Natural Image-Orientated Hybrid Filter Using Pulse Coupled Neural Network
Noise introduced during capture and transmission is inevitable for natural images generated by Complementary Metal-Oxide-Semiconductor (CMOS) sensors, including quantization error during digitalization, transmission disturbance and other sources of noise.
- PDF / 174,218 Bytes
- 7 Pages / 439.37 x 666.142 pts Page_size
- 37 Downloads / 215 Views
Abstract Noise introduced during capture and transmission is inevitable for natural images generated by Complementary Metal-Oxide-Semiconductor (CMOS) sensors, including quantization error during digitalization, transmission disturbance and other sources of noise. To process natural images from a CMOS sensor, a hybrid filter combining Pulse Coupled Neural Network (PCNN), median filter and Wiener filter is proposed in this paper. First, salt-and-pepper noise is located via PCNN, and processed by a median filter. Then, Gaussian noise is removed by a self-adaptive Wiener filter. Simulation results indicated that compared to other methods (hybrid filter containing median and Wiener filter, hybrid filter containing median and wavelet filter), the hybrid filter with PCNN demonstrates better performance in the preservation of image detail and edge in the premise of similar Signal-Noise Ratios (SNRs). Keywords PCNN
Median filter Wiener filter Image denoising
1 Introduction Image denoising is a persistent focus in the field of digital image processing, since the sources of noise in natural images are complex and a single filter is not able to remove all noise. The common noise in natural images can be classified as additive noise and multiplicative noise. Traditional median filters, mean filters, and Gaussian filters can improve or remove noise derived from certain sources, but often blur the image during denoising. The object of natural image denoising is to remove noise and preserve image details as much as possible.
Y.-D. Li (&) J.-H. Pan North China University of Technology, Beijing, China e-mail: [email protected] J.-H. Pan e-mail: [email protected] © Springer Science+Business Media Singapore 2016 A. Hussain (ed.), Electronics, Communications and Networks V, Lecture Notes in Electrical Engineering 382, DOI 10.1007/978-981-10-0740-8_31
271
272
Y.-D. Li and J.-H. Pan
Currently, the most popular denoising tool is the use of mean filters and median filters [1]. The self-adaptive centre weighted median filter proposed by Jin et al. [2] can smooth self-adaptively per noise scale. Gao and Gao [3] used PCNN to identify a salt-and-pepper point, then applied a median filter accordingly to the impulsive positions. Zou [4] proposed an improved PCNN method to remove the salt-and-pepper point, which harms the image detail to some degree. Cuo and Wang [5] removed noise by using PCNN and the wavelet transform method to maintain the image details; however, this method is difficult to implement on Field Programmable Gate Array (FPGA). The above mentioned methods all have certain advantages, but not for natural images. Natural images usually contain both Gaussian and salt-and-pepper noises, thus making it difficult to achieve satisfactory results with a single filter. In order to remove noise and retain image details simultaneously, hybrid filters are proposed. This paper proposes a novel hybrid median and Wiener filter with PCNN for natural image denoising implemented on FPGA. The denoising results verify the improved denoising performance of
Data Loading...
