Laser & Optoelectronics Progress, Volume. 59, Issue 16, 1615007(2022)
Deep Forgery Detection Using CutMix Algorithm and Improved Xception Network
Fig. 1. CutMix augmented image example. (a)(b) Original samples; (c) augmented sample
Fig. 2. Sampler for unbalanced data sets
Fig. 3. Proposed network structure
Fig. 4. ROC curves and AUC values of different models on the validation set
Fig. 5. Influence of hyper-parameter α and probability p on the detection model. (a) CutMix; (b) Mixup
Fig. 6. Results of data enhancement
|
|
|
|
|
|
|
Get Citation
Copy Citation Text
Pengzhi Geng, Yunqi Tang, Hongxing Fan, Xintong Zhu. Deep Forgery Detection Using CutMix Algorithm and Improved Xception Network[J]. Laser & Optoelectronics Progress, 2022, 59(16): 1615007
Category: Machine Vision
Received: Jul. 8, 2021
Accepted: Jul. 28, 2021
Published Online: Jul. 22, 2022
The Author Email: Tang Yunqi (tangyunqi@ppsuc.edu.cn)