Related Experiment Video
Updated: Nov 27, 2025
13:19
Deep Neural Networks for Image-Based Dietary Assessment
Published on: March 13, 2021
9.7K
AD or Non-AD: A Deep Learning Approach to Detect Advertisements from Magazines
Khaled Almgren1, Murali Krishnan2, Fatima Aljanobi2
1College of Computing and Informatics, Saudi Electronic University, Riyadh 11673, Saudi Arabia.
Entropy (Basel, Switzerland)
|December 3, 2020
Summary
This study uses deep learning to identify magazine advertisements from scanned images. Convolutional neural networks accurately distinguish ads from articles, improving marketing strategies.
Area of Science:
- Computer Science
- Artificial Intelligence
- Multimedia Analysis
Background:
- Deep learning advancements drive multimedia data processing research.
- Computer vision tasks like image recognition are enhanced by deep learning.
Purpose of the Study:
- To analyze visual features for detecting advertising images in scanned magazines.
- To classify images as advertisements or non-advertisements (articles).
- To enhance marketing strategies through accurate ad classification.
Main Methods:
- Analysis of visual image features.
- Application of convolutional neural networks (CNNs) for image classification.
Main Results:
- The proposed CNN approach achieved high accuracy in distinguishing advertisements from articles.
- The method outperformed existing classifiers and related work.
Conclusions:
- Deep learning, specifically CNNs, is effective for automated advertisement detection in magazines.
- This technology can significantly improve marketing strategy development and execution.
