Performance Analysis of Image Denoising using Deep Convolutional Neural Network
| Autor: | S. Priyadarsini, K Ramalakshmi, G Sankaramalliga, S. Thayammal |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Image quality
business.industry Computer science Noise reduction media_common.quotation_subject ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Pattern recognition Convolutional neural network Image (mathematics) Contrast (vision) Quality (business) Artificial intelligence Noise (video) Enhanced Data Rates for GSM Evolution business media_common |
| Zdroj: | IOP Conference Series: Materials Science and Engineering. 1070:012085 |
| ISSN: | 1757-899X 1757-8981 |
| Popis: | A performance analysis of conventional Convolutional Neural Network (CNN) based denoising method is proposed. In this image denoising method, the contrast of images is adaptively enhanced. Generally, it is not possible to capture the imageswith good quality for all situations. Because they are captured in various light conditions.So, the captured images are suffered by noise, which results in poor perceived image quality. Thus, it is necessary to improve the quality of images with edge detail preservation as much as possible. The convolutional neural network model for low light image enhancement is already developed and is named as DnCNNs. Here, the performance analysis of image denoising using the DnCNNmodel is presented. The DnCNN implicitly removes the noise in the image. The simulation results afford better reference for application developers. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|