Content Bias in Deep Learning Image Age Approximation: A new Approach Towards better Explainability
Preprint
Abstract
In the context of temporal image forensics, it is not evident that a neural network, trained on images from different time-slots (classes), exploits solely image age related features. Usually, images taken in close temporal proximity (e.g., belonging to the same age class) share some common content properties. Such content bias can be exploited by a neural network. In this work, a novel approach is proposed that evaluates the influence of image content. This approach is verified using synthetic images (where content bias can be ruled out) with an age signal embedded. Based on the proposed approach, it is shown that a deep learning approach proposed in the context of age classification is most likely highly dependent on the image content. As a possible countermeasure, two different models from the field of image steganalysis, along with three different preprocessing techniques to increase the signal-to-noise ratio (age signal to image content), are evaluated using the proposed method.Keywords
All Related Versions
This publication has 14 references indexed in Scilit:
- Digital image steganography survey and investigation (goal, assessment, method, development, and dataset)Signal Processing, 2022
- A machine learning-based approach for picture acquisition timeslot prediction using defective pixelsForensic Science International: Digital Investigation, 2021
- Perturbation-based methods for explaining deep neural networks: A surveyPattern Recognition Letters, 2021
- Temporal Image Forensic Analysis for Picture Dating with Deep Learning2020
- The ‘Northumbria Temporal Image Forensics’ Database: Description and Analysis2020
- Deep Residual Network for Steganalysis of Digital ImagesIEEE Transactions on Information Forensics and Security, 2018
- Constrained Convolutional Neural Networks: A New Approach Towards General Purpose Image Manipulation DetectionIEEE Transactions on Information Forensics and Security, 2018
- On the SPN Estimation in Image Forensics: A Systematic Empirical EvaluationIEEE Transactions on Information Forensics and Security, 2016
- Sensor Pattern Noise Estimation Based on Improved Locally Adaptive DCT Filtering and Weighted Averaging for Source Camera Identification and VerificationIEEE Transactions on Information Forensics and Security, 2016
- Rich Models for Steganalysis of Digital ImagesIEEE Transactions on Information Forensics and Security, 2012