Browsing by Author "Zhang, X."
Now showing 1 - 6 of 6
Results Per Page
Sort Options
Item Open Access Additive neural network for forest fire detection(Springer, 2020) Pan, H.; Badawi, D.; Zhang, X.; Çetin, Ahmet EnisIn this paper, we introduce a video-based wildfire detection scheme based on a computationally efficient additive deep neural network, which we call AddNet. This AddNet is based on a multiplication-free vector operator, which performs only addition and sign manipulation operations. In this regard, we construct a dot product-like operation from the mf-operator and use it to define dense and convolutional feed-forwarding passes in AddNet. We train AddNet on images taken from forestry surveillance cameras. Our experiments show that AddNet can achieve a time-saving by 12.4% when compared to an equivalent regular convolutional neural network (CNN). Furthermore, the smoke recognition performance of AddNet is as good as regular CNNs and substantially better than binary-weight neural networks.Item Open Access A hole accelerator for InGaN/GaN light-emitting diodes(AIP Publishing, 2014) Zhang, Z. H.; Liu, W.; Tan, S. T.; Ji, Y.; Wang, L.; Zhu, B.; Zhang, Y.; Lu, S.; Zhang, X.; Hasanov, N.; Sun, X. W.; Demir, Hilmi VolkanThe quantum efficiency of InGaN/GaN light-emitting diodes (LEDs) has been significantly limited by the insufficient hole injection, and this is caused by the inefficient p-type doping and the low hole mobility. The low hole mobility makes the holes less energetic, which hinders the hole injection into the multiple quantum wells (MQWs) especially when a p-type AlGaN electron blocking layer (EBL) is adopted. In this work, we report a hole accelerator to accelerate the holes so that the holes can obtain adequate kinetic energy, travel across the p-type EBL, and then enter the MQWs more efficiently and smoothly. In addition to the numerical study, the effectiveness of the hole accelerator is experimentally shown through achieving improved optical output power and reduced efficiency droop for the proposed InGaN/GaN LED. (C) 2014 AIP Publishing LLC.Item Open Access InGaN/GaN multiple-quantum-well light-emitting diodes with a grading InN composition suppressing the Auger recombination(AIP Publishing, 2014) Zhang Z.-H.; Liu W.; Ju, Z.; Tan S.T.; Ji Y.; Kyaw, Z.; Zhang, X.; Wang, L.; Sun, X. W.; Demir, Hilmi VolkanIn conventional InGaN/GaN light-emitting diodes (LEDs), thin InGaN quantum wells are usually adopted to mitigate the quantum confined Stark effect (QCSE), caused due to strong polarization induced electric field, through spatially confining electrons and holes in small recombination volumes. However, this inevitably increases the carrier density in quantum wells, which in turn aggravates the Auger recombination, since the Auger recombination scales with the third power of the carrier density. As a result, the efficiency droop of the Auger recombination severely limits the LED performance. Here, we proposed and showed wide InGaN quantum wells with the InN composition linearly grading along the growth orientation in LED structures suppressing the Auger recombination and the QCSE simultaneously. Theoretically, the physical mechanisms behind the Auger recombination suppression are also revealed. The proposed LED structure has experimentally demonstrated significant improvement in optical output power and efficiency droop, proving to be an effective solution to this important problem of Auger recombination.Item Open Access On the mechanisms of InGaN electron cooler in InGaN/GaN light-emitting diodes(Optical Society of America, 2014) Zhang, Z. H.; W. L.; Tan, S. T.; Ju, Z.; Ji, Y.; Kyaw, Z.; Zhang, X.; Hasanov, N.; Zhu, B.; Lu, S.; Zhang, Y.; Sun, X. W.; Demir, Hilmi VolkanElectron overflow limits the quantum efficiency of InGaN/GaN light-emitting diodes. InGaN electron cooler (EC) can be inserted before growing InGaN/GaN multiple quantum wells (MQWs) to reduce electron overflow. However, detailed mechanisms of how the InGaN EC contributes to the efficiency improvement have remained unclear so far. In this work, we theoretically propose and experimentally demonstrate an electron mean-free-path model, which reveals the InGaN EC reduces the electron mean free path in MQWs, increases the electron capture rate and also reduces the valence band barrier heights of the MQWs, in turn promoting the hole transport into MQWs. (C) 2014 Optical Society of AmericaItem Open Access Polygenic risk score increases schizophrenia liability through cognition-relevant pathways(Oxford University Press, 2019) Toulopoulou, Timothea; Zhang, X.; Cherny, S.; Dickinson, D.; Berman, K. F.; Straub, R. E.; Sham, P.; Weinberger, D. R.Cognitive deficit is thought to represent, at least in part, genetic mechanisms of risk for schizophrenia, with recent evidence from statistical modelling of twin data suggesting direct causality from the former to the latter. However, earlier evidence was based on inferences from twin not molecular genetic data and it is unclear how much genetic influence ‘passes through’ cognition on the way to diagnosis. Thus, we included direct measurements of genetic risk (e.g. schizophrenia polygenic risk scores) in causation models to assess the extent to which cognitive deficit mediates some of the effect of polygenic risk scores on the disorder. Causal models of family data tested relationships among key variables and allowed parsing of genetic variance components. Polygenic risk scores were calculated from summary statistics from the current largest genome-wide association study of schizophrenia and were represented as a latent trait. Cognition was also modelled as a latent trait. Participants were 1313 members of 1078 families: 416 patients with schizophrenia, 290 unaffected siblings, and 607 controls. Modelling supported earlier findings that cognitive deficit has a putatively causal role in schizophrenia. In total, polygenic risk score explained 8.07% [confidence interval (CI) 5.45–10.74%] of schizophrenia risk in our sample. Of this, more than a third (2.71%, CI 2.41–3.85%) of the polygenic risk score influence was mediated through cognition paths, exceeding the direct influence of polygenic risk score on schizophrenia risk (1.43%, CI 0.46–3.08%). The remainder of the polygenic risk score influence (3.93%, CI 2.37–4.48%) reflected reciprocal causation between schizophrenia liability and cognition (e.g. mutual influences in a cyclical manner). Analysis of genetic variance components of schizophrenia liability indicated that 26.87% (CI 21.45–32.57%) was associated with cognition-related pathways not captured by polygenic risk score. The remaining variance in schizophrenia was through pathways other than cognition-related and polygenic risk score. Although our results are based on inference through statistical modelling and do not provide an absolute proof of causality, we find that cognition pathways mediate a significant part of the influence of cumulative genetic risk on schizophrenia. We estimate from our model that 33.51% (CI 27.34–43.82%) of overall genetic risk is mediated through influences on cognition, but this requires further studies and analyses as the genetics of schizophrenia becomes better characterized.Item Open Access Strain-reduced micro-LEDs grown directly using partitioned growth(Frontiers Media S.A., 2021-03-10) Lu, S.; Zhang, Y.; Zhang, Z.-H.; Tsai, P. C.; Zhang, X.; Zhang, S. T.; Demir, Hilmi VolkanStrain-reduced micro-LEDs in 50 μm × 50 μm, 100 μm × 100 μm, 200 μm × 200 μm, 500 μm × 500 μm, and 1,000 μm × 1,000 μm sizes were grown on a patterned c-plane sapphire substrate using partitioned growth with the metal-organic chemical-vapor deposition (MOCVD) technique. The size effect on the optical properties and the indium concentration for the quantum wells were studied experimentally. Here, we revealed that the optical properties can be improved by decreasing the chip size (from 1,000 to 100 µm), which can correspondingly reduce the in-plane compressive stress. However, when the chip size is further reduced to 50 μm × 50 μm, the benefit of strain release is overridden by additional defects induced by the higher indium incorporation in the quantum wells and the efficiency of the device decreases. The underlying mechanisms of the changing output power are uncovered based on different methods of characterization. This work shows the rules of thumb to achieve optimal power performance for strain-reduced micro-LEDs through the proposed partitioned growth process.