Noise Level Map Research Articles

FFDNet: Toward a Fast and Flexible Solution for CNN based Image Denoising.

Due to the fast inference and good performance, discriminative learning methods have been widely studied in image denoising. However, these methods mostly learn a specific model for each noise level, and require multiple models for denoising images with different noise levels. They also lack flexibility to deal with spatially variant noise, limiting their applications in practical denoising. To address these issues, we present a fast and flexible denoising convolutional neural network, namely FFDNet, with a tunable noise level map as the input. The proposed FFDNet works on downsampled subimages, achieving a good trade-off between inference speed and denoising performance. In contrast to the existing discriminative denoisers, FFDNet enjoys several desirable properties, including (i) the ability to handle a wide range of noise levels (i.e., [0, 75]) effectively with a single network, (ii) the ability to remove spatially variant noise by specifying a non-uniform noise level map, and (iii) faster speed than benchmark BM3D even on CPU without sacrificing denoising performance. Extensive experiments on synthetic and real noisy images are conducted to evaluate FFDNet in comparison with state-of-the-art denoisers. The results show that FFDNet is effective and efficient, making it highly attractive for practical denoising applications.

Work Rotation to Reduce the Effect of Noise Exposure for Operators in Sugar Factory

This study investigates noise level at the grinding station in a sugar factory by measuring sound pressure level. This study aims to obtain noise level map, so that operators’ work hour can be regulated accordingly. Noise level is measured for four days using sound level meter. Observation is conducted at 9 points at 09.00, 11.00, 13.00, and 15.00 o’clock. The number of point of measurement is determined using contour mapping technique using grid size of 3m x 3m. Noise distribution pattern is obtained from surfer software v11.0. The results show that Daily Noise Dose (DND) exceeds the standard set by Occupational Safety and Health (OSHA). Every point of measurement has different noise level, which means that the maximum exposure duration allowed is also different. Maximum exposure duration allowed for the 9 points is ranging from 72.48 minutes to 143.54 minutes. This infers that the problem can be solved by work hour regulation of operators at each station; operators work for a different period of time at different stations.

Description of Anomalous Noise Events for Reliable Dynamic Traffic Noise Mapping in Real-Life Urban and Suburban Soundscapes

Traffic noise is one of the main pollutants in urban and suburban areas. European authorities have driven several initiatives to study, prevent and reduce the effects of exposure of population to traffic. Recent technological advances have allowed the dynamic computation of noise levels by means of Wireless Acoustic Sensor Networks (WASN) such as that developed within the European LIFE DYNAMAP project. Those WASN should be capable of detecting and discarding non-desired sound sources from road traffic noise, denoted as anomalous noise events (ANE), in order to generate reliable noise level maps. Due to the local, occasional and diverse nature of ANE, some works have opted to artificially build ANE databases at the cost of misrepresentation. This work presents the production and analysis of a real-life environmental audio database in two urban and suburban areas specifically conceived for anomalous noise events’ collection. A total of 9 h 8 min of labelled audio data is obtained differentiating among road traffic noise, background city noise and ANE. After delimiting their boundaries manually, the acoustic salience of the ANE samples is automatically computed as a contextual signal-to-noise ratio (SNR). The analysis of the real-life environmental database shows high diversity of ANEs in terms of occurrences, durations and SNRs, as well as confirming both the expected differences between the urban and suburban soundscapes in terms of occurrences and SNRs, and the rare nature of ANE.

A method to assess spatially variant noise in dynamic MR image series

Accurate measurement of spatially variant noise in MR images acquired using parallel imaging techniques is challenging. Image-based noise measurement methods such as the subtraction method proposed by the National Electrical Manufacturers Association or the multiple acquisition method often cannot be applied in vivo due to motion and/or dynamic contrast changes. Based on the Karhunen-Loeve transform and random matrix theory, we propose a novel method to accurately assess the noise variance in image series bearing temporal redundancy. The method fits the probability density function of eigenvalues from the temporal covariance matrix of the image series to the Marcenko-Pastur distribution. The accuracy of our method was validated using numerical simulation and an MR noise measurement experiment. The ability of this method to derive the g-factor map of a static phantom was validated against the multiple acquisition method. The method was applied to in vivo cardiac and brain image series and the results agreed with subtraction and multiple acquisition methods, respectively. This new image-based noise measurement method provides a practical means of retrospectively evaluating the noise level and/or g-factor map from multiframe image series.

Noise Mapping and Noise Scoring - Software Techniques and Result Presentation

Noise scoring is based on noise level maps and on exposure distribution. Different techniques can be used to develop the areal distribution of these values and to present them as coloured maps. Especially more complex ratings that are not only based on an exposure level and on population densities need relatively complex procedures to develop them from noise maps. It must be taken into account that there are two important steps in this procedure: One is the calculation of the noise score itself and the other is the presentation as coloured map to detect and focus on Hot Spots. Different techniques have been developed, were integrated in the software and have been applied in the frame of the QCity project. The consequences of different parameter settings are presented and discussed. The second step is the assignment of colours to a given scale of result values and even this simple process must be handled carefully to transport the needed information.

The development of prediction models to control noise levels around UK firing ranges

Since 1981, many UK firing ranges have used computer-based prediction models to control noise levels in the surrounding countryside. These models take into account local surface conditions and meteorology to provide peak noise level contour maps over distances of up to 30 km. A feature of these models is that peak noise level maps can be produced within a few minutes of the input of meteorological data obtained directly from radiosonde ascents or from mesoscale models. Several new prediction models have been developed at Salford over the years, taking advantage of the ever increasing power of personal computers to improve accuracy and user input and output facilities. This paper will discuss the model developments, including the meteorological preprocessor, illustrate some of the techniques adopted to improve accuracy and speed, and outline future improvements.