A novel illumination normalisation (IN) method using a convolutional neural network (CNN) is proposed. The proposed network is composed of the local pattern extraction (LPE) and illumination elimination (IE) layers. The LPE layers model the relationships between the pixels in each local region in order to handle various types of local shadow and shading in the face image. Based on the commonly used assumption about the illumination field, the IE layers generate illumination‐insensitive ratio images by calculating the ratio between the output pairs produced from the LPE layers. The final feature map obtained by combining the ratio images can possess an improved discriminative ability for face recognition (FR). For training the proposed network, the results produced by the Weber fraction‐based IN methods as ground truths are utilised. The experimental results demonstrate that the proposed network performs better in terms of FR accuracy compared with the conventional non‐CNN‐based method and it can be combined with any CNN‐based face classifier.
Aerobic degradation of high strength piggery waste elevated the reactor temperature inhibiting nitrification. This study included anaerobic pretreatment with various influent by-pass rates to control the temperature and to minimize the external carbon requirement for denitrification. To find the optimum operating conditions, both lab-scale AnSBR (anaerobic sequencing batch reactor) and Ax/Ox (anoxic/oxic) SBR were operated at 35 degrees C. The heat energy released from Ax/Ox SBR was assumed to be used for heating the AnSBR, with which the Ax/Ox reactor temperature could successfully be controlled below 40 degrees C. The optimum rates of by-pass were 1.0 for winter, 0.4 for spring/fall and 0.2-0.4 for summer, respectively. Applying the correction factors for the measured AUR2 (nitrite nitrification rate) and AUR (nitrate nitrification) at the predicted temperatures, the required oxic HRTs were computed. The required Ax/Ox HRT ratios were respectively 0.5 for COD/TKN>8, 1.0 for COD/TKN ratio of 5.5-8 and 3.5 for below 5.5. The optimum HRTs were 16 days for AnSBR and 17 days for Ax/Ox SBR with the corrected AUR2.
In order to characterize the nitrogen conversion characteristics in a thermophilic aerobic digestion (TAD) system, a laboratory study has been conducted with the analysis of effluent gas and microbial community in the sludge samples. The lab TAD system was operated with HRT of 3 days and 60 degrees C. Based on the nitrogen mass balance, it has been found that about 2/3 of the daily load of nitrogen was converted to the gaseous form of nitrogen whereas cellular transformation and unmetabolized nitrogen accounted for about 1/3. Among the gaseous nitrogen transformation, significant amount of influent nitrogen had been converted to N2 gas (29% of influent N) and N2O (9% of influent N). Ammonia conversion was only 28% of influent N. The detection of N2O gas is a clear indication of the biological nitrogen reduction process in the thermophilic aerobic digester. No conclusive evidence for the existence of aerobic deammonification has been found. The microbial community analysis showed that thermophilic bacteria such as Bacillus thermocloacae, Bacillus sp. and Clostridial groups dominated in this TAD reactor. The diverse microbial community in TAD sludge may play an important role in removing both strong organics and nitrogen from piggery waste.
The main concern of user-guided segmentation (UGS) is to achieve high segmentation accuracy with minimal user interaction. A novel convolutional neural network (CNN)-based UGS method is proposed, which employs a single click as the user interaction. In the proposed method, the input image in the Cartesian coordinate system is first converted into the polar transformed image with the user-guided point (UGP) as the origin of the polar coordinate system. The transformed image not only effectively delivers the UGP to the CNN, but also enables a single-scale convolution kernel to act as a multi-scale kernel, whose receptive field in the Cartesian coordinate system is altered based on the UGP without any extra parameters. In addition, a feature selection module (FSM) is introduced and utilised to additionally extract radial and angular features from the polar transformed image. Experimental results demonstrate that the proposed CNN using the polar transformed image improves the segmentation accuracy (mean intersection over union) by 3.69% on PASCAL VOC 2012 dataset compared with the CNN using the Cartesian coordinate image. The FSM achieves additional performance improvement of 1.32%. Moreover, the proposed method outperforms the conventional non-CNN-based UGS methods by 12.61% on average.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.