Urban flooding is escalating worldwide due to the increasing impervious surfaces from urban developments and frequency of extreme rainfall events by climate change. Traditional flood extent prediction and mapping meth...
详细信息
Urban flooding is escalating worldwide due to the increasing impervious surfaces from urban developments and frequency of extreme rainfall events by climate change. Traditional flood extent prediction and mapping methods based on physical-based hydrological principles often face limitations due to model complexity and computational burden. In response to these challenges, there has been a notable shift toward satellite imageprocessing and Artificial Intelligence (AI) based approaches, such as deeplearning (DL) models, including architectures like Convolutional Neural Networks (CNN). The objective of this research is to predict near real-time (NRT) flood extents within urban areas. This research integrated CNN (U-Net) with Sentinel-1 satellite imagery, Digital Elevation Model (DEM), hydrologic soil group (HSG), imperviousness, and rainfall data to create a flood extent prediction model. To detect flooded areas, a binary raster map was created using calibrated backscatter values derived from the VV (vertical transmit and vertical receive) polarization mode of Sentinel-1 imagery, which was highlighted as having a significant impact on backscatter behavior and prediction results. Application of the model was demonstrated in urban areas of Miami-Dade County, Florida. The results demonstrated the capability of the model to provide rapid and accurate flood extent predictions at a spatial resolution of 10 m, with an overall accuracy of 97.05 %, F-1 Score of 92.49 %, and AUC of 93 % in the study area. The U-Net model's flood predictions were compared with historical floodplain data and then using GIS overlay analysis, resulting in a Ground Truth Index of 84.05 % that shows the accuracy of the model in identifying flooded areas. The research incorporated crucial flood-influencing data (including rainfall) to the flood extent prediction models and expanded the focus models beyond major rainfall events only to encompass a wider range of flood events. The presented NRT flood e
Sludge morphology plays a vital role in activated sludge (AS) settleability. However, the presence of filamentous bacteria has been a significant issue that degrades the performance of AS systems. In this study, a cos...
详细信息
Sludge morphology plays a vital role in activated sludge (AS) settleability. However, the presence of filamentous bacteria has been a significant issue that degrades the performance of AS systems. In this study, a cost-effective setup with an automated stage was used to capture AS floc images from three wastewater treatment plants in Sapporo, Japan, from October 2021 to September 2022. The AS floc parameters, such as aspect ratio, roundness, equivalent diameter, and the area of filamentous and weak flocs (A F &Wfs-QIA ), were quantified using imageprocessing software. The A F &Wfs-QIA parameter served as a quantitative measure of filaments in the software- based approach. Indications of filamentous bulking in spring and winter were observed by analyzing the correlation between the sludge volume index (SVI) and the morphological parameters. This study successfully demonstrates that with an increase in the fraction of small and less dense flocs, the SVI elevates. Furthermore, sludge with excess filamentous bacteria made supernatants cleaner in the settling tests. These results led to a new hypothesis involving a suspended-solid-trapping mechanism with filamentous bacteria. In addition, this study extends its scope to the segmentation and detection of areas covered by filaments through a U-Net deep-learning multi-label semantic segmentation technique. In the U-Net-based method, the area of filaments (A F-UNet ) was used to quantify protruding filaments. It achieved performance comparable to that in the literature with a minimal training dataset while reducing the processingtime. Both the software-based and U-Net-based approaches were evaluated against SVI to assess the impact of filament growth on sludge settleability.
Background and objective: Early detection and grading of Diabetic Retinopathy (DR) is essential to determine an adequate treatment and prevent severe vision loss. However, the manual analysis of fundus images is time ...
详细信息
Background and objective: Early detection and grading of Diabetic Retinopathy (DR) is essential to determine an adequate treatment and prevent severe vision loss. However, the manual analysis of fundus images is time consuming and DR screening programs are challenged by the availability of human graders. Current automatic approaches for DR grading attempt the joint detection of all signs at the same time. However, the classification can be optimized if red lesions and bright lesions are independently processed since the task gets divided and simplified. Furthermore, clinicians would greatly benefit from explainable artificial intelligence (XAI) to support the automatic model predictions, especially when the type of lesion is specified. As a novelty, we propose an endto-end deeplearning framework for automatic DR grading (5 severity degrees) based on separating the attention of the dark structures from the bright structures of the retina. As the main contribution, this approach allowed us to generate independent interpretable attention maps for red lesions, such as microaneurysms and hemorrhages, and bright lesions, such as hard exudates, while using image-level labels only. Methods: Our approach is based on a novel attention mechanism which focuses separately on the dark and the bright structures of the retina by performing a previous image decomposition. This mechanism can be seen as a XAI approach which generates independent attention maps for red lesions and bright lesions. The framework includes an image quality assessment stage and deeplearning-related techniques, such as data augmentation, transfer learning and fine-tuning. We used the architecture Xception as a feature extractor and the focal loss function to deal with data imbalance. Results: The Kaggle DR detection dataset was used for method development and validation. The proposed approach achieved 83.7 % accuracy and a Quadratic Weighted Kappa of 0.78 to classify DR among 5 severity degrees, which outp
This paper presents an innovative approach to automatic volume control using imageprocessing and deeplearning techniques. The ability to automatically adjust volume levels based on environmental factors and user pre...
详细信息
deeplearning (DL) being popularly used in computer vision applications is still in its early stage in chemometric domain for spectral imageprocessing. Often the challenge is that there are too few samples from analy...
详细信息
deeplearning (DL) being popularly used in computer vision applications is still in its early stage in chemometric domain for spectral imageprocessing. Often the challenge is that there are too few samples from analytical laboratory experiments to preform DL. In this study, we present a novel combination of DL and chemometrics to process spectral images even with as few as < 100 spectral images. We divided the imageprocessing part such as object detection and recognition as the DL task and prediction of chemical property as the chemometric task based on latent space modelling. For imageprocessing tasks of object detection and recognition, transfer learning was performed on the pretrained YOLOv4 object detection network weights to adapt the model to work well on spectral images captured in laboratory settings. Once the object is identified with DL, a background query is performed for the pre-built chemometric models to select the model for predicting the properties for specific object. The obtained results showed good potential of using DL and chemometric approaches in conjunction to reap the best of both scientific domains. This approach is of high interest to whoever involved in spectral imaging and dealing with object detection and physicochemical properties prediction of the samples with chemometric approaches.(c) 2022 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://***/licenses/by/4.0/).
Garbage collection in urban areas has become a major challenge due to the increase in trash production. New technologies, including the application of deeplearning and imageprocessing methods, have been created to s...
详细信息
Quite possibly the most and best measures to contain the new popular episode is that the upkeep of the purported Social Distancing. The widespread Covid infection 2019 (COVID-19) has carried worldwide emergency with i...
详细信息
This paper describes how advanced deeplearning based computer vision algorithms are applied to enable real-time on-board sensor processing for small UAVs. Four use cases are considered: target detection, classificati...
详细信息
Agricultural robots are rapidly becoming more advanced with the development of relevant technologies and in great demand to guarantee food supply. As such, they are slated to play an important role in precision agricu...
详细信息
Agricultural robots are rapidly becoming more advanced with the development of relevant technologies and in great demand to guarantee food supply. As such, they are slated to play an important role in precision agriculture. For tomato production, harvesting employs over 40% of the total workforce. Therefore, it is meaningful to develop a robot harvester to assist workers. The objective of this work is to understand the factors restricting the recognition accuracy using imageprocessing and deeplearning methods, and improve the performance of crop detection in agricultural complex environment. With the accurate recognition of the growing status and location of crops, temporal management of the crop and selective harvesting can be available, and issues caused by the growing shortage of agricultural labour can be alleviated. In this respect, this work integrates the classic imageprocessing methods with the YOLOv5 (You only look once version 5) network to increase the accuracy and robustness of tomato and stem perception. As a consequence, an algorithm to estimate the degree of maturity of truss tomatoes (clusters of individual tomatoes) and an integrated method to locate stems based on the resultant experiments error of each individual method were proposed. Both indoor and real-filed tests were carried out using a robot harvester. The results proved the high accuracy of the proposed algorithms under varied illumination conditions, with an average deviation of 2 mm from the ground-truth. The robot can be guided to harvest truss tomatoes efficiently, with an average operating time of 9 s/cluster.
暂无评论