A cloud image detection method based on SVM vector machine

Li, Pengfei; Dong, Limin; Xiao, Huaibao; Xu, Mingliang

doi:10.1016/j.neucom.2014.09.102

Cited by 131 publications

(66 citation statements)

References 28 publications

(14 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For instance, the edge of the cloud contains gray level jump characteristics, and the part of the cloud is similarity to the whole. The cloud cluster has a certain fractals similarity [30]. On the other hand, the texture of snow in remote sensing images usually is bumpy because the influence from terrain relief, vegetation or man-made features.…”

Section: The Usefulness Of Spatial Contextual Information For Cloud Amentioning

confidence: 99%

Cloud and Snow Discrimination for CCD Images of HJ-1A/B Constellation Based on Spectral Signature and Spatio-Temporal Context

Bian

Qiannan

et al. 2016

Remote Sensing

View full text Add to dashboard Cite

It is highly desirable to accurately detect the clouds in satellite images before any kind of applications. However, clouds and snow discrimination in remote sensing images is a challenging task because of their similar spectral signature. The shortwave infrared (SWIR, e.g., Landsat TM 1.55-1.75 µm band) band is widely used for the separation of cloud and snow. However, for some sensors such as the CBERS-2 (China-Brazil Earth Resources Satellite), CBERS-4 and HJ-1A/B (HuanJing (HJ), which means environment in Chinese) that are designed without SWIR band, such methods are no longer practical. In this paper, a new practical method was proposed to discriminate clouds from snow through combining the spectral reflectance with the spatio-temporal contextual information. Taking the Mt. Gongga region, where there is frequent clouds and snow cover, in China as a case area, the detailed methodology was introduced on how to use the 181 scenes of HJ-1A/B CCD images in the year 2011 to discriminate clouds and snow in these images. Visual inspection revealed that clouds and snow pixels can be accurately separated by the proposed method. The pixel-level quantitative accuracy validation was conducted by comparing the detection results with the reference cloud masks generated by a random-tile validation scheme. The pixel-level validation results showed that the coefficient of determination (R 2 ) between the reference cloud masks and the detection results was 0.95, and the average overall accuracy, precision and recall for clouds were 91.32%, 85.33% and 81.82%, respectively. The experimental results confirmed that the proposed method was effective at providing reasonable cloud mask for the SWIR-lacking HJ-1A/B CCD images. Since HJ-1A/B have been in orbit for over seven years and these satellites still run well, the proposed method is helpful for the cloud mask generation of the historical archive HJ-1A/B images and even similar sensors.

show abstract

Section: The Usefulness Of Spatial Contextual Information For Cloud Amentioning

confidence: 99%

Cloud and Snow Discrimination for CCD Images of HJ-1A/B Constellation Based on Spectral Signature and Spatio-Temporal Context

Bian

Qiannan

et al. 2016

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…In [36], cloudy image was divided into sub-blocks of the same size according to the spatial position, then, the brightness and textural features were calculated to predict the class for each sub-block. However, the rigid division cannot be adaptive for the varied shape and size of clouds.…”

Section: Superpixel Segmentationmentioning

confidence: 99%

“…Clouds on the remote sensing images are mainly uniform flat, and a large proportion of them are brighter than most of the Earth's surface [9,36]. We define a brightness feature to enhance the difference between cloud pixels and non-cloud pixels:…”

Section: Superpixel Coarse Classificationmentioning

confidence: 99%

“…Hu et al [35] adopted a visual attention technique in computer vision based on random forest to automatically identify images with a significant cloud cover. Li et al [36] exploited the brightness and texture features to detect clouds based on SVM. An and Shi [37] proposed an automatic supervised remote-sensing image cloud detection approach based on the scene-learning scheme with two modules: feature data simulating and cloud detector learning and applying.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Cloud Detection Method for Landsat 8 Images Based on PCANet

Xie

Jiang

2018

Remote Sensing

View full text Add to dashboard Cite

Cloud detection for remote sensing images is often a necessary process, because cloud is widespread in optical remote sensing images and causes a lot of difficulty to many remote sensing activities, such as land cover monitoring, environmental monitoring and target recognizing. In this paper, a novel cloud detection method is proposed for multispectral remote sensing images from Landsat 8. Firstly, the color composite image of Bands 6, 3 and 2 is divided into superpixel sub-regions through Simple Linear Iterative Cluster (SLIC) method. Then, a two-step superpixel classification strategy is used to predict each superpixel as cloud or non-cloud. Thirdly, a fully connected Conditional Random Field (CRF) model is used to refine the cloud detection result, and accurate cloud borders are obtained. In the two-step superpixel classification strategy, the bright and thick cloud superpixels, as well as the obvious non-cloud superpixels, are firstly separated from potential cloud superpixels through a threshold function, which greatly speeds up the detection. The designed double-branch PCA Network (PCANet) architecture can extract the high-level information of cloud, then combined with a Support Vector Machine (SVM) classifier, the potential superpixels are correctly classified. Visual and quantitative comparison experiments are conducted on the Landsat 8 Cloud Cover Assessment (L8 CCA) dataset; the results indicate that our proposed method can accurately detect clouds under different conditions, which is more effective and robust than the compared state-of-the-art methods.

show abstract

“…However, because of different objects with similar spectral profiles and image noises, spectral features are not enough for cloud detection. Therefore, for high-accuracy cloud detection, we adequately consider both the spectral and spatial information of remote sensing imagery in this paper, mainly including spectral, texture and structure features [37,38].…”

Section: Preprocessing: Feature Selectionmentioning

confidence: 99%

Fuzzy AutoEncode Based Cloud Detection for Remote Sensing Imagery

et al. 2017

View full text Add to dashboard Cite

Cloud detection of remote sensing imagery is quite challenging due to the influence of complicated underlying surfaces and the variety of cloud types. Currently, most of the methods mainly rely on prior knowledge to extract features artificially for cloud detection. However, these features may not be able to accurately represent the cloud characteristics under complex environment. In this paper, we adopt an innovative model named Fuzzy Autoencode Model (FAEM) to integrate the feature learning ability of stacked autoencode networks and the detection ability of fuzzy function for highly accurate cloud detection on remote sensing imagery. Our proposed method begins by selecting and fusing spectral, texture, and structure information. Thereafter, the proposed technique established a FAEM to learn the deep discriminative features from a great deal of selected information. Finally, the learned features are mapped to the corresponding cloud density map with a fuzzy function. To demonstrate the effectiveness of the proposed method, 172 Landsat ETM+ images and 25 GF-1 images with different spatial resolutions are used in this paper. For the convenience of accuracy assessment, ground truth data are manually outlined. Results show that the average RER (ratio of right rate and error rate) on Landsat images is greater than 29, while the average RER of Support Vector Machine (SVM) is 21.8 and Random Forest (RF) is 23. The results on GF-1 images exhibit similar performance as Landsat images with the average RER of 25.9, which is much higher than the results of SVM and RF. Compared to traditional methods, our technique has attained higher average cloud detection accuracy for either different spatial resolutions or various land surfaces.

show abstract

A cloud image detection method based on SVM vector machine

Cited by 131 publications

References 28 publications

Cloud and Snow Discrimination for CCD Images of HJ-1A/B Constellation Based on Spectral Signature and Spatio-Temporal Context

Cloud and Snow Discrimination for CCD Images of HJ-1A/B Constellation Based on Spectral Signature and Spatio-Temporal Context

A Cloud Detection Method for Landsat 8 Images Based on PCANet

Fuzzy AutoEncode Based Cloud Detection for Remote Sensing Imagery

Contact Info

Product

Resources

About