A New Feature Descriptor for Image Denoising
- PDF / 567,199 Bytes
- 6 Pages / 595.276 x 790.866 pts Page_size
- 49 Downloads / 226 Views
(0123456789().,-volV)(0123456789(). ,- volV)
RESEARCH PAPER
A New Feature Descriptor for Image Denoising Neda Mohamadi1 • Ali R. Soheili2 • Faezeh Toutounian2 Received: 28 October 2019 / Accepted: 4 September 2020 Shiraz University 2020
Abstract One of the fundamental problems in the field of image processing is denoising. The underlying goal of image denoising is to effectively suppress noise while keeping intact the significant features of the image, such as texture and edge information. The gradient of image is a famous feature descriptor in denoising models to distinguish edges and ramps. If the received signal of an image is very noisy, the gradient cannot effectively distinguish between the image edges and the image ramps. In this paper, based on the difference curvature and the gradient of the image, we introduce a new feature descriptor. For demonstrating the effectiveness of the new feature descriptor, we use it in constructing a new diffusionbased denoising model. Experimental results show the effectiveness of the method. Keywords Difference curvature Feature descriptor Image denoising
1 Introduction Digital images can be acquired by different sensors, and in the process of image acquisition and transmission, all recording devices have traits which make them susceptible to noise. Noise degrades the quality of image and causes difficulty in image observation and analysis. In order to effectively reduce noise, various methods have been proposed. A large number of image denoising methods are conveniently formulated using some partial differential equations (PDEs) (Barbu 2019; Black et al. 1998; Catte et al. 1992; Jin et al. 2000; Mokhtari and Mostajeran 2020; Prasath and Singh 2010). The gradient of image is a famous feature descriptor for using in diffusion-based denoising models. The gradient of a very noisy image has a great
& Neda Mohamadi [email protected]; [email protected] 1
Department of Mathematics and Statistics, Mashhad Branch, Islamic Azad University, Mashhad, Iran
2
Department of Applied Mathematics, Faculty of Mathematical Sciences, Ferdowsi University of Mashhad, Mashhad, Iran
number of irrelevant maxima, and it cannot effectively distinguish between the image edges and the image ramps. Another feature descriptor which is constructed by the second-order derivatives along the direction of gradient of an image and perpendicular to the gradient is difference curvature (DC) (Chen et al. 2010). In this paper in order to apply both gradient and difference curvature to distinguish the edges from the flat areas, we will introduce a new feature descriptor based on gradient magnitude and difference curvature and use it in constructing a diffusionbased denoising model. The new diffusion-based model has been designed to perform more diffusion in the flat areas of the image and less diffusion in the edges of the image. So we can get rid of the noise and preserve the edges of the image simultaneously. The rest of this paper is organized as follows: In Sect. 2, we briefly describe
Data Loading...
