2016

DOI: 10.1016/j.neucom.2015.09.135

|View full text |Cite

|

Sign up to set email alerts

|

Pornographic image detection utilizing deep convolutional neural networks

¹

,

²

,

³

et al.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...

Literature Review2

Introduction2

I1

Citation Types

Supporting

1

Mentioning

31

Contrasting

0

Unclassified

2

Year Published

2017

2017

2023

2023

Publication Types

Select...

Book5

Article2

Other1

Relationship

Self Cite0

Independent8

Authors

Journals

Cited by 64 publications

(34 citation statements)

References 23 publications

Supporting

1

Mentioning

31

Contrasting

0

Unclassified

2

Order By: Relevance

“…Subsequently the mask is passed through a filter based on the geometric shape of the body. Several other methods have been proposed in this field [6,7,8]. For example, Nian, et al [6] propose a method for pornographic image detection where a dataset with more than 13 000 pornographic images with adults and over 35 000 normal images was used for training a deep convolutional neural network.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Several other methods have been proposed in this field [6,7,8]. For example, Nian, et al [6] propose a method for pornographic image detection where a dataset with more than 13 000 pornographic images with adults and over 35 000 normal images was used for training a deep convolutional neural network. It should be noted that all techniques and datasets related to pornography reported in the literature refer to pornographic material displaying adults rather than children.…”

Section: Literature Reviewmentioning

confidence: 99%

“…However, the development of highly accurate CPD's is inhibited due to the unavailability of image datasets suitable for the necessary training and testing of CPD's. In the case of child pornographic material, the creation, possession, and dissemination of real data is illegal [2,3] and unethical hence unlike other related cases [4,5,6,7,8] it is impossible to use real images for experimentation. Furthermore, within-class variability of images containing child pornographic material is huge making the process of machine-based classification a highly challenging task.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

On the detection of images containing child-pornographic material

¹

,

²

,

³

2017

2017 24th International Conference on Telecommunications (ICT)

View full text Add to dashboard Cite

The vast increase in the use of social networks and other internet-based communication tools contributed to the escalation of the problem of exchanging child pornographic material over the internet. The problem of dissemination of child pornographic material could be addressed using dedicated image detection algorithms capable of rating the inappropriateness level of images exchanged through computer networks so that images with inappropriate content involving children are blocked. However, the complexity of the image detection task coupled with the nonexistence of suitable datasets, inhibit the development of efficient algorithms that can be used for detecting offensive images containing children. To deal with the problem, we propose a methodological approach that can be used for supporting the development of child pornography detectors through the generation of synthetic datasets and through the decomposition of the task into a set of simpler tasks for which training data is available. Preliminary results show the promise of the proposed approach.

“…Subsequently the mask is passed through a filter based on the geometric shape of the body. Several other methods have been proposed in this field [6,7,8]. For example, Nian, et al [6] propose a method for pornographic image detection where a dataset with more than 13 000 pornographic images with adults and over 35 000 normal images was used for training a deep convolutional neural network.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Several other methods have been proposed in this field [6,7,8]. For example, Nian, et al [6] propose a method for pornographic image detection where a dataset with more than 13 000 pornographic images with adults and over 35 000 normal images was used for training a deep convolutional neural network. It should be noted that all techniques and datasets related to pornography reported in the literature refer to pornographic material displaying adults rather than children.…”

Section: Literature Reviewmentioning

confidence: 99%

“…However, the development of highly accurate CPD's is inhibited due to the unavailability of image datasets suitable for the necessary training and testing of CPD's. In the case of child pornographic material, the creation, possession, and dissemination of real data is illegal [2,3] and unethical hence unlike other related cases [4,5,6,7,8] it is impossible to use real images for experimentation. Furthermore, within-class variability of images containing child pornographic material is huge making the process of machine-based classification a highly challenging task.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

On the detection of images containing child-pornographic material

¹

,

²

,

³

2017

2017 24th International Conference on Telecommunications (ICT)

View full text Add to dashboard Cite

The vast increase in the use of social networks and other internet-based communication tools contributed to the escalation of the problem of exchanging child pornographic material over the internet. The problem of dissemination of child pornographic material could be addressed using dedicated image detection algorithms capable of rating the inappropriateness level of images exchanged through computer networks so that images with inappropriate content involving children are blocked. However, the complexity of the image detection task coupled with the nonexistence of suitable datasets, inhibit the development of efficient algorithms that can be used for detecting offensive images containing children. To deal with the problem, we propose a methodological approach that can be used for supporting the development of child pornography detectors through the generation of synthetic datasets and through the decomposition of the task into a set of simpler tasks for which training data is available. Preliminary results show the promise of the proposed approach.

“…Oleh karena itu, penelitian tentang deteksi kulit masih dikembangkan untuk menangani masalah tersebut. Dalam penelitian deteksi kulit, [2] dan [3] menggunakan ektrasi kulit untuk mendeteksi wajah, [4] dan [5] mendeteksi citra pornografi, sedangkan [6] mendeteksi wajah, mata, hidung, dan tangan kemudian melacak bagian-bagian tubuh tersebut.…”

Section: Iunclassified

Model Warna HSCbCrAB untuk Deteksi Kulit Menggunakan PCA-kNN

2017

View full text Add to dashboard Cite

Abstract--Skin detection is a process to determine a region whether including skin or not skin. Some digital cameras produce RGB images. In many cases skin detection is transformed from RGB to other color spaces, such as HSV, YCbCr, and CIELAB. Some color spaces have two separate components, namely luminance and chrominance components, whereas the color of the human skin is more commonly in the chrominance component. In this paper, we undertake skin detection studies using chrominance components from the color spaces of HSV, YCbCr, and CIELAB, under the name HSCbCrAB. We use PCA to reduce dimensions and NN as classifier. The results of the study showed good performance in HSCbCrAB color space for skin detection.Keywords -skin detection, pca-knn, hsv, ycbcr, cielab Abstrak-Deteksi kulit merupakan suatu proses untuk menentukan suatu wilayah apakah termasuk kulit atau bukan kulit. Beberapa kamera digital menghasilkan citra RGB. Dalam berbagai kasus deteksi kulit dilakukan transformasi dari RGB ke ruang warna lainnya, seperti HSV, YCbCr, dan CIELAB. Beberapa ruang warna memiliki dua komponen yang terpisah, yaitu komponen luminan dan krominan, sedangkan warna kulit manusia lebih sering berada pada komponen krominan. Dalam paper ini, kami melakukan penelitian deteksi kulit menggunakan komponen krominan dari ruang warna HSV, YCbCr, dan CIELAB, dengan nama HSCbCrAB. Kami menggunakan PCA untuk mengurangi dimensi dank NN sebagai klasifier. Hasil dari penelitian menunjukkan performa yang bagus pada ruang warna HSCbCrAB untuk deteksi kulit.Kata Kunci -deteksi kulit, pca-knn, hsv, ycbcr, cielab I. PENDAHULUAN Banyak penelitian yang tertarik dan tertantang dengan topik deteksi kulit. Topik ini merupakan langkah awal untuk memisahkan bagian antara aktor dengan bagian latar, pelacakan pergerakan aktor, deteksi citra pornografi, pengenalan wajah, identifikasi umur, augmented reality, dan masih banyak lagi lainnya. Kakumanu et al. [1] menyatakan bahwa penelitian tentang deteksi kulit mengalami masalah besar, seperti pencahayaan, karakteristik kamera, suku/etnis, karakteristik individu, dan faktor lain, seperti makeup, gaya rambut, kacamata, keringat, dan warna latar yang menyerupai warna kulit manusia. Oleh karena itu, penelitian tentang deteksi kulit masih dikembangkan untuk menangani masalah tersebut. Suatu citra dapat dipisahkan antara piksel kulit dan bukan kulit dengan menggunakan klasifikasi, seperti teknik threshold, Bayesian Network, dan k Nearest Neighbor (k-NN). Teknik threshold memerlukan waktu komputasi yang sangat cepat karena hanya membutuhkan beberapa peraturan klasifikasi [7], akan tetapi teknik ini memiliki kelemahan, karena kurang fleksibel akibat dari peraturan yang bersifat tetap. Dengan jumlah data latih yang sangat besar, waktu komputasi yang dibutuhkan oleh Neural Network lebih besar daripada Bayesian Network.[8] melakukan kombinasi antara Back-Propagation Neural Network (BPNN) dengan Bayesian Network. Bayesian Network digunakan untuk mendapatkan akurasi dan kehandalan yang lebih baik meskipun dalam kondisi...

“…All these have made recognizing and filtering out pornographic images a rising concern nowadays. While considerable progress has been made in the last two decades [11], [31], [9], [24], [18], [27], [28] vision. There are large intra-class variations in pornographic images, due to the changes in background, scale, scenario, and human pose.…”

Section: Introductionmentioning

confidence: 99%

Pornographic Image Recognition via Weighted Multiple Instance Learning

Jin¹,

²

,

³

2019

IEEE Trans. Cybern.

View full text Add to dashboard Cite

In the era of Internet, recognizing pornographic images is of great significance for protecting children's physical and mental health. However, this task is very challenging as the key pornographic contents (e.g., breast and private part) in an image often lie in local regions of small size. In this paper, we model each image as a bag of regions, and follow a multiple instance learning (MIL) approach to train a generic region-based recognition model. Specifically, we take into account the regions' degree of pornography, and make three main contributions. First, we show that based on very few annotations of the key pornographic contents in a training image, we can generate a bag of properly sized regions, among which the potential positive regions usually contain useful contexts that can aid recognition. Second, we present a simple quantitative measure of a region's degree of pornography, which can be used to weigh the importance of different regions in a positive image. Third, we formulate the recognition task as a weighted MIL problem under the convolutional neural network framework, with a bag probability function introduced to combine the importance of different regions. Experiments on our newly collected large scale dataset demonstrate the effectiveness of the proposed method, achieving an accuracy with 97.52% true positive rate at 1% false positive rate, tested on 100K pornographic images and 100K normal images.

scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.

Contact Info

hi@scite.ai

10624 S. Eastern Ave., Ste. A-614

Henderson, NV 89052, USA

Product

Browser Extension Assistant by scite Citation Statement Search Reference Check Visualizations Dashboards Explore Journals Explore Organizations Explore Funders Embedding Badge Embedding Citation Search Pricing

Resources

Blog Help & FAQ Accessibility Statement API Terms For Universities & Governments For Researchers For Publishers For Corporate, Pharma & Enterprise Author Marketing Become an Affiliate Get an organization trial or quote scite Data & Services

About

News & Press Careers Read our Paper Coverage

Blog Terms and Conditions API Terms Privacy Policy Contact Cookie Preferences Do Not Sell or Share My Personal Information

Copyright © 2024 scite LLC. All rights reserved.

Made with 💙 for researchers

Part of the Research Solutions Family.