A fast image dehazing algorithm based on negative correction

Gao, Yuanyuan; Hu, Hai‐Miao; Wang, Shuhang; Li, Bo

doi:10.1016/j.sigpro.2014.02.016

Cited by 63 publications

(27 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The RS data were downloaded from the United States Geological Survey. The images were rectified by polynomial geometric precision correction and mosaicked and trimmed to cover the study area [37]. The data included Landsat Thematic Mapper (Landsat TM) visible red band 3 (B3, 0.63-0.69 µm), near infrared band 4 (B4, 0.78-0.90 µm) and short-wave infrared band 5 (B5, 1.55-1.75 µm).…”

Section: Predictor Variablesmentioning

confidence: 99%

Spatial-Temporal Changes of Soil Organic Carbon Content in Wafangdian, China

Wang

Adhikari

et al. 2016

Sustainability

View full text Add to dashboard Cite

Soil organic carbon (SOC) plays an important role in soil fertility and the global carbon cycle. A better understanding of spatial-temporal changes of SOC content is essential for soil resource management, emission studies, and carbon accounting. In this study, we used a boosted regression trees (BRT) model to map distributions of SOC content in the topsoil (0-20 cm) and evaluated its temporal dynamics from 1990-2010 in Wafangdian City, northeast of China. A set of 110 (1990) and 127 (2010) soil samples were collected and nine environment variables (including topography and vegetation) were used. A 10-fold cross-validation was used to evaluate model performance as well as predictive uncertainty. Accuracy assessments showed that R 2 of 0.53 and RMSE (Root-mean-square error) of 9.7 g·kg −1 for 1990, and 0.55, and 5.2 g·kg −1 for 2010. Elevation and NDVI (Normalized Difference Vegetation Index) were the two important variables affecting SOC distribution. Results showed that mean SOC content decreased from 19 ± 14 to 18 ± 8 g·kg −1 over a 20 year period. The maps of SOC represented a decreasing trend from south to north across the study area in both periods. Rapid urbanization and land-use changes were accountable for declining SOC levels. We believe predicted maps of SOC can help local land managers and government agencies to evaluate soil quality and assess carbon sequestration potential and carbon credits.

show abstract

Section: Predictor Variablesmentioning

confidence: 99%

Spatial-Temporal Changes of Soil Organic Carbon Content in Wafangdian, China

Wang

Adhikari

et al. 2016

Sustainability

View full text Add to dashboard Cite

show abstract

“…As we can see, our algorithm is much faster than those of Choi [7], Tarel [13], He [14], Meng [25], Cai [28] and slightly faster than He [23], Yu [24], Gao [47], Zhu [26], Ren [30], and it has high efficiency even when the resolution of the nput image is large. This advantage of efficiency mainly benefits from the proposed linear model …”

Section: Quantitative Comparisonmentioning

confidence: 90%

“…13(h), the Gao's [47] method can enhance details of hazy images and can achieve fine effect generally. However, for the desk image, the floor looks too dark and haze remains in the upper part.…”

Section: Qualitative Comparison On Synthetic Imagesmentioning

confidence: 99%

“…Fig. 12(b-k) show the results of the methods of Choi [7], Tarel [13], He [14], He [23], Yu [24], Meng [25], Gao [47], Zhu [26], Cai [28], Ren [30], respectively. The results of the proposed algorithm are given in Fig.…”

Section: Qualitative Comparison On Real-world Imagesmentioning

confidence: 99%

“…In Fig. 12(h), we can notice that the image recovered using Gao's method [47] generally looks dim, since the haze removal magnitude which is determined via a negative correction strategy would lead to the over-decreasing of the true scene brightness (see the third image). As we discussed in Section 3.1, the fixed scattering coefficient does not always conform to the real world, and for this reason, the dehazing results of Zhu [26] are not quite satisfactory for dense hazy image (see the zoom-in of Fig.…”

Section: Accepted Manuscriptmentioning

confidence: 99%

See 2 more Smart Citations