Subsign detection with region-growing from contrasted seeds

Puthon, Anne-Sophie; Moutarde, Fabien; Nashashibi, Fawzi

doi:10.1109/itsc.2012.6338826

Cited by 4 publications

(8 citation statements)

References 9 publications

(8 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Puthon et al [11] in 2012 have proposed and tested a new original and performant technique , based on region-growing from…”

Section: Recognition Of Traffic Sumentioning

confidence: 99%

“…This "dark proportion" is computed using the same binarization technique as [11] for extracting contrasted seed from which is grown detected region. A morphological reconstruction used to find highly contrasted pixels in the image, then "dark" pixels are selected as pixels with intensity included in the [t 1 ;t 2 ] interval, with t 1 =µ+0.5σ and the mean of pixel values, and σ Examples of pixels considered as "dark" by this method are shown on second column of Figure 6 We nickname pHoG-Dp this new proposed descriptor concatenating the standard pHoG with the "darkness proportion" information computed on the same pyramid.…”

Section: Of the Number L Of Levels Of The Pyramid: L=2 Shows Much Betmentioning

confidence: 99%

“…From the application point of view, it is essential to also test performance of our recognition approach when applied on the candidate sub-sign regions found by our sub detection technique. As described in more details in [11], our detection method is based on region highly contrasted seeds (hopefully lying on border of black pictogram or text of the sub-sign). Figure 10 shows, for our global chain consisting of detection by region-growing from contrasted seeds followed by recognition with SVM applied on p f(FPPP) curves separately for eac (FPPP = False Positive Per Positive).…”

Section: Recall Curves Separately For Each Meta-class Computed On Tementioning

confidence: 99%

“…From the application point of view, it is essential to also test formance of our recognition approach when applied on found by our sub-sign detection technique. As described in more details in [11], our detection method is based on region-growing applied from ully lying on border of black for our global chain consisting of detection growing from contrasted seeds followed by cognition with SVM applied on pHoG-Dp, the recall = separately for each meta-class obtained Positive). One can see that maximum recall is rather low (~45%) for Arrow category, rate for this type of subsigns.…”

Section: Confusion Matrixmentioning

confidence: 99%

See 3 more Smart Citations

Recognition of supplementary signs for correct interpretation of traffic signs

Puthon

Moutarde

Nashashibi

2013

2013 IEEE Intelligent Vehicles Symposium (IV)

Self Cite

View full text Add to dashboard Cite

Abstract-Traffic Sign Recognition (TSR) is now relatively well-handled by several approaches. However, traffic signs are often completed by one (or several) supplementary placed below. They are essential for correct interpretation of main sign, as they specify its applicability scope. difficulty of supplementary sub-sign recognition potentially infinite number of classes, as nearly any can be written on them. In this paper, we propose and evaluate a hierarchical approach for recognition of supplementary signs, in which the "meta-class" of the sub-sign (Arrow, P Text or Mixed) is first determined. The classification is based on the pyramid-HOG feature, completed by dark area proportion measured on the same pyramid. large database of images with and without supp shows that the classification accuracy of our approach 95% precision and recall. When used on output of our sub specific detection algorithm, the global correct detection and recognition rate is 91%.

show abstract

“…Puthon et al [11] in 2012 have proposed and tested a new original and performant technique , based on region-growing from…”

Section: Recognition Of Traffic Sumentioning

confidence: 99%

Section: Of the Number L Of Levels Of The Pyramid: L=2 Shows Much Betmentioning

confidence: 99%

Section: Recall Curves Separately For Each Meta-class Computed On Tementioning

confidence: 99%

Section: Confusion Matrixmentioning

confidence: 99%

See 2 more Smart Citations

Recognition of supplementary signs for correct interpretation of traffic signs

Puthon

Moutarde

Nashashibi

2013

2013 IEEE Intelligent Vehicles Symposium (IV)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Puthon et al [11] in 2012 have proposed and tested a new original and performant technique , based on region-growing from…”

Section: Recognition Of Traffic Sumentioning

confidence: 99%

Recognition of supplementary signs for correct interpretation of traffic signs

Puthon

Moutarde

Nashashibi

2013

2013 IEEE Intelligent Vehicles Symposium Workshops (IV Workshops)

Self Cite

View full text Add to dashboard Cite

Traffic Sign Recognition (TSR) is now relatively well-handled by several approaches. However, traffic signs are often completed by one (or several) supplementary placed below. They are essential for correct interpretation of main sign, as they specify its applicability scope. difficulty of supplementary sub-sign recognition potentially infinite number of classes, as nearly any can be written on them. In this paper, we propose and evaluate a hierarchical approach for recognition of supplementary signs, in which the "meta-class" of the sub-sign (Arrow, P Text or Mixed) is first determined. The classification is based on the pyramid-HOG feature, completed by dark area proportion measured on the same pyramid. large database of images with and without supp shows that the classification accuracy of our approach 95% precision and recall. When used on output of our sub specific detection algorithm, the global correct detection and recognition rate is 91%.

show abstract