A study of cloud classification with neural networks using spectral and textural features

Tian, Bin; Shaikh, Mohammad Abuzar; Azimi-Sadjadi, M.R.; Haar, Thomas H. Vonder; Reinke, Donald L.

doi:10.1109/72.737500

Cited by 142 publications

(17 citation statements)

References 39 publications

(62 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, to overcome the difficulties in dealing with non-linearity and complexity of atmospheric prediction, researchers studied the use of artificial intelligence models, such as neural networks, as surrogate methods to traditional methods. Early studies applied these soft techniques to cloud classification [63][64][65][66], wind speed forecast [67][68][69], and storm prediction [70,71].…”

Section: Atmospheric-relatedmentioning

confidence: 99%

Estimation of renewable energy and built environment-related variables using neural networks – A review

Rodrigues

Gomes

Gaspar

et al. 2018

Renewable and Sustainable Energy Reviews

View full text Add to dashboard Cite

This paper presents a review on the application of neural networks for the estimation, forecasting, monitoring, and classification of exogenous environmental variables that affect the performance, salubrity, and security of cities, buildings, and infrastructures. The forecast of these variables allows to explore renewable energy and water resources, to prevent potentially hazardous construction locations, and to find the healthiest places, thus promoting a more sustainable future. Five research themes are covered-solar, atmospheric, hydrologic, geologic, and climate change. The solar section comprises solar radiation, direct and diffuse radiation, infrared and ultraviolet radiation, clearness index, and sky luminance and luminous efficacy. The atmospheric section reviews wind, temperature, humidity, cloud classification, and storm prediction. The hydrologic section focuses on precipitation, rainfall-runoff, hail, snow, drought, flood, tides, water levels, and other variables. The geologic section covers works on landslides, earthquakes, liquefaction, erosion, soil classification, soil mechanics, and other properties. Finally, climate change forecasting and downscaling of climate models are reviewed. This work demonstrates the wide range of applications of these methods in different research fields. Some research gaps and interdisciplinary research opportunities are identified for future development of comprehensive forecast and evaluation approaches regarding the estimation of renewable energy and built environment-related variables.

show abstract

Section: Atmospheric-relatedmentioning

confidence: 99%

Estimation of renewable energy and built environment-related variables using neural networks – A review

Rodrigues

Gomes

Gaspar

et al. 2018

Renewable and Sustainable Energy Reviews

View full text Add to dashboard Cite

show abstract

“…The results were compared to those obtained from Fisher discriminant function showing that numerical results illustrate the capabilities of ANNs in solving linear and nonlinear classification problems. Bin Tian et al [5] proposed in their study of cloud data classification from satellite imagery using neural networks with spectral and textural features. Two different neural-network paradigms namely probability neural network (PNN) and unsupervised Kohonen self-organized feature map (SOM) were examined in the study.…”

Section: Introductionmentioning

confidence: 99%

Performance Comparison of Different Multilayer Perceptron Network Activation Functions in Automated Weather Classification

Isa

Omar

Saad

et al. 2010

2010 Fourth Asia International Conference on Mathematical/Analytical Modelling and Computer Simulation

View full text Add to dashboard Cite

Multilayer perceptron network (MLP) has been recognized as a powerful tool for many applications including classification. Selection of the activation functions in the multilayer perceptron (MLP) network plays an essential role on the network performance. This paper presents a comparison study of two commonly used MLP activation function; sigmoid and hyperbolic tangent for weather classification. Meteorological data such as solar radiation, ambient temperature, current, surface temperature, voltage, wind direction and wind speed data are acquired from a photovoltaic (PV) system. Then, the meteorological data are input to the MLP network to classify the weather condition. In this study, weather conditions are classified into four types; rain, cloudy, dry day and storm. Levenberg-Marquardt algorithm is used to train the MLP network since it is the fastest training and ensure the best converges towards a minimum error. Experimental results show that hyperbolic tangent activation function is more efficient compared to sigmoid activation function. The MLP network using hyperbolic tangent function has achieved higher classification accuracy with less number of hidden nodes compared to sigmoid activation function.

show abstract

“…Yhann and Simpson [10] combined top-of-atmosphere reflectance and radiance from the National Oceanic and Atmospheric Administration's (NOAA) Advanced Very High-Resolution Radiometer (AVHRR) to detect cloudy pixels. Bankert [23] and Tian et al [24] investigated the performance of neural-network-based cloud classifications on AVHRR measurements and multispectral GOES-8 satellite imagery, respectively. Cloud Automated Neural Network (CANN) was presented by Miller and Emery [25] which defined different thresholds along with six textural features to distinguish between nine cloud classes.…”

Section: Introductionmentioning

confidence: 99%

“…Supervised neural networks classifiers are flexible and fast in learning [34] and they reveal their capability of dealing with consecutive images from GEO satellites with promising results for meteorological application [24,31,35]. In recent years, deep learning algorithms have been a popular solution to overcome the complexity of real-time data mining problems in earth and atmospheric sciences.…”

Section: Introductionmentioning

confidence: 99%

Deep Neural Network Cloud-Type Classification (DeepCTC) Model and Its Application in Evaluating PERSIANN-CCS

et al. 2020

View full text Add to dashboard Cite

Satellite remote sensing plays a pivotal role in characterizing hydrometeorological components including cloud types and their associated precipitation. The Cloud Profiling Radar (CPR) on the Polar Orbiting CloudSat satellite has provided a unique dataset to characterize cloud types. However, data from this nadir-looking radar offers limited capability for estimating precipitation because of the narrow satellite swath coverage and low temporal frequency. We use these high-quality observations to build a Deep Neural Network Cloud-Type Classification (DeepCTC) model to estimate cloud types from multispectral data from the Advanced Baseline Imager (ABI) onboard the GOES-16 platform. The DeepCTC model is trained and tested using coincident data from both CloudSat and ABI over the CONUS region. Evaluations of DeepCTC indicate that the model performs well for a variety of cloud types including Altostratus, Altocumulus, Cumulus, Nimbostratus, Deep Convective and High clouds. However, capturing low-level clouds remains a challenge for the model. Results from simulated GOES-16 ABI imageries of the Hurricane Harvey event show a large-scale perspective of the rapid and consistent cloud-type monitoring is possible using the DeepCTC model. Additionally, assessments using half-hourly Multi-Radar/Multi-Sensor (MRMS) precipitation rate data (for Hurricane Harvey as a case study) show the ability of DeepCTC in identifying rainy clouds, including Deep Convective and Nimbostratus and their precipitation potential. We also use DeepCTC to evaluate the performance of the Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks-Cloud Classification System (PERSIANN-CCS) product over different cloud types with respect to MRMS referenced at a half-hourly time scale for July 2018. Our analysis suggests that DeepCTC provides supplementary insights into the variability of cloud types to diagnose the weakness and strength of near real-time GEO-based precipitation retrievals. With additional training and testing, we believe DeepCTC has the potential to augment the widely used PERSIANN-CCS algorithm for estimating precipitation.

show abstract

A study of cloud classification with neural networks using spectral and textural features

Cited by 142 publications

References 39 publications

Estimation of renewable energy and built environment-related variables using neural networks – A review

Estimation of renewable energy and built environment-related variables using neural networks – A review

Performance Comparison of Different Multilayer Perceptron Network Activation Functions in Automated Weather Classification

Deep Neural Network Cloud-Type Classification (DeepCTC) Model and Its Application in Evaluating PERSIANN-CCS

Contact Info

Product

Resources

About