scholarly journals Mapping the Topographic Features of Mining-Related Valley Fills Using Mask R-CNN Deep Learning and Digital Elevation Data

2020 ◽  
Vol 12 (3) ◽  
pp. 547 ◽  
Author(s):  
Aaron E. Maxwell ◽  
Pariya Pourmohammadi ◽  
Joey D. Poyner
Keyword(s):  
Deep Learning ◽  
Predictor Variable ◽  
Training Data ◽  
Lidar Data ◽  
Eastern United States ◽  
Topographic Features ◽  
Model Generalization ◽  
Wide Range ◽  
Elevation Data ◽  
Study Sites

Modern elevation-determining remote sensing technologies such as light-detection and ranging (LiDAR) produce a wealth of topographic information that is increasingly being used in a wide range of disciplines, including archaeology and geomorphology. However, automated methods for mapping topographic features have remained a significant challenge. Deep learning (DL) mask regional-convolutional neural networks (Mask R-CNN), which provides context-based instance mapping, offers the potential to overcome many of the difficulties of previous approaches to topographic mapping. We therefore explore the application of Mask R-CNN to extract valley fill faces (VFFs), which are a product of mountaintop removal (MTR) coal mining in the Appalachian region of the eastern United States. LiDAR-derived slopeshades are provided as the only predictor variable in the model. Model generalization is evaluated by mapping multiple study sites outside the training data region. A range of assessment methods, including precision, recall, and F1 score, all based on VFF counts, as well as area- and a fuzzy area-based user’s and producer’s accuracy, indicate that the model was successful in mapping VFFs in new geographic regions, using elevation data derived from different LiDAR sensors. Precision, recall, and F1-score values were above 0.85 using VFF counts while user’s and producer’s accuracy were above 0.75 and 0.85 when using the area- and fuzzy area-based methods, respectively, when averaged across all study areas characterized with LiDAR data. Due to the limited availability of LiDAR data until relatively recently, we also assessed how well the model generalizes to terrain data created using photogrammetric methods that characterize past terrain conditions. Unfortunately, the model was not sufficiently general to allow successful mapping of VFFs using photogrammetrically-derived slopeshades, as all assessment metrics were lower than 0.60; however, this may partially be attributed to the quality of the photogrammetric data. The overall results suggest that the combination of Mask R-CNN and LiDAR has great potential for mapping anthropogenic and natural landscape features. To realize this vision, however, research on the mapping of other topographic features is needed, as well as the development of large topographic training datasets including a variety of features for calibrating and testing new methods.

scholarly journals Unsupervised content-preserving transformation for optical microscopy

2021 ◽  
Vol 10 (1) ◽  
Author(s):  
Xinyang Li ◽  
Guoxun Zhang ◽  
Hui Qiao ◽  
Feng Bao ◽  
Yue Deng ◽  
...  
Keyword(s):  
Deep Learning ◽  
Optical Microscopy ◽  
Training Data ◽  
Fluorescence Labeling ◽  
Imaging Data ◽  
Image Transformation ◽  
General Applicability ◽  
Data Annotation ◽  
Biomedical Image ◽  
Wide Range

AbstractThe development of deep learning and open access to a substantial collection of imaging data together provide a potential solution for computational image transformation, which is gradually changing the landscape of optical imaging and biomedical research. However, current implementations of deep learning usually operate in a supervised manner, and their reliance on laborious and error-prone data annotation procedures remains a barrier to more general applicability. Here, we propose an unsupervised image transformation to facilitate the utilization of deep learning for optical microscopy, even in some cases in which supervised models cannot be applied. Through the introduction of a saliency constraint, the unsupervised model, named Unsupervised content-preserving Transformation for Optical Microscopy (UTOM), can learn the mapping between two image domains without requiring paired training data while avoiding distortions of the image content. UTOM shows promising performance in a wide range of biomedical image transformation tasks, including in silico histological staining, fluorescence image restoration, and virtual fluorescence labeling. Quantitative evaluations reveal that UTOM achieves stable and high-fidelity image transformations across different imaging conditions and modalities. We anticipate that our framework will encourage a paradigm shift in training neural networks and enable more applications of artificial intelligence in biomedical imaging.

scholarly journals Improving generalization of deep learning models for diagnostic pathology by increasing variability in training data: experiments on osteosarcoma subtypes

2020 ◽  
Author(s):  
Haiming Tang ◽  
Nanfei Sun ◽  
Steven Shen
Keyword(s):  
Deep Learning ◽  
Model Performance ◽  
High Variability ◽  
Training Data ◽  
Classification Model ◽  
Training Dataset ◽  
Learning Models ◽  
Diagnostic Pathology ◽  
Model Generalization ◽  
Histopathological Images

Artificial intelligence (AI) has an emerging progress in diagnostic pathology. A large number of studies of applying deep learning models to histopathological images have been published in recent years. While many studies claim high accuracies, they may fall into the pitfalls of overfitting and lack of generalization due to the high variability of the histopathological images. We use the example of Osteosarcoma to illustrate the pitfalls and how the addition of model input variability can help improve model performance. We use the publicly available osteosarcoma dataset to retrain a previously published classification model for osteosarcoma. We partition the same set of images into the training and testing datasets differently than the original study: the test dataset consists of images from one patient while the training dataset consists images of all other patients. The performance of the model on the test set using the new partition schema declines dramatically, indicating a lack of model generalization and overfitting.We also show the influence of training data variability on model performance by collecting a minimal dataset of 10 osteosarcoma subtypes as well as benign tissues and benign bone tumors of differentiation. We show the additions of more and more subtypes into the training data step by step under the same model schema yield a series of coherent models with increasing performances. In conclusion, we bring forward data preprocessing and collection tactics for histopathological images of high variability to avoid the pitfalls of overfitting and build deep learning models of higher generalization abilities.

scholarly journals Intensity Thresholding and Deep Learning Based Lane Marking Extraction and Lane Width Estimation from Mobile Light Detection and Ranging (LiDAR) Point Clouds

2020 ◽  
Vol 12 (9) ◽  
pp. 1379 ◽  
Author(s):  
Yi-Ting Cheng ◽  
Ankit Patel ◽  
Chenglu Wen ◽  
Darcy Bullock ◽  
Ayman Habib
Keyword(s):  
Deep Learning ◽  
Learning Strategies ◽  
Point Clouds ◽  
Concrete Pavement ◽  
Training Data ◽  
Light Detection And Ranging ◽  
Light Detection ◽  
Lane Marking ◽  
Wide Range ◽  
Lane Width

Lane markings are one of the essential elements of road information, which is useful for a wide range of transportation applications. Several studies have been conducted to extract lane markings through intensity thresholding of Light Detection and Ranging (LiDAR) point clouds acquired by mobile mapping systems (MMS). This paper proposes an intensity thresholding strategy using unsupervised intensity normalization and a deep learning strategy using automatically labeled training data for lane marking extraction. For comparative evaluation, original intensity thresholding and deep learning using manually established labels strategies are also implemented. A pavement surface-based assessment of lane marking extraction by the four strategies is conducted in asphalt and concrete pavement areas covered by MMS equipped with multiple LiDAR scanners. Additionally, the extracted lane markings are used for lane width estimation and reporting lane marking gaps along various highways. The normalized intensity thresholding leads to a better lane marking extraction with an F1-score of 78.9% in comparison to the original intensity thresholding with an F1-score of 72.3%. On the other hand, the deep learning model trained with automatically generated labels achieves a higher F1-score of 85.9% than the one trained on manually established labels with an F1-score of 75.1%. In concrete pavement area, the normalized intensity thresholding and both deep learning strategies obtain better lane marking extraction (i.e., lane markings along longer segments of the highway have been extracted) than the original intensity thresholding approach. For the lane width results, more estimates are observed, especially in areas with poor edge lane marking, using the two deep learning models when compared with the intensity thresholding strategies due to the higher recall rates for the former. The outcome of the proposed strategies is used to develop a framework for reporting lane marking gap regions, which can be subsequently visualized in RGB imagery to identify their cause.

scholarly journals Hybrid Deep Learning Models for Sentiment Analysis

Complexity ◽  
2021 ◽  
Vol 2021 ◽  
pp. 1-16
Author(s):  
Cach N. Dang ◽  
María N. Moreno-García ◽  
Fernando De la Prieta
Keyword(s):  
Deep Learning ◽  
Sentiment Analysis ◽  
Short Term Memory ◽  
Computation Time ◽  
Hybrid Models ◽  
Training Data ◽  
Support Vector ◽  
Learning Models ◽  
Hybrid Techniques ◽  
Wide Range

Sentiment analysis on public opinion expressed in social networks, such as Twitter or Facebook, has been developed into a wide range of applications, but there are still many challenges to be addressed. Hybrid techniques have shown to be potential models for reducing sentiment errors on increasingly complex training data. This paper aims to test the reliability of several hybrid techniques on various datasets of different domains. Our research questions are aimed at determining whether it is possible to produce hybrid models that outperform single models with different domains and types of datasets. Hybrid deep sentiment analysis learning models that combine long short-term memory (LSTM) networks, convolutional neural networks (CNN), and support vector machines (SVM) are built and tested on eight textual tweets and review datasets of different domains. The hybrid models are compared against three single models, SVM, LSTM, and CNN. Both reliability and computation time were considered in the evaluation of each technique. The hybrid models increased the accuracy for sentiment analysis compared with single models on all types of datasets, especially the combination of deep learning models with SVM. The reliability of the latter was significantly higher.

scholarly journals Stacking Ensemble Methodology Using Deep Learning and ARIMA Models for Short-Term Load Forecasting

Energies ◽  
2021 ◽  
Vol 14 (21) ◽  
pp. 7378
Author(s):  
Pedro M. R. Bento ◽  
Jose A. N. Pombo ◽  
Maria R. A. Calado ◽  
Silvio J. P. S. Mariano
Keyword(s):  
Deep Learning ◽  
New England ◽  
Network Architecture ◽  
Prediction Models ◽  
Load Forecasting ◽  
Training Data ◽  
Constant Field ◽  
Short Term ◽  
Wide Range ◽  
Short Term Load Forecasting

Short-Term Load Forecasting is critical for reliable power system operation, and the search for enhanced methodologies has been a constant field of investigation, particularly in an increasingly competitive environment where the market operator and its participants need to better inform their decisions. Hence, it is important to continue advancing in terms of forecasting accuracy and consistency. This paper presents a new deep learning-based ensemble methodology for 24 h ahead load forecasting, where an automatic framework is proposed to select the best Box-Jenkins models (ARIMA Forecasters), from a wide-range of combinations. The method is distinct in its parameters but more importantly in considering different batches of historical (training) data, thus benefiting from prediction models focused on recent and longer load trends. Afterwards, these accurate predictions, mainly the linear components of the load time-series, are fed to the ensemble Deep Forward Neural Network. This flexible type of network architecture not only functions as a combiner but also receives additional historical and auxiliary data to further its generalization capabilities. Numerical testing using New England market data validated the proposed ensemble approach with diverse base forecasters, achieving promising results in comparison with other state-of-the-art methods.

scholarly journals LiDAR Odometry by Deep Learning-based Feature Points with Two-step Pose Estimation

2021 ◽  
Author(s):  
Tianyi Liu ◽  
Yan Wang ◽  
xiaoji niu ◽  
Chang Le ◽  
Tisheng Zhang ◽  
...  
Keyword(s):  
Deep Learning ◽  
Field Experiment ◽  
Pose Estimation ◽  
Ground Truth ◽  
Feature Point ◽  
Training Data ◽  
Lidar Data ◽  
Feature Points ◽  
Experiment Data ◽  
Research And Innovation

KITTI dataset is collected from three types of environments, i.e., country, urban and highway The types of feature point cover a variety of scenes. The KITTI dataset provides 22 sequences of LiDAR data. 11 sequences of them from sequence 00 to sequence 10 are "training" data. The training data are provided with ground truth translation and rotation. In addition, field experiment data is collected by low-resolution LiDAR, VLP-16 in Wuhan Research and Innovation Center.

scholarly journals Pano-RSOD: A Dataset and Benchmark for Panoramic Road Scene Object Detection

Electronics ◽  
2019 ◽  
Vol 8 (3) ◽  
pp. 329 ◽  
Author(s):  
Yong Li ◽  
Guofeng Tong ◽  
Huashuai Gao ◽  
Yuebin Wang ◽  
Liqiang Zhang ◽  
...  
Keyword(s):  
Deep Learning ◽  
Object Detection ◽  
Training Data ◽  
Training Dataset ◽  
Panoramic Image ◽  
Information Object ◽  
Wide Range ◽  
Panoramic Images ◽  
Bounding Boxes ◽  
Scene Object

Panoramic images have a wide range of applications in many fields with their ability to perceive all-round information. Object detection based on panoramic images has certain advantages in terms of environment perception due to the characteristics of panoramic images, e.g., lager perspective. In recent years, deep learning methods have achieved remarkable results in image classification and object detection. Their performance depends on the large amount of training data. Therefore, a good training dataset is a prerequisite for the methods to achieve better recognition results. Then, we construct a benchmark named Pano-RSOD for panoramic road scene object detection. Pano-RSOD contains vehicles, pedestrians, traffic signs and guiding arrows. The objects of Pano-RSOD are labelled by bounding boxes in the images. Different from traditional object detection datasets, Pano-RSOD contains more objects in a panoramic image, and the high-resolution images have 360-degree environmental perception, more annotations, more small objects and diverse road scenes. The state-of-the-art deep learning algorithms are trained on Pano-RSOD for object detection, which demonstrates that Pano-RSOD is a useful benchmark, and it provides a better panoramic image training dataset for object detection tasks, especially for small and deformed objects.

DeepHeart

2021 ◽  
Vol 17 (2) ◽  
pp. 1-18
Author(s):  
Xiangmao Chang ◽  
Gangkai Li ◽  
Guoliang Xing ◽  
Kun Zhu ◽  
Linlin Tu
Keyword(s):  
Signal Processing ◽  
Deep Learning ◽  
Spectrum Analysis ◽  
Absolute Error ◽  
Physical Activities ◽  
Motion Artifacts ◽  
Training Data ◽  
Data Set ◽  
Electrocardiogram Signals ◽  
Wide Range

Heart rate (HR) estimation based on photoplethysmography (PPG) signals has been widely adopted in wrist-worn devices. However, the motion artifacts caused by the user’s physical activities make it difficult to get the accurate HR estimation from contaminated PPG signals. Although many signal processing methods have been proposed to address this challenge, they are often highly optimized for specific scenarios, making them impractical in real-world settings where a user may perform a wide range of physical activities. In this article, we propose DeepHeart, a new HR estimation approach that features deep-learning-based denoising and spectrum-analysis-based calibration. DeepHeart generates clean PPG signals from electrocardiogram signals based on a training data set. Then a set of denoising convolutional neural networks (DCNNs) are trained with the contaminated PPG signals and their corresponding clean PPG signals. Contaminated PPG signals are then denoised by an ensemble of DCNNs and a spectrum-analysis-based calibration is performed to estimate the final HR. We evaluate DeepHeart on the IEEE Signal Processing Cup training data set with 12 records collected during various physical activities. DeepHeart achieves an average absolute error of 1.61 beats per minute (bpm), outperforming a state-of-the-art deep learning approach (4 bpm) and a classical signal processing approach (2.34 bpm).

scholarly journals Cellpose: a generalist algorithm for cellular segmentation

2020 ◽  
Author(s):  
Carsen Stringer ◽  
Tim Wang ◽  
Michalis Michaelos ◽  
Marius Pachitariu
Keyword(s):  
Deep Learning ◽  
Training Data ◽  
Data Repository ◽  
Biological Applications ◽  
Segmentation Method ◽  
Wide Range ◽  
Great Progress ◽  
Microscopy Images ◽  
2D Model

Many biological applications require the segmentation of cell bodies, membranes and nuclei from microscopy images. Deep learning has enabled great progress on this problem, but current methods are specialized for images that have large training datasets. Here we introduce a generalist, deep learning-based segmentation method called Cellpose, which can precisely segment cells from a wide range of image types and does not require model retraining or parameter adjustments. We trained Cellpose on a new dataset of highly-varied images of cells, containing over 70,000 segmented objects. We also demonstrate a 3D extension of Cellpose which reuses the 2D model and does not require 3D-labelled data. To support community contributions to the training data, we developed software for manual labelling and for curation of the automated results, with optional direct upload to our data repository. Periodically retraining the model on the community-contributed data will ensure that Cellpose improves constantly.

scholarly journals Brightness Invariant Deep Spectral Super-Resolution

Sensors ◽  
2020 ◽  
Vol 20 (20) ◽  
pp. 5789
Author(s):  
Tarek Stiebel ◽  
Dorit Merhof
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Scale Invariance ◽  
Super Resolution ◽  
Training Data ◽  
Essential Property ◽  
Practical Applications ◽  
Wide Range ◽  
Cheap Alternative ◽  
The Moment

Spectral reconstruction from RGB or spectral super-resolution (SSR) offers a cheap alternative to otherwise costly and more complex spectral imaging devices. In recent years, deep learning based methods consistently achieved the best reconstruction quality in terms of spectral error metrics. However, there are important properties that are not maintained by deep neural networks. This work is primarily dedicated to scale invariance, also known as brightness invariance or exposure invariance. When RGB signals only differ in their absolute scale, they should lead to identical spectral reconstructions apart from the scaling factor. Scale invariance is an essential property that signal processing must guarantee for a wide range of practical applications. At the moment, scale invariance can only be achieved by relying on a diverse database during network training that covers all possibly occurring signal intensities. In contrast, we propose and evaluate a fundamental approach for deep learning based SSR that holds the property of scale invariance by design and is independent of the training data. The approach is independent of concrete network architectures and instead focuses on reevaluating what neural networks should actually predict. The key insight is that signal magnitudes are irrelevant for acquiring spectral reconstructions from camera signals and are only useful for a potential signal denoising.

Sign in / Sign up

Export Citation Format

Share Document