Nudity detection based on image zoning

Santos, Clayton; Santos, Eulanda M. dos; Souto, Eduardo

doi:10.1109/isspa.2012.6310454

Cited by 15 publications

(8 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This claim has clearly not been reassessed. The field has assumed that it should train computers to see hairless skin and smooth-textured, skin-colored regions of images (e.g., Santos et al 2012). Indeed, skin maps of subjects do not include head hair.…”

Section: Discussionmentioning

confidence: 99%

Training Computers to See Internet Pornography: Gender and Sexual Discrimination in Computer Vision Science

Gehl

Moyer‐Horner

Yeo

2016

Television & New Media

View full text Add to dashboard Cite

This article critically examines computer vision–based pornography filtering (CVPF), a subfield in computer science seeking to train computers on how to recognize the difference between digital pornographic images and nonpornographic images. Based on a review of 102 peer-reviewed CVPF articles, we argue that CVPF has as a whole trained computers to “see” a very specific, idealized form of pornography: pictures of lone, thin, naked women. The article supports this argument by closely reading the algorithms proposed in the CVPF literature and quantitatively analyzing the images included as illustrations of these algorithms. Drawing on pornography studies, we also compare the CVPF pornographic imagination with “noisy” pornography that exceeds computer vision. Ultimately, the article argues that this very narrow imagination of porn in CVPF reflects and reinforces larger gender and sexual inequalities in the technology industry as a whole.

show abstract

Section: Discussionmentioning

confidence: 99%

Training Computers to See Internet Pornography: Gender and Sexual Discrimination in Computer Vision Science

Gehl

Moyer‐Horner

Yeo

2016

Television & New Media

View full text Add to dashboard Cite

show abstract

“…This was not done in related studies, such as by Wang, et al [2], Karavarsamis, et al [10], Basilio, et al [6]. Santos, et al used GLCM features based on LDA classification [4].…”

Section: Discussionmentioning

confidence: 99%

“…Santos, et al [4] employed skin segmentation using GLCM features on YC b C r color space. This approach was able to separate skin objects from similar-toskin color objects based on a linear discriminant analysis (LDA) classifier.…”

Section: Introductionmentioning

confidence: 99%

Texture Analysis for Skin Classification in Pornography Content Filtering Based on Support Vector Machine

Nugroho

Rahadian

Adji

et al. 2016

J. Eng. Technol. Sci.

View full text Add to dashboard Cite

Abstract. Nowadays, the Internet is one of the most important things in a human's life. The unlimited access to information has the potential for people to gather any data related to their needs. However, this sophisticated technology also bears a bad side, for instance negative content information. Negative content can come in the form of images that contain pornography. This paper presents the development of a skin classification scheme as part of a negative content filtering system. The data are trained by grey-level co-occurrence matrices (GLCM) texture features and then used to classify skin color by support vector machine (SVM). The tests on skin classification in the skin and non-skin categories achieved an accuracy of 100% and 97.03%, respectively. These results indicate that the proposed scheme has potential to be implemented as part of a negative content filtering system.

show abstract

“…Com base nisso, foi definida a seguinte hipótese: em imagens de nudez, a região central é composta predominantemente por regiões de pele. A partir desta suposição e para introduzir informação local no vetor de características a ser gerado, foi adotado o conceito de zoneamento em imagens, utilizado em [15].…”

Section: Zoneamento Em Imagensunclassified

ANDImage: An adaptive architecture for nude detection in image

Santos¹,

Souto²,

Santos³

2015

2015 10th Iberian Conference on Information Systems and Technologies (CISTI)

Self Cite

View full text Add to dashboard Cite

Resumo -Com a popularização do acesso a Internet, instituições e pais têm encontrado sérios problemas para evitar o acesso de funcionários e crianças a conteúdos impróprios como páginas pornográficas. Uma forma de combater este tipo de abuso é empregar filtros que possibilitem a detecção de nudez em imagens digitais. Este artigo propõe ANDImage, uma arquitetura adaptativa para detecção de nudez em imagens, que utiliza um módulo de zoneamento e aplica informações de cor para descrever uma imagem de nudez. O classificador SVM é utilizado na classificação das amostras. Resultados experimentais mostram que a eficiência do modelo pode alcançar até 94,3% de acertos, em uma base composta de 2.250 imagens de nudez e 3.110 imagens neutras. Palavras Chave -imagens; detecção de nudez; zoneamento em imagens.Abstract -With the popularization of Internet access, institutions and parents have encountered serious problems to prevent access by employers and children to inappropriate content such as pornographic pages. The detection mechanisms try to circumvent or at least mitigate this problem by using of filters or mechanisms that enable the nudity detection in digital images. This paper proposes ANDImage, an adaptative architecture for nude detection in images, which uses a module zoning with color information to describe a nudity image. The SVM classifier is used in classifying samples. Experimental results show the efficiency of the model can achieve accuracy rates of up to 94.3% in a base composed of 2.250 nudity images and 3.110 neutral images.

show abstract

Nudity detection based on image zoning

Cited by 15 publications

References 16 publications

Training Computers to See Internet Pornography: Gender and Sexual Discrimination in Computer Vision Science

Training Computers to See Internet Pornography: Gender and Sexual Discrimination in Computer Vision Science

Texture Analysis for Skin Classification in Pornography Content Filtering Based on Support Vector Machine

ANDImage: An adaptive architecture for nude detection in image

Contact Info

Product

Resources

About