An adult image recognizing algorithm based on naked body detection

Abstract

The nude image is an important part in adult images spreading on the Internet. A new recognizing algorithm of nude image based on the navel and body feature is proposed in this paper. The algorithm considers the naked body which is composed by trunk, limb and face as the object to be recognized. Body is taken as a combination of predefined key rectangles. The intersection of the key rectangles is the navel. Firstly, the algorithm recognizes the location of navel in an image. Then it constructs the feature vector of key rectangles. Finally, it classifies the features by Forward Propagation neural network to determine whether there is a naked body in the image or not. Experiments show that the algorithm can recognize nude images accurately, and the recognizing speed can satisfy the requirement of practical systems.

Nudity detection based on image zoning

“…It is not surprising that this idea has already been investigated in the literature [7,11,13]. These previous works, as well as other approaches, have one characteristic in common: they employ filters to detect skin as a first level of nudity classification processes.…”

An Automated Anti-Pornography System Using a Skin Detector Based on Artificial Intelligence: A Review

“…It is used to train a classi¯er on these features to determine whether nudity or pornography for any given image is present. The labeled image classi¯cation can be categorized using di®erent machine learning to identify pornographic image based on skin color and erotic part features, such as support vector machine as adopted by Duan et al, 36 Lin et al, 69 Zeng et al, 134 Staniszewski et al, 107 Zhu et al, 141 Dhakal et al, 34 Zhao and Cai, 136 Bouirouga et al 15 Decision tree classi¯er as adopted by Shen et al, 97 Shen et al, 98 Shiwei et al 101 Arti¯cial neural network as adopted by Kim et al, 61 Kim et al, 62 Zheng et al, 140 Lefebvre et al, 68 Sayed et al, 91 Wang et al, 117 and Jang et al 52 According to Subramanian et al, 108 the naïve Bayesian and neural network show promising and better techniques that can be applied to¯ltering spam, for instance the pornography spam. Duan et al 36 The authors proposed a skin color model (SCM) using a combination of YUV and YIQ color spaces.…”

Objectionable Image Detection in Cloud Computing Paradigm-a Review

“…The detection rate of this approach was 75.6%. Wang et al (2009) depicted a new method for identifying adult images based on naked body detection. In this work they considered navel and body features, which are composed by trunk, limb and face.…”