Skin detection is the process of discriminating skin and non-skin regions in a digital image and it is widely used in several applications ranging from hand gesture analysis to track body parts and face detection. Skin detection is a challenging problem which has drawn extensive attention from the research community in the context of expert and intelligent systems, nevertheless a fair comparison among approaches is very difficult due to the lack of a common benchmark and a unified testing protocol. In the recent era, the success of deep convolutional neural network (CNN) has strongly influenced the field of image segmentation and gave us various successful models to date. Anyway, due to the lack of large ground truth for skin detection only few works have addressed the skin detection problem using CNN models. In this work, we investigate the most recent researches in this field, and we propose a fair comparison among approaches using several different datasets. The major contributions of this work are (i) an exhaustive literature review of skin color detection approaches and a comparison of approaches that can be useful to researchers and practitioners to select the most suitable method for their application, (ii) the collection and examination of many datasets with ground truth for skin detection that can be useful to produce a training set for CNN models, (iii) a framework to evaluate and combine different skin detector approaches, whose source code is made freely available for future research, and (iv) an extensive experimental comparison among several recent methods which have also been used to define an ensemble that works well in many different problems. Experiments are carried out in 10 different datasets including more than 10,000 labelled images: experimental results confirm that the best method here proposed obtains a very good performance with respect to other stand-alone approaches, without requiring ad hoc parameter tuning. A MATLAB version of the framework for testing and of the methods proposed in this paper will be freely available from https://github.com/LorisNanni.
Lumini A., Nanni L. (2020). Fair comparison of skin detection approaches on publicly available datasets. EXPERT SYSTEMS WITH APPLICATIONS, 160, 1-11 [10.1016/j.eswa.2020.113677].
Fair comparison of skin detection approaches on publicly available datasets
Lumini A.
Primo
;
2020
Abstract
Skin detection is the process of discriminating skin and non-skin regions in a digital image and it is widely used in several applications ranging from hand gesture analysis to track body parts and face detection. Skin detection is a challenging problem which has drawn extensive attention from the research community in the context of expert and intelligent systems, nevertheless a fair comparison among approaches is very difficult due to the lack of a common benchmark and a unified testing protocol. In the recent era, the success of deep convolutional neural network (CNN) has strongly influenced the field of image segmentation and gave us various successful models to date. Anyway, due to the lack of large ground truth for skin detection only few works have addressed the skin detection problem using CNN models. In this work, we investigate the most recent researches in this field, and we propose a fair comparison among approaches using several different datasets. The major contributions of this work are (i) an exhaustive literature review of skin color detection approaches and a comparison of approaches that can be useful to researchers and practitioners to select the most suitable method for their application, (ii) the collection and examination of many datasets with ground truth for skin detection that can be useful to produce a training set for CNN models, (iii) a framework to evaluate and combine different skin detector approaches, whose source code is made freely available for future research, and (iv) an extensive experimental comparison among several recent methods which have also been used to define an ensemble that works well in many different problems. Experiments are carried out in 10 different datasets including more than 10,000 labelled images: experimental results confirm that the best method here proposed obtains a very good performance with respect to other stand-alone approaches, without requiring ad hoc parameter tuning. A MATLAB version of the framework for testing and of the methods proposed in this paper will be freely available from https://github.com/LorisNanni.File | Dimensione | Formato | |
---|---|---|---|
main.pdf
accesso aperto
Tipo:
Postprint
Licenza:
Licenza per Accesso Aperto. Creative Commons Attribuzione - Non commerciale - Non opere derivate (CCBYNCND)
Dimensione
1.43 MB
Formato
Adobe PDF
|
1.43 MB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.