A5085556321- Retinal Imaging and Analysis
- Glaucoma and retinal disorders
- Digital Imaging for Blood Diseases
- Recycling and utilization of industrial and municipal waste in materials production
- Concrete and Cement Materials Research
- CO2 Sequestration and Geologic Interactions
- Retinal and Optic Conditions
- Grouting, Rheology, and Soil Mechanics
- Rock Mechanics and Modeling
- Cloud Data Security Solutions
- Adsorption and biosorption for pollutant removal
- Innovative concrete reinforcement materials
- Recycling and Waste Management Techniques
- TiO2 Photocatalysis and Solar Cells
- Municipal Solid Waste Management
- Medical Image Segmentation Techniques
- Iron and Steelmaking Processes
- Advanced Photocatalysis Techniques
- Gas Sensing Nanomaterials and Sensors
- Microbial Applications in Construction Materials
- Recycled Aggregate Concrete Performance
- Chemical Looping and Thermochemical Processes
- Graphite, nuclear technology, radiation studies
- Chromium effects and bioremediation
Xi'an University of Architecture and Technology
2020-2023
Northwest Normal University
2020-2021
Accurate segmentation of retinal blood vessels is a key step in the diagnosis fundus diseases, among which cataracts, glaucoma, and diabetic retinopathy (DR) are main diseases that cause blindness. Most methods based on deep convolutional neural networks can effectively extract features. However, convolution pooling operations also filter out some useful information, final segmented have problems such as low classification accuracy. In this paper, we propose multi-scale residual attention...
The accurate segmentation of retinal images is a basic step in screening for retinopathy and glaucoma. Most existing image methods have insufficient feature information extraction. They are susceptible to the impact lesion area poor quality, resulting recovery contextual information. This also causes results model be noisy low accuracy. Therefore, this paper proposes multi-scale multi-branch convolutional neural network (multi-scale (MSMB-Net)) segmentation. uses atrous convolution with...
Segmentation of retinal vessels is important for doctors to diagnose some diseases. The segmentation accuracy can be effectively improved by using deep learning methods. However, most the existing methods are incomplete shallow feature extraction, and superficial features lost, resulting in blurred vessel boundaries inaccurate capillaries results. At same time, "layer-by-layer" information fusion between encoder decoder makes extracted from layer network cannot smoothly transferred network,...
With the increasing power generation capacity of circulating fluidized bed boilers and electrolytic aluminum production worldwide, emissions coal combustion fly ash (FBCF) bayer red mud (BRM) have increased dramatically, resulting in extremely severe environmental issues as it is difficult to utilize these resources. To address this issue, study uses industrial solid wastes such BRM, FBCF, silica fume mineral admixtures develop a green low-carbon early strength multi-solid waste composite...
Accurate segmentation of retinal vessels is critical to the mechanism, diagnosis, and treatment many ocular pathologies. Due poor contrast inhomogeneous background fundus imaging complex structure images, this makes accurate blood from images still challenging. In paper, we propose an effective framework for vascular segmentation, which innovative mainly in image pre-processing stage stage. First, perform enhancement on three publicly available datasets based multiscale retinex with color...
Retinal vessel segmentation has high value for the research on diagnosis of diabetic retinopathy, hypertension, and cardiovascular cerebrovascular diseases. Most methods based deep convolutional neural networks (DCNN) do not have large receptive fields or rich spatial information cannot capture global context larger areas. Therefore, it is difficult to identify lesion area, efficiency poor. This paper presents a butterfly fully network (BFCN). First, in view low contrast between blood...
At present, the rising level of carbon dioxide (CO2) in atmosphere has become a global concern, which urges researchers to find possible solutions reduce or capture CO2 emissions. Mineralization is an important method for reducing In this study, investigation provided opportunity advance understanding improving mechanism spherical ultrafine fly ash (UFA) on mineralization waste slag (WS). The results show that adding UFA could significantly improve efficiency(E), with increased rate was...