TY - GEN
T1 - Automatic detection of acne scars
T2 - 1st IEEE-EMBS Conference on Point-of-Care Healthcare Technologies, PHT 2013
AU - Dey, Biman Chandra
AU - Nirmal, B.
AU - Galigekere, Ramesh R.
PY - 2013/3/8
Y1 - 2013/3/8
N2 - Acne scars are of great cosmetic concern, and there are several methods of treatment towards reducing the appearance of scars. Automatic detection of acne scar-pixels from digital color images would be helpful in quantitative assessment of the success of treatment. This paper addresses detection of acne scar-pixels based on color image processing. The RGB model is used to representing the data. Pixels from the background (skin) and from the lesions of interest (acne scars) were recorded from the images of 7 subjects, to build a knowledge-base i.e., clusters associated with the skin and acne scars, respectively. The clusters were found to be fairly distinct in the RGB space. Consequently, classification (segmentation) is performed by minimum-distance- rule in the RGB space, by using Mahalanobis distance (MD). We have also implemented Bayes' method. The results have been validated with respect to the ground-truth extracted by manual segmentation of scars. The classifier based on MD performs better than that based on Bayes, with the average values of sensitivity and specificity of the former being 90.36 and 93.82, respectively.
AB - Acne scars are of great cosmetic concern, and there are several methods of treatment towards reducing the appearance of scars. Automatic detection of acne scar-pixels from digital color images would be helpful in quantitative assessment of the success of treatment. This paper addresses detection of acne scar-pixels based on color image processing. The RGB model is used to representing the data. Pixels from the background (skin) and from the lesions of interest (acne scars) were recorded from the images of 7 subjects, to build a knowledge-base i.e., clusters associated with the skin and acne scars, respectively. The clusters were found to be fairly distinct in the RGB space. Consequently, classification (segmentation) is performed by minimum-distance- rule in the RGB space, by using Mahalanobis distance (MD). We have also implemented Bayes' method. The results have been validated with respect to the ground-truth extracted by manual segmentation of scars. The classifier based on MD performs better than that based on Bayes, with the average values of sensitivity and specificity of the former being 90.36 and 93.82, respectively.
UR - https://www.scopus.com/pages/publications/84874555171
UR - https://www.scopus.com/inward/citedby.url?scp=84874555171&partnerID=8YFLogxK
U2 - 10.1109/PHT.2013.6461325
DO - 10.1109/PHT.2013.6461325
M3 - Conference contribution
AN - SCOPUS:84874555171
SN - 9781467327664
T3 - IEEE EMBS Special Topic Conference on Point-of-Care (POC) Healthcare Technologies: Synergy Towards Better Global Healthcare, PHT 2013
SP - 224
EP - 227
BT - IEEE EMBS Special Topic Conference on Point-of-Care (POC) Healthcare Technologies
Y2 - 16 January 2013 through 18 January 2013
ER -