Int. J. Mach. Learn. & Cyber.

2020

DOI: 10.1007/s13042-020-01157-9

|View full text |Cite

|

Sign up to set email alerts

|

Fine-grained pornographic image recognition with multiple feature fusion transfer learning

¹

,

²

,

³

et al.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...

Methods Based On Deep Learning1

Transfer Learning1

Vision Attention1

Research Progress Of Sensitive Image Detection1

Citation Types

Supporting

0

Mentioning

5

Contrasting

0

Year Published

2022

2022

2023

2023

Publication Types

Select...

Article8

Book2

Relationship

Self Cite0

Independent10

Authors

Journals

Cited by 19 publications

(5 citation statements)

References 30 publications

Supporting

0

Mentioning

5

Contrasting

0

Order By: Relevance

“…A CNN network was trained for bad image recognition, and a method based on a deconvolution network was used to optimize performance in different scenarios. Li et al [24] performed pornography detection by fusing four DenseNet121 models [25]. Compared with a single DenseNet121 network, the model was improved by about 1%.…”

Section: Methods Based On Deep Learningmentioning

confidence: 99%

RepVGG-SimAM: An Efficient Bad Image Classification Method Based on RepVGG with Simple Parameter-Free Attention Module

Cai,

Qiao,

Zhang

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

With the rapid development of Internet technology, the number of global Internet users is rapidly increasing, and the scale of the Internet is also expanding. The huge Internet system has accelerated the spread of bad information, including bad images. Bad images reflect the vulgar culture of the Internet. They will not only pollute the Internet environment and impact the core culture of society but also endanger the physical and mental health of young people. In addition, some criminals use bad images to induce users to download software containing computer viruses, which also greatly endanger the security of cyberspace. Cyberspace governance faces enormous challenges. Most existing methods for classifying bad images face problems such as low classification accuracy and long inference times, and these limitations are not conducive to effectively curbing the spread of bad images and reducing their harm. To address this issue, this paper proposes a classification method (RepVGG-SimAM) based on RepVGG and a simple parameter-free attention mechanism (SimAM). This method uses RepVGG as the backbone network and embeds the SimAM attention mechanism in the network so that the neural network can obtain more effective information and suppress useless information. We used pornographic images publicly disclosed by data scientist Alexander Kim and violent images collected from the internet to construct the dataset for our experiment. The experimental results prove that the classification accuracy of the method proposed in this paper can reach 94.5% for bad images, that the false positive rate of bad images is only 4.3%, and that the inference speed is doubled compared with the ResNet101 network. Our proposed method can effectively identify bad images and provide efficient and powerful support for cyberspace governance.

“…A CNN network was trained for bad image recognition, and a method based on a deconvolution network was used to optimize performance in different scenarios. Li et al [24] performed pornography detection by fusing four DenseNet121 models [25]. Compared with a single DenseNet121 network, the model was improved by about 1%.…”

Section: Methods Based On Deep Learningmentioning

confidence: 99%

RepVGG-SimAM: An Efficient Bad Image Classification Method Based on RepVGG with Simple Parameter-Free Attention Module

Cai,

Qiao,

Zhang

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

With the rapid development of Internet technology, the number of global Internet users is rapidly increasing, and the scale of the Internet is also expanding. The huge Internet system has accelerated the spread of bad information, including bad images. Bad images reflect the vulgar culture of the Internet. They will not only pollute the Internet environment and impact the core culture of society but also endanger the physical and mental health of young people. In addition, some criminals use bad images to induce users to download software containing computer viruses, which also greatly endanger the security of cyberspace. Cyberspace governance faces enormous challenges. Most existing methods for classifying bad images face problems such as low classification accuracy and long inference times, and these limitations are not conducive to effectively curbing the spread of bad images and reducing their harm. To address this issue, this paper proposes a classification method (RepVGG-SimAM) based on RepVGG and a simple parameter-free attention mechanism (SimAM). This method uses RepVGG as the backbone network and embeds the SimAM attention mechanism in the network so that the neural network can obtain more effective information and suppress useless information. We used pornographic images publicly disclosed by data scientist Alexander Kim and violent images collected from the internet to construct the dataset for our experiment. The experimental results prove that the classification accuracy of the method proposed in this paper can reach 94.5% for bad images, that the false positive rate of bad images is only 4.3%, and that the inference speed is doubled compared with the ResNet101 network. Our proposed method can effectively identify bad images and provide efficient and powerful support for cyberspace governance.

“…By truncating the bottleneck layer of the pretrained network, the useful neurons of the reusable layer are retained to mine more classification features. In addition, migration learning can keep the training data in the same feature space or have the same distribution as the future data, avoiding the problem of overfitting [40].…”

Section: Transfer Learningmentioning

confidence: 99%

Integration of Face and Gait Recognition via Transfer Learning: A Multiscale Biometric Identification Approach

Ahmed,

Mahmood

2023

View full text Add to dashboard Cite

The ubiquity of biometric identification systems and their applications is evident in today's world. Among various biometric features, face and gait are readily obtainable and thus hold significant value. Advances in computational vision and deep learning have paved the way for the integration of these biometric features at multiple scales. This study introduces a system for biometric recognition that synergises face and gait recognition through the lens of transfer learning. Feature extraction was accomplished using Inception_v3 and DenseNet201 algorithms, while classification was performed employing machine learning algorithms such as K-Nearest Neighbours (KNN) and Support Vector Classification (SVC). A unique dataset was constructed for this research, consisting of face and gait information extracted from video clips. The findings underscore the efficacy of integrating face and gait recognition, primarily through feature and score fusion, resulting in enhanced recognition accuracy. Specifically, the Inception_v3 algorithm was found to excel in feature extraction, and SVC was superior for classification purposes. The system achieved an accuracy of 98% when feature-level fusion was performed, and 97% accuracy was observed with score fusion using Decision Trees. The results highlight the potential of transfer learning in advancing multiscale biometric recognition systems.

“…On the negative side, the authors could have tested their proposal with a larger and more challenging dataset, such as the pornography-2K dataset (composed of 2000 videos with scenes that can give a high false positive rate, such as sumo, swimming, etc.). In another recent work, Lin et al [126] presented a strategy employing transfer learning with multi-feature fusion. For sensitive-content detection, they fused four separately trained DenseNet-121 [127] models by freezing different surface layers using 120,000 images divided into five classes: porn, sexy, hentai, cartoon and neutral.…”

Section: Vision Attentionmentioning

confidence: 99%

Learning Strategies for Sensitive Content Detection

¹

,

²

,

³

et al. 2023

View full text Add to dashboard Cite

Currently, the volume of sensitive content on the Internet, such as pornography and child pornography, and the amount of time that people spend online (especially children) have led to an increase in the distribution of such content (e.g., images of children being sexually abused, real-time videos of such abuse, grooming activities, etc.). It is therefore essential to have effective IT tools that automate the detection and blocking of this type of material, as manual filtering of huge volumes of data is practically impossible. The goal of this study is to carry out a comprehensive review of different learning strategies for the detection of sensitive content available in the literature, from the most conventional techniques to the most cutting-edge deep learning algorithms, highlighting the strengths and weaknesses of each, as well as the datasets used. The performance and scalability of the different strategies proposed in this work depend on the heterogeneity of the dataset, the feature extraction techniques (hashes, visual, audio, etc.) and the learning algorithms. Finally, new lines of research in sensitive-content detection are presented.

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

customersupport@researchsolutions.com

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

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

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

Copyright © 2025 scite LLC. All rights reserved.

Made with 💙 for researchers

Part of the Research Solutions Family.