Chromatic Shadow Detection and Tracking for Moving Foreground Segmentation

Ivan Huerta, Michael Boelstoft Holte, Thomas B. Moeslund, Jordi Gonzàlez

Research output: Contribution to journalJournal articleResearchpeer-review

18 Citations (Scopus)

Abstract

Advanced segmentation techniques in the surveillance domain deal with shadows to avoid distortions when detecting moving objects. Most approaches for shadow detection are still typically restricted to penumbra shadows and cannot cope well with umbra shadows. Consequently, umbra shadow regions are usually detected as part of moving objects, thus affecting the performance of the final detection. In this paper we address the detection of both penumbra and umbra shadow regions. First, a novel bottom-up approach is presented based on gradient and colour models, which successfully discriminates between chromatic moving cast shadow regions and those regions detected as moving objects. In essence, those regions corresponding to potential shadows are detected based on edge partitioning and colour statistics. Subsequently (i) temporal similarities between textures and (ii) spatial similarities between chrominance angle and brightness distortions are analysed for each potential shadow region for detecting the umbra shadow regions. Our second contribution refines even further the segmentation results: a tracking-based top-down approach increases the performance of our bottom-up chromatic shadow detection algorithm by properly correcting non-detected shadows. To do so, a combination of motion filters in a data association framework exploits the temporal consistency between objects and shadows to increase the shadow detection rate. Experimental results exceed current state-of-the-art in shadow accuracy for multiple well-known surveillance image databases which contain different shadowed materials and illumination conditions.
Original languageEnglish
JournalImage and Vision Computing
Volume41
Pages (from-to)42–53
Number of pages12
ISSN0262-8856
DOIs
Publication statusPublished - 2015

Fingerprint

Color
Luminance
Textures
Lighting
Statistics

Cite this

@article{fadb012d816d44b583c1e5a7e53753da,
title = "Chromatic Shadow Detection and Tracking for Moving Foreground Segmentation",
abstract = "Advanced segmentation techniques in the surveillance domain deal with shadows to avoid distortions when detecting moving objects. Most approaches for shadow detection are still typically restricted to penumbra shadows and cannot cope well with umbra shadows. Consequently, umbra shadow regions are usually detected as part of moving objects, thus affecting the performance of the final detection. In this paper we address the detection of both penumbra and umbra shadow regions. First, a novel bottom-up approach is presented based on gradient and colour models, which successfully discriminates between chromatic moving cast shadow regions and those regions detected as moving objects. In essence, those regions corresponding to potential shadows are detected based on edge partitioning and colour statistics. Subsequently (i) temporal similarities between textures and (ii) spatial similarities between chrominance angle and brightness distortions are analysed for each potential shadow region for detecting the umbra shadow regions. Our second contribution refines even further the segmentation results: a tracking-based top-down approach increases the performance of our bottom-up chromatic shadow detection algorithm by properly correcting non-detected shadows. To do so, a combination of motion filters in a data association framework exploits the temporal consistency between objects and shadows to increase the shadow detection rate. Experimental results exceed current state-of-the-art in shadow accuracy for multiple well-known surveillance image databases which contain different shadowed materials and illumination conditions.",
author = "Ivan Huerta and Holte, {Michael Boelstoft} and Moeslund, {Thomas B.} and Jordi Gonz{\`a}lez",
year = "2015",
doi = "10.1016/j.imavis.2015.06.003",
language = "English",
volume = "41",
pages = "42–53",
journal = "Image and Vision Computing",
issn = "0262-8856",
publisher = "Elsevier",

}

Chromatic Shadow Detection and Tracking for Moving Foreground Segmentation. / Huerta, Ivan; Holte, Michael Boelstoft; Moeslund, Thomas B.; Gonzàlez, Jordi.

In: Image and Vision Computing, Vol. 41, 2015, p. 42–53.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - Chromatic Shadow Detection and Tracking for Moving Foreground Segmentation

AU - Huerta, Ivan

AU - Holte, Michael Boelstoft

AU - Moeslund, Thomas B.

AU - Gonzàlez, Jordi

PY - 2015

Y1 - 2015

N2 - Advanced segmentation techniques in the surveillance domain deal with shadows to avoid distortions when detecting moving objects. Most approaches for shadow detection are still typically restricted to penumbra shadows and cannot cope well with umbra shadows. Consequently, umbra shadow regions are usually detected as part of moving objects, thus affecting the performance of the final detection. In this paper we address the detection of both penumbra and umbra shadow regions. First, a novel bottom-up approach is presented based on gradient and colour models, which successfully discriminates between chromatic moving cast shadow regions and those regions detected as moving objects. In essence, those regions corresponding to potential shadows are detected based on edge partitioning and colour statistics. Subsequently (i) temporal similarities between textures and (ii) spatial similarities between chrominance angle and brightness distortions are analysed for each potential shadow region for detecting the umbra shadow regions. Our second contribution refines even further the segmentation results: a tracking-based top-down approach increases the performance of our bottom-up chromatic shadow detection algorithm by properly correcting non-detected shadows. To do so, a combination of motion filters in a data association framework exploits the temporal consistency between objects and shadows to increase the shadow detection rate. Experimental results exceed current state-of-the-art in shadow accuracy for multiple well-known surveillance image databases which contain different shadowed materials and illumination conditions.

AB - Advanced segmentation techniques in the surveillance domain deal with shadows to avoid distortions when detecting moving objects. Most approaches for shadow detection are still typically restricted to penumbra shadows and cannot cope well with umbra shadows. Consequently, umbra shadow regions are usually detected as part of moving objects, thus affecting the performance of the final detection. In this paper we address the detection of both penumbra and umbra shadow regions. First, a novel bottom-up approach is presented based on gradient and colour models, which successfully discriminates between chromatic moving cast shadow regions and those regions detected as moving objects. In essence, those regions corresponding to potential shadows are detected based on edge partitioning and colour statistics. Subsequently (i) temporal similarities between textures and (ii) spatial similarities between chrominance angle and brightness distortions are analysed for each potential shadow region for detecting the umbra shadow regions. Our second contribution refines even further the segmentation results: a tracking-based top-down approach increases the performance of our bottom-up chromatic shadow detection algorithm by properly correcting non-detected shadows. To do so, a combination of motion filters in a data association framework exploits the temporal consistency between objects and shadows to increase the shadow detection rate. Experimental results exceed current state-of-the-art in shadow accuracy for multiple well-known surveillance image databases which contain different shadowed materials and illumination conditions.

U2 - 10.1016/j.imavis.2015.06.003

DO - 10.1016/j.imavis.2015.06.003

M3 - Journal article

VL - 41

SP - 42

EP - 53

JO - Image and Vision Computing

JF - Image and Vision Computing

SN - 0262-8856

ER -