Ground-based Cloud Image Research Articles

Ground-Based Cloud-Type Recognition Using Manifold Kernel Sparse Coding and Dictionary Learning

Recognizing cloud type of ground-based images automatically has a great influence on the weather service but poses a significant challenge. Based on the symmetric positive definite (SPD) matrix manifold, a novel method named “manifold kernel sparse coding and dictionary learning” (MKSCDL) is proposed for cloud classification. Different from classical features extracted in the Euclidean space, the SPD matrix fuses multiple features and represents non-Euclidean geometric characteristics. MKSCDL is composed of three steps: feature extraction, dictionary learning, and classification. With the learned dictionary, the SPD matrix of the cloud image can be described with the sparse code. The experiments are conducted on two different ground-based cloud image datasets. Benefitting from the sparse representation on the Riemannian matrix manifold, compared to the recent baselines, experimental results demonstrate that MKSCDL possesses a more competitive performance on both grayscale and colour image datasets.

A new ultra-short-term photovoltaic power prediction model based on ground-based cloud images

The cloud shading on the photovoltaic (PV) power station is one of the main factors that cause random changes in the PV output power, and thereby greatly influences an ultra-short-term photovoltaic power prediction. This paper presents an ultra-short-term prediction model for photovoltaic power generation based on dynamic characteristics of the cloud that is sheltering the sun. The proposed prediction model consists of three stages. In the first stage, the moving trajectory of the cloud is predicted using the motion vector and the cloud that shelters the sun is selected. In the second stage, the dynamic characteristics of target cloud, which have a great influence on the photovoltaic power generation, are extracted using the digital image processing. In the third stage, a prediction model based on the radial basis function (RBF) neural network, which is trained with processed sample data, is designed. Finally, the performance of RBF prediction model is compared with the performance of auto regressive (AR) model. The comparison shows that the power prediction accuracy of RBF model is 7.4% and the power prediction accuracy of AR model is 13.6%. The proposed ultra-short-term PV power prediction model can significantly improve the power prediction performance, especially in cloudy weather.

Salient Dual Activations Aggregation for Ground-Based Cloud Classification in Weather Station Networks

Since appearances of clouds are always changeable, ground-based cloud classification is still in urgent need of development in weather station networks. Many existing methods resort to convolutional neural networks to improve the classification accuracy. However, these methods just carry out the feature extraction from one convolutional layer, hence making it difficult to obtain complete information of ground-based cloud images. To address this limitation, in this paper, we propose a novel method named salient dual activations aggregation (SDA2) to extract ground-based cloud features from different convolutional layers, which could learn the structural, textural, and high-level semantic information for ground-based cloud representation, simultaneously. Specifically, the salient patch selection strategy is first applied to select salient vectors from one shallow convolutional layer. Then, corresponding weights are learned from one deep convolutional layer. After obtaining a set of salient vectors with various weights, this paper is further designed to aggregate them into a representative vector for each ground-based cloud image by explicitly modeling the relationship among salient vectors. The proposed SDA2 is validated on three ground-based cloud databases, and the experimental results prove its effectiveness. Especially, we obtain the promising classification results of 91.24% on the MOC_e database, 91.15% on the IAP_e database, and 88.73% on the CAMS_e database.

Short-term cloud coverage prediction using the ARIMA time series model

ABSTRACTIn view of the important role of cloud coverage on the solar (energy) irradiance, the total cloud coverage prediction based on ground-based cloud images is studied in this paper. In traditional prediction techniques, the correlation between cloud coverage over continue time is always neglected. Thus, an autoregressive integrated moving average (ARIMA) time series model is used to predict the short-term cloud coverage. Experimental results on a collected time series database of cloud coverage computed from ground-based cloud images show that the correlation information of time series is useful for cloud coverage prediction. Additionally, the ARIMA model gains a superior prediction performance for forecasts of one minute or longer 20 and 30 minutes. We are able to predict the cloud coverage with an approximate error of 5%, 7%, and 9% for 1, 5, and 20 and 30 minute forecasts, respectively. Furthermore, we found that there are different error rates of predictions for different cloud coverage intervals. High cloud coverage always suffers from a higher error rate.

DeepCloud: Ground-Based Cloud Image Categorization Using Deep Convolutional Features

Accurate ground-based cloud image categorization is a critical but challenging task that has not been well addressed. One of the essential issues that affect the performance is to extract the representative visual features. Nearly all of the existing methods rely on the hand-crafted descriptors (e.g., local binary patterns, CENsus TRsansform hISTogram, and scale-invariant feature transform). Their limited discriminative power indeed leads to the unsatisfactory performance. To alleviate this, we propose “DeepCloud” as a novel cloud image feature extraction approach by resorting to the deep convolutional visual features. In the recent years, the deep convolutional neural network (CNN) has achieved the promising results in lots of computer vision and image understanding fields. Nevertheless, it has not been applied to cloud image classification yet. Thus, we actually pay the first effort to fill this blank. Since cloud image classification can be attributed to a multi-instance learning problem, simply employing the convolutional features within CNN cannot achieve the promising result. To address this, Fisher vector encoding is applied to executing the spatial feature aggregation and high-dimensional feature mapping on the raw deep convolutional features. Moreover, the hierarchical convolutional layers are used simultaneously to capture the fine textural characteristics and high-level semantic information in the unified manner. To further leverage the performance, a cloud pattern mining and selection method are also proposed. It targets at finding the discriminative local patterns to better distinguish the different kinds of clouds. The experiments on a challenging ground-based cloud image data set demonstrate the superiority of the proposition over the state-of-the-art methods.

Measure for the Difference Between LBP Features Extracted From Original and Resized Cloud Images With Varying Resolutions

Currently, ground-based cloud images taken by using a whole-sky imager are especially popular in the field of meteorology because of their high resolution and accurate cloud information. Cloud images are natural texture images, and thus texture features based on local binary patterns (LBPs) are widely used to analyze texture images. However, the high-computation cost of extracting LBP features from high-resolution cloud texture images may make this technique unacceptable in practical image processing. A commonly adopted method is to resize the original image to an appropriate version with a decreased resolution. But this process will inevitably result in information loss. Accordingly, a measure based on the Kullback-Leibler (KL) divergence of the difference between LBP histogram features extracted from the original and resized images with varying resolutions is reported in this letter. Furthermore, a confidence interval technique is introduced to validate the significance of such difference. Experiments based on real ground-based cloud images show the measurement results of KL divergence in LBP features extracted from original and resized images. The experimental results indicate that images should be resized with caution when performing image processing.

Ground-based cloud classification using weighted local binary patterns

Ground-based cloud classification plays an essential role in meteorological research, and, recently, texture classification techniques have been introduced to automate the process. As a typical texture descriptor, local binary patterns (LBP) have emerged as a very powerful tool due to their effective representation ability. However, it neglects the local contrast information of ground-based cloud images, which may hinder the classification performance. We propose a descriptor called weighted local binary patterns (WLBP) for ground-based cloud classification. The proposed WLBP not only inherits the advantages of LBP but also encodes the useful contrast information of local structures. We define the variance of a local patch as a rotation invariant measure and use this measure as an adaptive weight to adjust the contribution of each neighboring pixel in the process of histogram accumulation. The experimental results demonstrate that the proposed WLBP achieves a better performance than the state-of-the-art methods.

Enhancement of Ground-based Cloud Image by Genetic Algorithm with SURF Points

The ground-based cloud image is a valuable data source for meteorological observation. For the purpose of cloud type recognition and cloud height measurement, it is essential to enhance the ground-based cloud image. However, exiting image enhancement approach may not achieve optimal performance in this application. In this paper, we propose a genetic algorithm based method to enhance the ground-based cloud image. Inspired by the interest points matching technique of stereo, we propose a new criterion that use SURF interest points to construct fitness function of the genetic algorithm. In addition, in order to increase the diversity of the population, we propose a novel initialization that contains two strategies: random numbers and specified numbers according to the regularized incomplete beta function. The experimental results show that our method is suitable for enhancing the ground-based cloud image compare with other existing methods.

Classifiers for Ground-Based Cloud Images Using Texture Features

The classification of ground-based cloud images has received more attention recently. The result of this work applies to the analysis of climate change; a correct classification is, therefore, important. In this paper, we used 18 texture features to distinguish 7 sky conditions. The important parameters of two classifiers are fine-tuned in the experiment, namely, k-nearest neighbor (k-NN) and artificial neural network (ANN). The performances of the two classifications were compared. Advantages and limitations of both classifiers were discussed. Our result revealed that the k-NN model performed at 72.99% accuracy while the ANN model has higher performance at 86.93% accuracy. We showed that our result is better than previous studies. Finally, seven most effective texture features are recommended to be used in the field of cloud type classification.

Salient local binary pattern for ground-based cloud classification

Ground-based cloud classification is challenging due to extreme variations in the appearance of clouds under different atmospheric conditions. Texture classification techniques have recently been introduced to deal with this issue. A novel texture descriptor, the salient local binary pattern (SLBP), is proposed for ground-based cloud classification. The SLBP takes advantage of the most frequently occurring patterns (the salient patterns) to capture descriptive information. This feature makes the SLBP robust to noise. Experimental results using ground-based cloud images demonstrate that the proposed method can achieve better results than current state-of-the-art methods.

An Automated Cirrus Cloud Detection Method for a Ground-Based Cloud Image

Abstract Cloud detection is a basic research for achieving cloud-cover state and other cloud characteristics. Because of the influence of sunlight, the brightness of sky background on the ground-based cloud image is usually nonuniform, which increases the difficulty for cirrus cloud detection, and few detection methods perform well for thin cirrus clouds. This paper presents an effective background estimation method to eliminate the influence of variable illumination conditions and proposes a background subtraction adaptive threshold method (BSAT) to detect cirrus clouds in visible images for the small field of view and mixed clear–cloud scenes. The BSAT algorithm consists of red-to-blue band operation, background subtraction, adaptive threshold selection, and binarization. The experimental results show that the BSAT algorithm is robust for all types of cirrus clouds, and the quantitative evaluation results demonstrate that the BSAT algorithm outperforms the fixed threshold (FT) and adaptive threshold (AT) methods in cirrus cloud detection.

Study on Very Short-term Solar Irradiance Forecasting Methods using Ground based Cloud Image

Due to the fluctuating and intermittent nature of solar irradiance, the power output of PV power station fluctuates. It brings great difficulties for power output forecasting of PV power station. Traditional PV power output forecasting technologies using physical/statistical model combined with NWP failed to provide accurate results. Based on the research of on-line analysis on ground-based cloud image captured by Total Sky Imager (TSI), through the technical phases of cloud recognition, cloud blocking prediction, distortion correction, projection transformation and integration with cloud forcing analysis, the roadmap of PV very short term solar irradiance forecasting based on ground-based cloud image analysis was presented in this paper. Theoretical analysis proves that the method tends to provide reasonable accurate forecasting results and can be widely used.