Title: | Signal Extraction for Classification of Noisy Images Compressed using Autoencoders |
Authors: | Sebai, Dorsaf Missaoui, Nour Zouaghi, Asma |
Citation: | WSCG 2021: full papers proceedings: 29. International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision, p. 245-252. |
Issue Date: | 2021 |
Publisher: | Václav Skala - UNION Agency |
Document type: | conferenceObject konferenční příspěvek |
URI: | http://hdl.handle.net/11025/45030 |
ISBN: | 978-80-86943-34-3 |
ISSN: | 2464-4617 2464–4625(CD/DVD) |
Keywords: | autoencoder;klasifikace;datový soubor hluku;hluk |
Keywords in different language: | autoencoder;noise;classification;noise dataset |
Abstract in different language: | The world is experiencing an increasing boom in computer vision. This is more and more used in many domainssuch as robotics, medicine, industry, security systems, etc. In this context, Deep Neural Networks (DNNs) havegreat capabilities and are widely used. Convolutional Neural Networks (CNNs) present a particular class of DNNsthat is most commonly leveraged to analyzing visual imagery. However, CNN performances completely dependon two main issues. The first issue is related to the quality of the images generated by capture cameras. All imagescaptured by remote sensors and modern imaging systems are practically noisy, which can prevent the image frombeing correctly classified and identified by a CNN. The second issue is the throughput available for the transmissionof the large amount of data between capture sensors and units processing CNNs. A seamless transmission can beensured by compression techniques that help reducing the size of data, while affording the required quality forcomputer vision algorithms. Since lossy compression of noise-free and noisy images differ from each other, thiswork firstly raises the question of CNNs resilience to noisy images compression using the particular autoencoders.We secondly propose a method that aims to improve this resilience so that CNNs can achieve better classificationperformances. The compressed noisy images are passed, as a test set, along a model that is learnt from a noisedataset. The subtraction of the so captured noise from the noisy images is then performed to extract the usefulsignal to classify. This will be first work, where we learn the autoencoder from the noise sample, and not the noisysample, while denoising. Obtained results prove the efficiency of the proposed method. |
Rights: | © Václav Skala - UNION Agency |
Appears in Collections: | WSCG 2021: Full Papers Proceedings |
Please use this identifier to cite or link to this item:
http://hdl.handle.net/11025/45030
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.