China
Africa
Explore chapters and articles related to this topic
Understanding Audio Effects
Published in Brecht De Man, Ryan Stables, Joshua D. Reiss, IntelligentMusic Production, 2019
Brecht De Man, Ryan Stables, Joshua D. Reiss
As the name implies, dynamic range compression is concerned with mapping the dynamic range of an audio signal to a smaller range. Traditional dynamic range compressors achieve this goal by reducing the high signal levels while leaving the quieter parts untreated, in a process called downward compression. The same effect would be attained by upward compression, where quiet parts are amplified and loud parts are left alone, but this is a matter of (historical) convention. Compression sends the signal along two different paths, one that goes through the analysis circuit (the side-chain), and another that goes through the processing circuit (the voltage controlled amplifier or VCA). While the side-chain signal is often identical to the signal being processed, it can also be a filtered version or even an entirely different signal (external-adaptive).
Radio Studios
Published in Skip Pizzi, Graham A. Jones, A Broadcast Engineering Tutorial for Non-Engineers, 2014
A technique called audio data compression, based on perceptual coding, may be used to reduce the amount of data needed to transport or store digital audio (see Chapter 5). Note that this type of compression has nothing to do with the dynamic range compression that takes place in audio processing (see later sections in this chapter). There are many different audio compression systems, often referred to simply as codecs, and most have been standardized by some international standards organization. A specific codec is usually referred to by some abbreviation of its standard's name (such as “MP3” for the MPEG-2 Audio Layer 3 coding standard).
ODMNet: Automated Glaucoma Detection and Classification Model Using Heuristically-Aided Optimized DenseNet and MobileNet Transfer Learning
Published in Cybernetics and Systems, 2023
Felix Joseph Xavier, Fanax Femy F.
As the eye includes the complicated nature of glaucoma lesions, including structure, color, and size differentiations, the processing and identification of glaucoma are complex. Thus, there is a need for a standard approach that is used to increase the image contrast. In this model, the multi-scale Retinex-based image enhancement increases the contrast. It performs nonlinear color conversion to get better and clear details of the images and also highlights the necessary information owing to various illumination conditions. Some images may have imbalanced illumination of background due to the lack of loss of details and quality of acquired images. The multi-scale Retinex-based image enhancement (Zotin 2018) is defined as a weighted sum of several Single-Scale Retinex outputs. It is more efficient in balancing the tradeoff between the color rendition and dynamic range compression. It is mathematically formulated in Eq. (1).
Frequency component vectorisation for image dehazing
Published in Journal of Experimental & Theoretical Artificial Intelligence, 2021
Nazeer Muhammad, Hira Khan, Nargis Bibi, Muhammad Usman, Naseer Ahmed, Shahid Nawaz Khan, Zahid Mahmood
Deep convolutional neural network (D-CNN) (J. Li et al., 2018) is followed by the image enhancement based method. This method increases the overall contrast of haze distorted data by increasing the dynamic range of the grey intensities (Muhammad et al., 2018). Though, CNN struggles to restore the most favourable value regarding every local area, however, offers highly complex computation. The Aod-net (B. Li et al., 2017) model preserves the stability between the dynamic range compression and the colour sequency. However, it lacks the ability to preserve edges which results as halo regions in shape boundary. Homogeneous filtering targets the combine frequency filtering and greyscale transformation to enhance the image standard (Berman & Avidan, 2016; Zhang et al., 2017; Cai et al., 2016; Ancuti, 2018a; Muhammad et al., 2018). It can preserve the contour information in irregular regions in an effective manner. In essence, the main motive of image improvement is to enhance the visual perception and present inflated convenience for computer identification without taking in to account the deterioration model.
Benefits of incorporating the adaptive dynamic range optimization amplification scheme into an assistive listening device for people with mild or moderate hearing loss
Published in Assistive Technology, 2018
Hung-Yue Chang, Ching-Hsing Luo, Tun-Shin Lo, Hsiao-Chuan Chen, Kuo-You Huang, Wen-Huei Liao, Mao-Chang Su, Shu-Yu Liu, Nan-Mai Wang
Without a remote microphone, ALDs with a microphone fitted to the body-worn receiver were less beneficial than hearing aids (Boothroyd & Iglehart, 1998). Traditional ALDs provide a linear amplification scheme to increase the SNR; however, the limitations of the linear circuits entail an uncomfortable sensation when high-level input signals are received, especially in noisy environments (Farrow, Tatum, Michel, & McCabe, 2012). In a previous study, when linear ALDs were used for watching television, the loud background sound tracks appeared to be annoying and distracting; thus, the speech intelligibility was reduced (Aberdeen & Fereiro, 2014). Considering that the hearing dynamic range decreases with increasing hearing loss (Fowler, 1965; Moore & Glasberg, 1993), appropriate compressor and noise-reduction strategies commonly used in hearing-aid technology are a necessary consideration in ALD design. Some recently distributed ALDs include wide dynamic range compression (e.g., Etymotic The BEAN [Etymotic, 2015]) or output compression (e.g., Sound World Solution CS50 [Bailey, 2014]); however, their clinical benefits compared with those of a linear system remain unknown.