Development of an all-sky imaging system for cloud cover assessment

Fa, Tao; Xie, Wanyi; Wang, Yiren; Xia, Yingwei

doi:10.1364/ao.58.005516

Cited by 25 publications

(21 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In contrast with expert-designed algorithms for TCC retrieval, just a few data-driven schemes were presented lately for estimating TCC [49][50][51] or for clouds segmentation in optical all-sky imagery [52]. The researchers my consider the problem of TCC retrieval being solved or to be too simple for being addressed with complex machine learning algorithms.…”

Section: On the Optimization Nature Of Known Schemes For Tcc Retrievamentioning

confidence: 99%

“…In case of the schemes employing machine learning (ML) methods [49][50][51], the algorithms are inherently of optimization nature since the essence of almost any supervised ML algorithm is the optimization of empirical costs based on train data set.…”

Section: On the Optimization Nature Of Known Schemes For Tcc Retrievamentioning

confidence: 99%

“…A few data-driven methods were presented lately for estimating TCC from all-sky optical imagery [49,50,52]. In [50], the only improvement compared to simple schemes [33] is the application of clustering algorithm in the form of superpixel segmentation step.…”

Section: On the Data-driven Algorithms For Tcc Retrieval From All-skymentioning

confidence: 99%

See 2 more Smart Citations

On the Generalization Ability of Data-Driven Models in the Problem of Total Cloud Cover Retrieval

Krinitskiy¹,

Aleksandrova²,

Verezemskaya³

et al. 2020

Preprint

View full text Add to dashboard Cite

Total cloud cover (TCC) retrieval from ground-based optical imagery is a problem being tackled by a few generations of researchers. The number of human-designed algorithms for the estimation of TCC grows every year. However, there is not very much progress in terms of quality, mostly due to the lack of systematic approach to the design of the algorithms, to the assessment of their generalization ability, and to the assessment of the TCC retrieval quality. In this study, we discuss the optimization nature of data-driven schemes for TCC retrieval. In order to compare the algorithms, we propose the framework for the assessment of the algorithms characteristics. We present several new algorithms that are based on deep learning techniques: a model for outliers filtering, and a few models for TCC retrieval from all-sky imagery. For training and assessment of data-driven algorithms of this study, we present the Dataset of All-Sky Imagery over the Ocean (DASIO) containing over one million of all-sky optical images of visible sky dome taken in various regions of the World Ocean. The research campaigns contributed to DASIO collection took place in the Atlantic ocean, the Indian ocean the Red and Mediterranean seas, and also in the Arctic ocean. Optical imagery collected during these missions are accompanied by standard meteorological observations of cloudiness characteristics made by experienced observers. We assess the generalization ability of the presented models in several scenarios that differ in terms of the regions selected for the train and validation subsets. As a result, we demonstrate that our models based on convolutional neural networks deliver superior quality compared to all previously published schemes. As a key result, we demonstrate considerable drop of the ability to generalize the training data in case of strong covariate shift between training and validation subsets of imagery which may take place in case of region-aware subsampling.

show abstract

Section: On the Optimization Nature Of Known Schemes For Tcc Retrievamentioning

confidence: 99%

Section: On the Optimization Nature Of Known Schemes For Tcc Retrievamentioning

confidence: 99%

See 1 more Smart Citation

On the Generalization Ability of Data-Driven Models in the Problem of Total Cloud Cover Retrieval

Krinitskiy¹,

Aleksandrova²,

Verezemskaya³

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…In this work we use the sky images captured by the ground-based all-sky-view camera modeled as ASC100. The ASC100 is developed by Hefei Institute of Physical Science, Chinese Academy of Sciences (Tao et al, 2019), and is deployed on the rooftop of Anhui Air Traffic Management Bureau, Civil Aviation Administration of China. The appearance and functional specifications of ASC100 are shown in Fig.…”

Section: Instrument and Data Descriptionmentioning

confidence: 99%

SegCloud: a novel cloud image segmentation model using deep Convolutional Neural Network for ground-based all-sky-view camera observation

Xie¹,

Liu²,

Yang³

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

Abstract. Cloud detection and cloud properties have significant applications in weather forecast, signal attenuation analysis, and other cloud-related fields. Cloud image segmentation is the fundamental and important step to derive cloud cover. However, traditional segmentation methods rely on low-level visual features of clouds, and often fail to achieve satisfactory performance. Deep Convolutional Neural Networks (CNNs) are able to extract high-level feature information of object and have become the dominant methods in many image segmentation fields. Inspired by that, a novel deep CNN model named SegCloud is proposed and applied to accurate cloud segmentation based on ground-based observation. Architecturally, SegCloud possesses symmetric encoder-decoder structure. The encoder network combines low-level cloud features to form high-level cloud feature maps with low resolution, and the decoder network restores the obtained high-level cloud feature maps to the same resolution of input images. The softmax classifier finally achieves pixel-wise classification and outputs segmentation results. SegCloud has powerful cloud discrimination ability and can automatically segment the whole sky images obtained by a ground-based all-sky-view camera. Furthermore, a new database, which includes 400 whole sky images and manual-marked labels, is built to train and test the SegCloud model. The performance of SegCloud is validated by extensive experiments, which show that SegCloud is effective and accurate for ground-based cloud segmentation and achieves better results than traditional methods. Moreover, the accuracy and practicability of SegCloud is further proved by applying it to cloud cover estimation.

show abstract

“…The study of aerosol optical properties is another important field that uses sky cameras [10]. Some authors have created their own sky cameras for cloud cover assessment without using conventional solar occulting devices [11]. Furthermore, to counteract the temporal resolution limitations of satellite images, a digital camera was used for night-time cloud detection, detailing the percentage of cloud cover at 5-min intervals over the Manila Observatory [12].…”

Section: Introductionmentioning

confidence: 99%

Real-Time Automatic Cloud Detection Using a Low-Cost Sky Camera

Alonso-Montesinos

2020

Remote Sensing

View full text Add to dashboard Cite

Characterizing the atmosphere is one of the most complex studies one can undertake due to the non-linearity and phenomenological variability. Clouds are also among the most variable atmospheric constituents, changing their size and shape over a short period of time. There are several sectors in which the study of cloudiness is of vital importance. In the renewable field, the increasing development of solar technology and the emerging trend for constructing and operating solar plants across the earth’s surface requires very precise control systems that provide optimal energy production management. Similarly, airports are hubs where cloud coverage is required to provide high-precision periodic observations that inform airport operators about the state of the atmosphere. This work presents an autonomous cloud detection system, in real time, based on the digital image processing of a low-cost sky camera. An algorithm was developed to identify the clouds in the whole image using the relationships established between the channels of the RGB and Hue, Saturation, Value (HSV) color spaces. The system’s overall success rate is approximately 94% for all types of sky conditions; this is a novel development which makes it possible to identify clouds from a ground perspective without the use of radiometric parameters.

show abstract

Development of an all-sky imaging system for cloud cover assessment

Cited by 25 publications

References 14 publications

On the Generalization Ability of Data-Driven Models in the Problem of Total Cloud Cover Retrieval

On the Generalization Ability of Data-Driven Models in the Problem of Total Cloud Cover Retrieval

SegCloud: a novel cloud image segmentation model using deep Convolutional Neural Network for ground-based all-sky-view camera observation

Real-Time Automatic Cloud Detection Using a Low-Cost Sky Camera

Contact Info

Product

Resources

About