Optimized Dropkey-Based Grad-CAM: Toward Accurate Image Feature Localization
Sensors · 10.10.2023 · 23 (20) · 8351
DOI: 10.3390/s23208351 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, online erschienen am 10.10.2023. Verfasst von Yiwei Liu und 7 weiteren Autor:innen. Beteiligte Einrichtung: College of Mechanical and Electrical Engineering, Nanjing Forestry University, Nanjing 210037, China. Die Arbeit wurde bisher 21-mal zitiert (Crossref) und nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Postdoctoral Science Foundation of China, Fundamental Research Funds for the Central Universities.
Abstract (Originalsprache)
Regarding the interpretable techniques in the field of image recognition, Grad-CAM is widely used for feature localization in images to reflect the logical decision-making information behind the neural network due to its high applicability. However, extensive experimentation on a customized dataset revealed that the deep convolutional neural network (CNN) model based on Gradient-weighted Class Activation Mapping (Grad-CAM) technology cannot effectively resist the interference of large-scale noise. In this article, an optimization of the deep CNN model was proposed by incorporating the Dropkey and Dropout (as a comparison) algorithm. Compared with Grad-CAM, the improved Grad-CAM based on Dropkey applies an attention mechanism to the feature map before calculating the gradient, which can introduce randomness and eliminate some areas by applying a mask to the attention score. Experimental results show that the optimized Grad-CAM deep CNN model based on the Dropkey algorithm can effectively resist large-scale noise interference and achieve accurate localization of image features. For instance, under the interference of a noise variance of 0.6, the Dropkey-enhanced ResNet50 model achieves a confidence level of 0.878 in predicting results, while the other two models exhibit confidence levels of 0.766 and 0.481, respectively. Moreover, it exhibits excellent performance in visualizing tasks related to image features such as distortion, low contrast, and small object characteristics. Furthermore, it has promising prospects in practical computer vision applications. For instance, in the field of autonomous driving, it can assist in verifying whether deep learning models accurately understand and process crucial objects, road signs, pedestrians, or other elements in the environment.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- College of Mechanical and Electrical Engineering, Nanjing Forestry University, Nanjing 210037, China.
- SEU-FEI Nano-Pico Center, Key Lab of MEMS of Ministry of Education, Southeast University, Nanjing 210096, China.
- College of Electronic and Optical Engineering & College of Flexible Electronics (Future Technology), Nanjing University of Posts and Telecommunications, Nanjing 210023, China.
ORCID-Profile: Yingqing Guo
Förderung
- National Natural Science Foundation of China
- Postdoctoral Science Foundation of China
- Fundamental Research Funds for the Central Universities
- Nanjing Forestry University College Student Innovation Training Program
- NUPTSF
- Open Research Fund of key Laboratory of MEMS of Ministry of Education, Southeast University
Zitieren
APA
Liu, Y., Tang, L., Liao, C., Zhang, C., Guo, Y., Xia, Y., Zhang, Y., & Yao, S. (2023). Optimized Dropkey-Based Grad-CAM: Toward Accurate Image Feature Localization. Sensors, 23(20), 8351. https://doi.org/10.3390/s23208351
Vancouver
Liu Y, Tang L, Liao C, Zhang C, Guo Y, Xia Y, et al. Optimized Dropkey-Based Grad-CAM: Toward Accurate Image Feature Localization. Sensors. 2023;23(20):8351. doi:10.3390/s23208351
BibTeX
@article{Liu2023,
author = {Liu, Yiwei and Tang, Luping and Liao, Chen and Zhang, Chun and Guo, Yingqing and Xia, Yixuan and Zhang, Yangyang and Yao, Sisi},
title = {Optimized Dropkey-Based Grad-CAM: Toward Accurate Image Feature Localization},
journal = {Sensors},
year = {2023},
volume = {23},
number = {20},
pages = {8351},
doi = {10.3390/s23208351}
}
EUVIMED