Sky detection by effective context inference

Shang, Yuanyuan; Li, Ge; Luan, Zhong; Zhou, Xiuzhuang; Guo, Guodong

doi:10.1016/j.neucom.2015.12.126

Cited by 9 publications

(13 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To demonstrate the effectiveness of our proposed approach, we conduct extensive experiments on our HazySky dataset and a subset of the SkyFinder dataset [15]. On the HazySky dataset, we compare our method against three recently proposed sky detection algorithms (Shen et al's [1], Lu et al's [6] and Shang et al's [14]). On the SkyFinder dataset, we compare our method against the same baseline methods (Lu et al's [6], Hoiem et al's [8] and Tighe et al's [9]) evaluated by Mihail et al [15].…”

Section: Methodsmentioning

confidence: 99%

“…Besides, in some dense hazy scenes, Lu et al's method [6] is completely inefficient (as shown in the third to fifth row of Figure 6). Shang et al's method [14] also performs badly when the texture features and line features are weak (as shown in the fourth column of Figures 5 and 6). All three algorithms [1,6,14] adapt poorly to hazy scenes, and our method produces the best results under different sky scenes of hazy images.…”

Section: Performance On the Hazysky Datasetmentioning

confidence: 98%

“…The second stage is to label each uncertain sample as sky or non-sky. In this stage, the S sky_hc and S nsky_hc obtained in the first stage are taken as the reference, and a similarity measurement is conducted by computing the Euclidean distance of feature vectors between the candidate sample to the reference sky and reference non-sky regions, respectively, as shown in Equations (14) and (15).…”

Section: Uncertain Regions Labelingmentioning

confidence: 99%

See 2 more Smart Citations

Earthquake prediction based on community division

Ren

Liu

et al. 2018

Physica A: Statistical Mechanics and its Applications

View full text Add to dashboard Cite

Section: Methodsmentioning

confidence: 99%

Section: Performance On the Hazysky Datasetmentioning

confidence: 98%

Section: Uncertain Regions Labelingmentioning

confidence: 99%

See 1 more Smart Citation

Earthquake prediction based on community division

Ren

Liu

et al. 2018

Physica A: Statistical Mechanics and its Applications

View full text Add to dashboard Cite

“…Another challenge that previous methods seek to solve is the partial occlusion of the sky by foreground objects cutting the sky into many disconnected parts. In this case, horizon line detection-based method [ 1 ] will lose efficacy, and classification-based approaches [ 6 , 8 , 9 , 14 ] may be more effective. Therefore, methods combining hand-engineered features with a classifier were used to address this problem.…”

Section: Introductionmentioning

confidence: 99%

Sky Detection in Hazy Image

Song

Luo

et al. 2018

Sensors

View full text Add to dashboard Cite

Sky detection plays an essential role in various computer vision applications. Most existing sky detection approaches, being trained on ideal dataset, may lose efficacy when facing unfavorable conditions like the effects of weather and lighting conditions. In this paper, a novel algorithm for sky detection in hazy images is proposed from the perspective of probing the density of haze. We address the problem by an image segmentation and a region-level classification. To characterize the sky of hazy scenes, we unprecedentedly introduce several haze-relevant features that reflect the perceptual hazy density and the scene depth. Based on these features, the sky is separated by two imbalance SVM classifiers and a similarity measurement. Moreover, a sky dataset (named HazySky) with 500 annotated hazy images is built for model training and performance evaluation. To evaluate the performance of our method, we conducted extensive experiments both on our HazySky dataset and the SkyFinder dataset. The results demonstrate that our method performs better on the detection accuracy than previous methods, not only under hazy scenes, but also under other weather conditions.

show abstract

“…Sky detection can improve accuracy of GNSS based localization [3,5,9], can be used for scene classification, and is often employed to achieve more effective contentbased image retrieval [6]. Most existing sky region detection algorithms are mainly based on color priors [1,2,11]. In the method proposed in [1], both image simplification and classification are employed.…”

Section: Introductionmentioning

confidence: 99%

An Efficient Sky Detection Algorithm From Fisheye Image Based on region classification and segment analysis

Bour

Merabet

Ruichek

et al. 2017

TMLAI

View full text Add to dashboard Cite

In this paper, an efficient approach for automatic and accurate sky region detection from fisheye images is proposed. The proposed approach starts by segmenting the acquired image into regions using Statistical Region Merging method. After that, the segmented regions are characterized using local RGB color descriptor using image quantization. The next step consists of classifying the characterized regions into sky and nonsky regions by using maximal similarity based region classification through Hellinger kernel based distance. In order to improve the obtained region classification results, a segment analysis based technique using Line Segment Detector is proposed. Experimental results prove the robustness and performance of the proposed procedure.

show abstract

Sky detection by effective context inference

Cited by 9 publications

References 17 publications

Earthquake prediction based on community division

Earthquake prediction based on community division

Sky Detection in Hazy Image

An Efficient Sky Detection Algorithm From Fisheye Image Based on region classification and segment analysis

Contact Info

Product

Resources

About