Logo image
On classification of distorted images with deep convolutional neural networks
Conference proceeding

On classification of distorted images with deep convolutional neural networks

Yiren Zhou, Sibo Song and Ngai-Man Cheung
The Institute of Electrical and Electronics Engineers, Inc. (IEEE) Conference Proceedings, p.1213
01/01/2017

Abstract

Acoustic noise Acoustics Artificial neural networks Classifiers Distortion Image acquisition Image classification Neural networks Noise prediction Signal processing Speech Training Tuning
Conference Title: ICASSP 2017 - 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) Conference Start Date: 2017, March 5 Conference End Date: 2017, March 9 Conference Location: New Orleans, LA, USA Image blur and image noise are common distortions during image acquisition. In this paper, we systematically study the effect of image distortions on the deep neural network (DNN) image classifiers. First, we examine the DNN classifier performance under four types of distortions. Second, we propose two approaches to alleviate the effect of image distortion: re-training and fine-tuning with noisy images. Our results suggest that, under certain conditions, fine-tuning with noisy images can alleviate much effect due to distorted inputs, and is more practical than re-training.

Metrics

1 Record Views

Details

Logo image