Wavelet Active Media Technology & Information Processing, Sichuan, China, 15–17 December 2017. [Google Scholar]
- Mohanty, N.; Rath, T.; Lee, A.; Manmatha, R. Learning shapes for image classification and retrieval. In Proceedings of the International Conference on Image & Video Retrieval, Singapore, 20–22 July 2005. [Google Scholar]
- Conoci, S.; Rundo, F.; Petralta, S.; Battiato, S. Advanced skin lesion discrimination pipeline for early melanoma cancer diagnosis towards PoC devices. In Proceedings of the European Conference on Circuit Theory & Design, Catania, Italy, 4–6 September 2017. [Google Scholar]
- El-ghazal, A.; Basir, O.; Belkasim, S. Farthest point distance: A new shape signature for fourier descriptors. Signal Process. Image Commun. 2009, 24, 572–586. [Google Scholar] [CrossRef]
68.21% | 68.21% | 68.21% | 68.21% | 68.21% | 68.21% | 68.21% | 68.21% | |
71.50% | 71.65% | 71.82% | 71.93% | 72.01% | 71.98% | 71.92% | 71.83% | |
72.35% | 72.74% | 73.16% | 73.48% | 73.72% | 73.80% | 73.75% | 73.76% | |
72.82% | 73.52% | 74.20% | 74.56% | 74.75% | 74.91% | 74.85% | 74.81% | |
72.10% | 73.45% | 74.40% | 75.15% | 75.45% | 75.44% | 75.40% | 75.17% | |
70.91% | 73.09% | 74.56% | 75.25% | 75.64% | 75.60% | 75.55% | 75.28% | |
70.17% | 72.91% | 74.55% | 75.32% | 75.73% | 75.70% | 75.58% | 75.28% | |
70.01% | 72.86% | 74.54% | 75.34% | 75.73% | 75.71% | 75.58% | 75.29% | |
70.00% | 72.87% | 74.55% | 75.36% | 75.73% | 75.71% | 75.58% | 75.29% |
1/6 | 2/6 | 3/6 | 4/6 | 5/6 | |
---|---|---|---|---|---|
Retrieval rate | 76.83% | 77.69% | 77.89% | 78.18% | 77.80% |
Method | Score | Matching Time (ms) |
---|---|---|
FMSCCD + FASD (ours) | 78.18% | 10.6 |
DIR [17] | 77.69% | 4.6 |
ASD&CCD [18] | 76.20% | 230.5 |
FASD | 73.56% | 5.6 |
MDM [16] | 70.46% | 30.2 |
FD-CCD [11] | 67.94% | 3.2 |
FPD [24] | 64.29% | 2.8 |
CCD [18] | 68.67% | 112.3 |
Method | 13.3% | 26.7% | 40.0% | 53.3% | 66.7% | 80.0% | 93.3% | 100.0% | Average |
---|---|---|---|---|---|---|---|---|---|
FMSCCD + FASD | 92.7% | 87.9% | 83.2% | 77.5% | 70.4% | 60.6% | 46.7% | 27.6% | 68.3% |
DIR [17] | 91.1% | 86.5% | 81.6% | 75.2% | 67.8% | 59.4% | 47.4% | 31.7% | 67.6% |
ASD&CCD [18] | 86.9% | 79.9% | 72.9% | 64.6% | 55.7% | 44.5% | 32.1% | 21.8% | 57.3% |
MDM [16] | 87.6% | 78.8% | 69.4% | 60.9% | 51.1% | 41.7% | 28.4% | 18.7% | 54.6% |
DALR [21] | 85.6% | 74.6% | 66.1% | 58.3% | 51.1% | 42.4% | 31.8% | 23.9% | 54.2% |
FD-CCD [11] | 78.4% | 69.1% | 61.4% | 54.2% | 46.4% | 37.7% | 27.1% | 17.7% | 49.0% |
© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Zheng, Y.; Guo, B.; Chen, Z.; Li, C. A Fourier Descriptor of 2D Shapes Based on Multiscale Centroid Contour Distances Used in Object Recognition in Remote Sensing Images. Sensors 2019, 19, 486. https://doi.org/10.3390/s19030486
Zheng Y, Guo B, Chen Z, Li C. A Fourier Descriptor of 2D Shapes Based on Multiscale Centroid Contour Distances Used in Object Recognition in Remote Sensing Images. Sensors. 2019; 19(3):486. https://doi.org/10.3390/s19030486
Chicago/Turabian StyleZheng, Yan, Baolong Guo, Zhijie Chen, and Cheng Li. 2019. "A Fourier Descriptor of 2D Shapes Based on Multiscale Centroid Contour Distances Used in Object Recognition in Remote Sensing Images" Sensors 19, no. 3: 486. https://doi.org/10.3390/s19030486