Feature-based active contour model and occluding object detection

Sara Memar; Riadh Ksantini; Boubakeur Boufama

doi:10.1364/JOSAA.33.000648

Journal of the Optical Society of America A
Vol. 33,
Issue 4,
pp. 648-662
(2016)
•https://doi.org/10.1364/JOSAA.33.000648

Feature-based active contour model and occluding object detection

Sara Memar, Riadh Ksantini, and Boubakeur Boufama

Not Accessible

Your library or personal account may give you access

Get PDF
Email
Share
Get Citation
Copy Citation Text
Sara Memar, Riadh Ksantini, and Boubakeur Boufama, "Feature-based active contour model and occluding object detection," J. Opt. Soc. Am. A 33, 648-662 (2016)

Export Citation
- BibTex
- Endnote (RIS)
- HTML
- Plain Text
Citation alert
Save article

Abstract

This paper presents a method for image segmentation and object detection. The proposed strategy consists of two major stages. The first one corresponds to image segmentation, which is based on the active contour model (ACM) algorithm, using an automatic selection of the best candidate features among gradient, polarity, and depth, coupled with a combination of them by the kernel support vector machine (KSVM). Although existing techniques, such as the ones based on ACM, perform well in the single-object case and non-noisy environments, these techniques fail when the scene consists of multiple occluding objects, with possibly similar colors. Thus, the second stage corresponds to the identification of salient and occluded objects based on the fuzzy C-mean algorithm (FCM). In this stage, the depth is included as another clue that allows us to estimate the cluster number and to make the clustering process more robust. In particular, complex occlusions can be handled this way, and the objects can be properly segmented and identified. Experimental results on real images and on several standard datasets have shown the success and effectiveness of the proposed method.

Full Article | PDF Article

More Like This

Cauchy graph embedding based diffusion model for salient object detection

Yihua Tan, Yansheng Li, Chen Chen, Jin-gang Yu, and Jinwen Tian
J. Opt. Soc. Am. A 33(5) 887-898 (2016)

Collaborative multicue fusion using the cross-diffusion process for salient object detection

Jin-Gang Yu, Changxin Gao, and Jinwen Tian
J. Opt. Soc. Am. A 33(3) 404-415 (2016)

Salient object detection fusing global and local information based on nonsubsampled contourlet transform

Dongmei Liu, Faliang Chang, and Chunsheng Liu
J. Opt. Soc. Am. A 33(8) 1430-1441 (2016)

Previous Article Next Article

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Figures (8)

You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Tables (11)

You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Equations (13)

You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Pixels of the Image	Feature Value	True Edge and Nonedge Vectors	Euclidean Distance Score
$P_{1} = {F_{g}, F_{D}}$	$P_{1} = [0.1, 0.2]$	True $edge = [0, 0]$	0.22
		True $nonedge = [1, 1]$	1.2
$P_{20} = {F_{g}, F_{D}}$	$P_{20} = [0.75, 0.9]$	True $edge = [0, 0]$	1.17
		True $nonedge = [1, 1]$	0.32

Image ID	$I_{texture}$	Threshold	Number of Peaks in Depth Histogram	Selected Feature
Image 43	0.42	0.44	3	${P D}$
Image 170	0.71	0.54	2	${G D}$
Image 205	0.59	0.46	4	${G D}$
Image 294	0.20	0.37	4	${P D}$
Image 344	0.70	0.45	1	${G}$
Image 455	0.41	0.45	1	${P}$
Image 800	0.18	0.38	1	${P}$
Image 835	0.59	0.51	1	${G}$
Image 841	0.33	0.40	2	${P D}$
Image 705	0.46	0.43	2	${G D}$

Image ID	$I_{texture}$	Threshold	Number of Peaks in Depth Histogram	Selected Feature
Scene-1	0.27	0.45	2	${P D}$
Scene-4	0.29	0.46	2	${P D}$
Scene-7	0.45	0.42	3	${G D}$
Scene-9	0.43	0.38	2	${G D}$
Scene-12	0.47	0.38	2	${G D}$
Scene-14	0.54	0.47	2	${G D}$

Image ID	[48]	[47]	[5] RD– GAC	[5] RD–CV	Proposed Method
Image 43	0	0.888	0.962	0.963	0.913
Image 170	0.032	0.682	0.734	0	0.747
Image 205	0.092	0.756	0.766	0.044	0.797
Image 294	0.615	0.938	0.607	0.490	0.911
Image 344	0.882	0.913	0.869	0.763	0.924
Image 455	0	0.928	0.977	0.843	0.940
Image 800	0.215	0.794	0.505	0.689	0.885
Image 835	0.195	0.834	1	0.946	0.985
Image 841	0	0.277	0.962	0.072	0.807
Image 705	0.444	0.771	0.923	0.342	0.941
Average	0.247	0.778	0.830	0.515	0.885

Image ID	[48]	[47]	[5] RD–GAC	[5] RD–CV	Proposed Method
Scene-1	0	0.794	0.693	0.579	0.699
Scene-4	0	0.587	0.571	0.697	0.672
Scene-7	0.804	0.671	0.608	0.401	0.854
Scene-9	0	0.758	0.895	0.142	0.899
Scene-12	0.9	0.957	0.737	0	0.913
Scene-14	0.873	0.803	0.884	0.565	0.894
Average	0.429	0.761	0.731	0.397	0.821

	[48]		[47]		RD–GAC		RD–CV		Proposed Method
Image ID	$α$	itr Number	$σ$	itr Number	$Δ t_{1}$	itr Number	$Δ t_{1}$	itr Number	$α$	itr Number
Image 43	36	100	10	400	0.9	300	0.1	150	1	260
Image 170	80	200	10	600	0.8	300	0.3	150	1.5	260
Image 205	80	200	10	300	0.9	300	0.3	150	0.8	285
Image 294	80	200	10	600	0.7	300	0.4	300	1.3	235
Image 344	60	150	10	200	0.9	300	0.4	70	1.3	160
Image 455	100	200	10	200	0.9	300	0.1	450	1.3	235
Image 800	100	150	10	400	0.9	300	0.1	600	1	155
Image 835	100	150	10	600	0.9	300	0.1	100	1.5	235
Image 841	100	250	10	400	0.7	300	0.4	200	2.5	260
Image 705	60	100	10	600	0.9	300	0.4	200	6	260

Image ID	[48]	[47]	[5] RD–GAC	[5] RD–CV	Proposed Method
Computer	0	0.699	0.656	0.603	0.688
Cone	0.009	0.809	0.846	0.833	0.879
Laundry	0	0.703	0.779	0.068	0.923
Monopoly	0.776	0.404	0.743	0.207	0.896
Star	0	0.804	0.838	0.555	0.959
Teddy	0.052	0.789	0.810	0.808	0.920
Tsukuba	0.754	0.882	0.966	0.604	0.949
Car	0.391	0.661	0.825	0.513	0.859
Chair	0.519	0.669	0.744	0.405	0.842
Flower	0.310	0.631	0.627	0.328	0.823
Average	0.281	0.704	0.783	0.492	0.873

Image ID	[48]	[47]	[5] RD–GAC	[5] RD–CV	Proposed Method
Computer	0	0.699	0.656	0.603	0.688
Cone	0.009	0.809	0.846	0.833	0.879
Laundry	0	0.703	0.779	0.068	0.923
Monopoly	0.776	0.404	0.743	0.207	0.896
Star	0	0.804	0.838	0.555	0.959
Teddy	0.052	0.789	0.810	0.808	0.920
Tsukuba	0.754	0.882	0.966	0.604	0.949
Car	0.391	0.661	0.825	0.513	0.859
Chair	0.519	0.669	0.744	0.405	0.842
Flower	0.310	0.631	0.627	0.328	0.823
Average	0.281	0.704	0.783	0.492	0.873

Abstract

Cited By

Figures (8)

Tables (11)

Equations (13)

Journal of the Optical Society of America A