Papers by Keyword: Denoise

Abstract: Total magnetic intensity (TMI) data is an important data for understanding the seafloor spreading of the South China Sea. In this paper, we studied to denoise the TMI data in the ocean subbasin of the SCS by using various filtering techniques. Based on the noise features occured in the data, we utilize the directional filter to reject noise of NS direction, and then apply the anisotropic diffusion filter to further elliminate the bead-like artifacts. The denoised result presents a well conherence of magnetic lineations in the ocean basin. The magnetic lineaments trend to nearly EW in the eastern subbasin while NE in southwestern subbasin. The intensity of the magnetic lineaments in the eastern subbasin is stronger than that in the southwestern subbasin.

Abstract: Eddy current testing is one of the five major routine nondestructive testing methods and it is convenient, fast and suitable for online detecting of the surface fatigue crack of bolt holes. However, the signals of eddy current testing are so weak that it is difficult to identify the signals. So more effective signal processing method must be adopted to deal with the weak signals. This paper used the wavelet analysis to process signals of the eddy current testing for the surface flaw of bolt holes. It can inhibit the noise and reinforce signal and make qualitative testing possible for quality evaluation of the surface fatigue crack of bolt holes.

Abstract: The basic principle of the NAS-RIF algorithm is described and this algorithm has a simple structure, but it tends to amplify the noise and produce excessive smoothing in the iterative process. This paper takes some measures and focuses on overcoming these shortcomings. Firstly we use an adaptive denoising method based on total variation to reduce noise in turbulence-degraded images. Then, a regularized term is added into the cost function to preserve the edges effectively and a new form of the cost function is also improved to guarantee the convergence of the algorithm. Comparing with the traditional NAS-RIF algorithm, the proposed algorithm has a positive improvement in restraining the noise enlargement, preserving the detail features and the image restoration effect is obviously better.

Abstract: Speckle noise is a common phenomenon in Optical Coherence Tomography (OCT) images. This paper describes a dynamic filtering approach for anterior chamber OCT images to reduce the speckle noise in wavelet domain. The approach proposed segments the OCT image into some parts and identifies if the parts have region of interest (ROI), which includes the anterior chamber tissues. Then, it suppresses noise with three different suppression strategies in part with ROI. For the part without ROI, it sets the discrete wavelet transform coefficients of this part to zero. Here, the sampling-based sub-band adaptive algorithm is used to distinguish the ROI; and the correlations of neighboring wavelet coefficients and the coefficients of the corresponding locations in adjacent decomposition levels are used to suppress the noise. The performance improvement over the previously published method is quantified in terms of noise suppression, image structural preservation and visual quality. The numerical values of the image quality metrics along with the qualitative analysis results indicated that the approach proposed has better performance.

Abstract: In the EMC interference environment of mine, we found the EMC is white noise and pulse interference. The UWB signal pulses is short and signal power spectral density distributed in wide band is very low. This article mainly proposes a de-noising method of Wavelet transform modulus maxima in signal band, focused on the traditional low-pass filter that can only remove the outside-band noise. Simulation results show that the scheme can effectively retain the signal singularity based on information to achieve obvious noise reduction

Abstract: Phase unwrapping is one of the key technologies in electronic speckle pattern interferometry. A new phase unwrapping algorithm, based on windowed Fourier transform is proposed. The high noisy phase map is denoised by the window Fourier transform approach and then is uwrapped by the discrete cosine transform. The method is tested in this paper using a circumferentially fixed circular plate with a point load at the centre and compared with not denoised wrapped phase map and unwrapped phase map. The result shows the new proposed phase unwrapping method in denoising and improving image quality has obvious superiority.

Abstract: In the PCB micro drilling, because the force signal is tiny, and when the micro-drill drill to a certain degree of multilayer PCB, alternate force signals will not appear obvious, through to the drilling force signal analysis, we can know the drill bit position and the materials to the influence of the drill failure, so the drilling force signals denoise seems extremely important. In the processing of the non-stationary signal, traditional signal processing method has a certain extent of insufficient, using the wavelet packet decomposition signal, the white noise variance and amplitude decrease with the increase of wavelet scales, but the signal variance and amplitude has nothing to do with the wavelet transform. According to the view of the signal energy, first of all, we make the multiscale decomposition of the signal, then, by using some of the wavelet packet that has efficient energy to reconstruct the original signal. Comparing with the traditional threshold denoising ,using this method in the test signal to deal with the noise can effectively eliminate the white noise interference, and has good denoising effects besides the simple calculation.

Abstract: The load time history signal of engineering vehicle is usually polluted by various nonstationary and stationary noises in the field test. An approach based on wavelet transform (WT) and fractal dimension (FD) is proposed in order to improve the adaptability and efficiency of denoising. This method initially decomposes the original signal into detail and approximation space in the WT domain by WT-based multiresolution decomposition. The short-time fractal dimension of detail coefficient is calculated at each scale. After the application of the binary processing to the short-time fractal dimensions, the locations where the thresholding of the detail coefficients has to be executed are ensured. The desired load signal is provided by applying WT-based multiresolution reconstruction to the processed detail coefficients and the unprocessed approximation coefficients. The proposed method is applied to an actual load time history signal of engineering vehicle. And the performance of this method is compared with that of the WT-based hard thresholding denoising method. The results show that this method is an alternative way to process the load time history signal of engineering vehicle.

Abstract: With the increase request of strip quality, the roller eccentricity has become the important factor of influencing product quality. On the practical model of Cold-rolling Mill of 300, this paper establishes eccentricity compensation control practice of Cold-rolling Mill of 300 which applies wavelet transform method. Based on mufti-resolution decomposition of roll force signal by use of wavelets, eccentricity signals are separated from disturbances and noise,and self-optimization is employed to real-time control the roller eccentricity. The results show that the method is effective.

Abstract: Many methods have been developed to remove image noises with wavelet. Here, the combination of those methods is considered to construct a new method that covers more aspects of the problem. Data fusion is chosen for such combination. With two classic existent methods as an example, a data-fusion based wavelet image denoising method is proposed. Experiment results show that the new method can provide better denoising performance thus suggests the potential of such a strategy.