Dual Attention Network for Pitch Estimation of Monophonic Music

The task of pitch estimation is an essential step in many audio signal processing applications. In this paper, we propose a data-driven pitch estimation network, the Dual Attention Network (DA-Net), which processes directly on the time-domain samples of monophonic music. DA-Net includes six Dual Attention Modules (DA-Modules), and each of them includes two kinds of attention: element-wise and channel-wise attention. DA-Net is to perform element attention and channel attention operations on convolution features, which reflects the idea of "symmetry". DA-Modules can model the semantic interdependencies between element-wise and channel-wise features. In the DA-Module, the element-wise attention mechanism is realized by a Convolutional Gated Linear Unit (ConvGLU), and the channel-wise attention mechanism is realized by a Squeeze-and-Excitation (SE) block. We explored three kinds of combination modes (serial mode, parallel mode, and tightly coupled mode) of the element-wise attention and channel-wise attention. Element-wise attention selectively emphasizes useful features by re-weighting the features at all positions. Channel-wise attention can learn to use global information to selectively emphasize the informative feature maps and suppress the less useful ones. Therefore, DA-Net adaptively integrates the local features with their global dependencies. The outputs of DA-Net are fed into a fully connected layer to generate a 360-dimensional vector corresponding to 360 pitches. We trained the proposed network on the iKala and MDB-stem-synth datasets, respectively. According to the experimental results, our proposed dual attention network with tightly coupled mode achieved the best performance.

Characterization of second-order bands in Raman scattering spectra of lead phthalocyanine thin films

The structure, optical absorption (500…950 nm) and resonance Raman spectra (within the range 100…3000 cm–1) of lead phthalocyanine (PbPc) thin solid films with the thickness 190 nm were studied. The films were deposited using thermal evaporation in vacuum 6.5 mPa onto silica substrates held at room temperature. It was found that in the process of depositing the PbPc thin solid films monoclinic and triclinic PbPc crystallites were grown, and the amount of crystallites in the triclinic phase in the as-deposited PbPc films was approximately two times less than those in the monoclinic one. The resonance Raman spectroscopy, with application of the He-Ne laser line 632.8 nm as an excitation source, was used for studying the 190-nm thick PbPc films. Due to resonance enhancement, the second-order Raman spectrum of PbPc films within the region 1700…2950 cm–1 was successfully registered and analyzed for the first time. It has been shown that the second-order PbPc Raman spectrum is mainly formed by the overtones and combination modes of B1 symmetry fundamental vibrations. The second-order Raman region of 2550…2900 cm–1 appeared to be highly specific for PbPc and could be used for its identification along with the finger-print region of fundamental vibrational modes.

Application of hybrid electrocoagulation–filtration methods in the pretreatment of marine aquaculture wastewater

The aim of this study was to provide technical means and data support for enhancing the filtration pretreatment capacity of a recirculating aquaculture system. A continuous flow electrocoagulation (EC)–filtration system was designed and its application in the pretreatment of marine aquaculture wastewater was studied. The influences of anode combination modes, hydraulic retention times (HRTs) of the EC reactor and filter pore sizes on the water treatment capacity were investigated. Results showed that EC could significantly enhance the treatment efficiency of the filtration equipment used in subsequent steps. Al-Fe electrodes used as anode led to better processing capacity of this system, and the optimum anode was 3Al + Fe. With the increase of HRT and decrease of filter pore size, the enhanced effect of the EC process on the filter was more obvious. When the current density was 19.22 A/m2, the anode was 3Al + Fe, the HRT was 4.5 min and the filter pore size was 45 μm, the removal efficiency of the system for Vibrio, chemical oxygen demand, total ammonia nitrogen, nitrite nitrogen (NO2−-N), nitrate nitrogen (NO3−-N) and total nitrogen was 69.55 ± 0.93%, 48.99 ± 1.42%, 57.06 ± 1.28%, 34.09 ± 2.27%, 18.47 ± 1.88% and 55.26 ± 1.42%, respectively, and the energy consumption was (26.25 ± 4.95) × 10−3kWh/m3.

Establishment of the Qy Absorption Spectrum of Chlorophyll a Extending to Near-Infrared

A weak absorption tail related to the Qy singlet electronic transition of solvated chlorophyll a is discovered using sensitive anti-Stokes fluorescence excitation spectroscopy. The quasi-exponentially decreasing tail was, at ambient temperature, readily observable as far as −2400 cm−1 from the absorption peak and at relative intensity of 10−7. The tail also weakened rapidly upon cooling the sample, implying its basic thermally activated nature. The shape of the spectrum as well as its temperature dependence were qualitatively well reproduced by quantum chemical calculations involving the pigment intramolecular vibrational modes, their overtones, and pairwise combination modes, but no intermolecular/solvent modes. A similar tail was observed earlier in the case of bacteriochlorophyll a, suggesting generality of this phenomenon. Long vibronic red tails are, thus, expected to exist in all pigments of light-harvesting relevance at physiological temperatures.

Interactions of Fish Proteins and Volatile Flavor Compounds: Mechanisms, Affecting Factors and Analytical Methods

: The removal of volatile compounds with unfavorable fishy odor from fish has attracted considerable research attention especially in surimi production industry. The binding of these volatile compounds by fish proteins are reported to affect the deodorization of fishy odor. To remove the protein-bound fishy substances effectively, the mechanisms underlying the interaction between proteins and flavor substances and methods that can extract and analyze the aromatic substances have been extensively studied. This paper briefly reviews the typical fishy compounds and the combination modes and mechanisms of fish proteins with these compounds. Moreover, factors affecting the interactions including protein conformation and solution conditions are discussed in detail.

Preferred Service Investment in the Dual-Channel Supply Chain: Leader-Follower Relationships and Product Characteristics

This paper examines the influence that members’ preferred service investments have on their decision-making and profits in a dual-channel supply chain. We discuss changes in the supplier’s and retailer’s decision-making and optimal profits with reference to their different preferred service investment combination modes and their different leader-follower relationships. We also provide classified comparisons of the supplier’s and retailer’s optimal profits by referring to four types of product characteristics combination modes that are assessed across two channels. By providing classifications and comparisons, we put forward operational insights for the supplier and retailer who have direct implications for their optimal decisions and profits, including selling products with higher homogeneity between two channels, choosing the retailer as the leader of the supply chain, and confirming appropriate preference combination mode under a certain product characteristics. We analyze the influences of the preferred service investment levels as well as preference intensities of the supplier and retailer on their strategies and profits in their respective advantaged states and obtain some enlightenment about the cooperation strategies and relationship-building methods of the supplier and retailer in the actual process of their participation in the dual-channel supply chain operation.

On-Orbit Calibration of Installation Parameter of Multiple Star Sensors System for Optical Remote Sensing Satellite with Ground Control Points

Owing to the vibrations and thermal shocks that arise during the launch and orbit penetration process, the on-orbit installation parameters of multiple star sensors are different from the on-ground measured parameters, causing inconsistencies in the attitude determinations from different combination modes and seriously affecting the geometric accuracy of high-resolution optical remote sensing images. This study presents an on-orbit calibration approach for the installation parameters of a multiple star sensors system using ground control points (GCPs). Based on the on-ground installation parameters of the optical axes of conventional star sensors, a fiducial coordinate system is proposed as the calibration coordinate system. The installation parameters of the conventional star sensors are calibrated using the statistical characteristics of angles between axes of the star sensor and three fiducial vectors in the J2000 celestial coordinate system. Based on the GCPs, the relative fiducial parameters are calculated, and the installation parameter of unconventional star sensor is then calibrated with the relative fiducial parameters and statistical characteristics of angles. It can be used for high-resolution optical remote sensing satellite measuring with only two star sensors to unify the fiducial coordinate system. The proposed method is tested using simulated data and on-orbit measurement data. The results demonstrate that the proposed method can calibrate the optical axis of the star sensor without the restriction of the accuracy of horizontal axis. Moreover, the star sensor with a large installation angle error can be calibrated well using the proposed approach. The results of attitude determinations from different star sensor combination modes are consistent, and the geometric accuracy of the remote sensing images is significantly improved.