A study on computer vision for facial emotion recognition.

Artificial intelligence has been successfully applied in various fields, one of which is computer vision. In this study, a deep neural network (DNN) was adopted for Facial emotion recognition (FER). One of the objectives in this study is to identify the critical facial features on which the DNN model focuses for FER. In particular, we utilized a convolutional neural network (CNN), the combination of squeeze-and-excitation network and the residual neural network, for the task of FER. We utilized AffectNet and the Real-World Affective Faces Database (RAF-DB) as the facial expression databases that provide learning samples for the CNN. The feature maps were extracted from the residual blocks for further analysis. Our analysis shows that the features around the nose and mouth are critical facial landmarks for the neural networks. Cross-database validations were conducted between the databases. The network model trained on AffectNet achieved 77.37% accuracy when validated on the RAF-DB, while the network model pretrained on AffectNet and then transfer learned on the RAF-DB results in validation accuracy of 83.37%. The outcomes of this study would improve the understanding of neural networks and assist with improving computer vision accuracy.

Real-world use of ranibizumab for neovascular age-related macular degeneration in Taiwan.

This study investigated the "real-world" use of ranibizumab for neovascular age-related macular degeneration (nAMD) in Taiwan and assessed the visual outcome. We reviewed the medical records at National Cheng Kung University Hospital, Taiwan, during 2012-2014 for 264 consecutive eyes of 229 patients with nAMD, who applied for ranibizumab covered by national health insurance. A total of 194 eyes (73.5%) in 179 patients (65.5% men; mean ± standard deviation age 69.4 ± 10.7 years) were pre-approved for treatment. Applications for treatment increased year by year, but approval rates decreased during this time. The major causes of rejection for funding were diseases mimicking nAMD, including macular pucker/epiretinal membrane, macular scarring, dry-type AMD, and possible polypoidal choroidal vasculopathy. After completion of three injections in 147 eyes, visual acuity significantly improved, gaining ≥1 line in 51.8% of eyes and stabilising in 38.3% of 141 eyes in which visual acuity was measured. The 114 eyes approved with only one application had a better visual outcome than the 27 eyes approved after the second or third applications. In conclusion, ranibizumab is effective for nAMD; however, approval after the second or third application for national health insurance cover is a less favourable predictor of visual outcome.

Fiber-Optic Thermal Sensor for TiN Film Crack Monitoring.

The study focuses on the thermal and temperature sensitivity behavior of an optical fiber sensor device. In this article, a titanium nitride (TiN)-coated fiber Bragg grating (FBG) sensor fabricated using an ion beam sputtering system was investigated. The reflection spectra of the FBG sensor were tested using R-soft optical software to simulate the refractive index sensitivity. In these experiments, the temperature sensitivity of the TiN FBG was measured at temperatures ranging from 100 to 500 °C using an optical spectrum analyzer (OSA). The results showed that the temperature sensitivity of the proposed TiN FBG sensor reached 12.8 pm/°C for the temperature range of 100 to 300 °C and 20.8 pm/°C for the temperature range of 300 to 500 °C. Additionally, we found that the produced oxidation at temperatures of 400-500 °C caused a crack, with the crack becoming more and more obvious at higher and higher temperatures.

Enhancement of mosquito trapping efficiency by using pulse width modulated light emitting diodes.

In this study, a light-driving bug zapper is presented for well controlling the diseases brought by insects, such as mosquitoes. In order to have the device efficient to trap the insect pests in off-grid areas, pulse width modulated light emitting diodes (PWM-LED) combined with a solar power module are proposed and implemented. With specific PWM electric signals to drive the LED, it is found that no matter what the ability of catching insects or the consumed power efficiency can be enhanced thus. It is demonstrated that 40% of the UV LED consumed power and 25.9% of the total load power consumption can be saved, and the trapped mosquitoes are about 250% increased when the PWM method is applied in the bug zapper experiments.

Design of asymmetric freeform lens for low glared LED street light with total internal reflection.

The study is focused on the asymmetric secondary freeform lens (ASFL) design for creating a low glared light-emitting diode (LED) street light. The lens is mounted on a chip on board (COB) LED as the new LED street light module to perform a non-axial symmetric light intensity distribution. The experimental results show that the street light can work without inclining lamps and reach Chinese National Standards (CNS) and Illuminating Engineering Society of North America (IESNA) standards at the same time.

Analysis and Optimization of Thermodiffusion of an FBG Sensor in the Gas Nitriding Process.

In this paper, we report the numerical calculations for a thermo-optical model and the temperature sensitivity of a fiber Bragg grating (FBG) sensor. The thermally-induced behaviors of a FBG sensor in the gas nitriding process were analyzed for temperatures ranging from 100⁻650 °C. The FBG consisted of properly chosen photosensitive fiber materials with an optimized thermo-optic coefficient. The experimental and optimized thermo-optic coefficient results were consistent in terms of temperature sensitivity. In these experiments, the temperature sensitivity of the FBG was found to be 11.9 pm/°C.

Start-up of simultaneous partial nitrification, anammox and denitrification (SNAD) process in sequencing batch biofilm reactor using novel biomass carriers.

Simultaneous partial nitrification, anammox and denitrification (SNAD) process was started-up in a 2.5L sequencing batch biofilm reactor (SBBR) using novel biomass carriers. The SNAD process took only 51d for start-up at nitrogen loading rate (NLR) and organic loading rate (OLR) of 120 and 60g/m(3)-d, respectively. Long-term stable operation of SNAD process was observed at NLR and OLR of 360 and 180g/m(3)-d with average total nitrogen and COD removal efficiencies of >88% and >90%, respectively. The values of conversion ratio [Formula: see text] remained below 0.11 after the start-up period, which further confirmed the long-term stability of SNAD process. Results of polymerase chain reaction (PCR), qualitative PCR, and scanning electron microscopic (SEM) analysis of sludge samples confirmed the co-existence and enrichment of AOB, anammox bacteria and denitrifying bacteria in the reactor and biofilm formation on to the carriers.

Effect of zinc on anammox activity and performance of simultaneous partial nitrification, anammox and denitrification (SNAD) process.

In the present study, short-term effects of zinc on anammox activities and long-term effect of zinc on the performance of simultaneous partial nitrification, anammox and denitrification (SNAD) process were evaluated. The anammox activity decreased with increasing zinc concentration and exposure time in short-term tests. The IC50 value of zinc was found to be 6.9mg/L. However, the presence of zinc (<10mg/L) in wastewater stimulated the microbial activities and nitrogen removal performance of SNAD process in sequencing batch biofilm reactor (SBBR). At first, inhibition of SNAD process was observed when influent zinc concentration increased to 20mg/L. The system recovered immediately, suggesting the acclimatization of microbial communities of SNAD process. The results showed that SBBR was well acclimatized under high zinc concentration (50-100mg/L) achieving 98% NH4(+)-N, 96% TN and 87% COD removal efficiencies.