Improvement of photogrammetric JRC data distributions based on parabolic error models

Kim, Dong Hyun; Poropat, George; Gratchev, Ivan; Balasubramaniam, A. S.

doi:10.1016/j.ijrmms.2015.09.007

Cited by 22 publications

(13 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…11. In contrast to the authors' previous laboratory study (Kim et al 2015a), it appears that the plotted data are quite dispersed according to each field survey condition and mostly distributed in the overestimated region, as demonstrated in Fig. 11.…”

Section: Accuracy and Precision Of Photogrammetric Jrc Valuescontrasting

confidence: 86%

“…6a, the obtained data on point cloud intervals from the field surveys generally show a gradual increase with camera-to-object distance. The point intervals show rather scattered patterns (R 2 = 0.66-0.84) compared to those obtained under laboratory conditions obtained in a previous study by the authors (Kim et al 2015a). The dispersion of data is clearly shown from the results obtained using the short focal length (24 mm).…”

Section: Profile Agreementsupporting

confidence: 37%

“…As a natural measure of average error magnitude, the advantages of MAE have also been reported (Willmott and Matsuura 2005;Chai and Draxler 2014). A parabolic error model, which describes the RMSE of photogrammetric JRC values according to camera-to-object distances, was developed in a previous laboratory study (Kim et al 2015a). Using the quadratic error functions, the ranges of JRC errors obtained from the employed focal lengths and camera-to-object distances can be illustrated by the sizes of the parabolas.…”

Section: Evaluation Of the Accuracy Of Photogrammetric Jrc Valuesmentioning

confidence: 99%

“…The photographs of a rock sample were taken with the perpendicular camera orientation to the object and keeping the same indoor lighting conditions. The background of the developed models is detailed in the literature (Kim et al 2015a).…”

Section: Evaluation Of the Accuracy Of Photogrammetric Jrc Valuesmentioning

confidence: 99%

“…Based on the methodology of stereo photogrammetry, a JRC error model was developed (Kim et al 2015a). This model demonstrates the relationships between the RMSE of photogrammetric JRC values and the proportion of overor underestimation of the JRC values considering camerato-object distances.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Assessment of the Accuracy of Close Distance Photogrammetric JRC Data

et al. 2016

Self Cite

View full text Add to dashboard Cite

By using close range photogrammetry, this article investigates the accuracy of the photogrammetric estimation of rock joint roughness coefficients (JRC), a measure of the degree of roughness of rock joint surfaces. This methodology has proven to be convenient both in laboratory and in site conditions. However, the accuracy and precision of roughness profiles obtained from photogrammetric 3D images have not been properly established due to the variances caused by factors such as measurement errors and systematic errors in photogrammetry. In this study, the influences of camera-to-object distance, focal length and profile orientation on the accuracy of JRC values are investigated using several photogrammetry field surveys. Directional photogrammetric JRC data are compared with data derived from the measured profiles, so as to determine their accuracy. The extent of the accuracy of JRC values was examined based on the error models which were previously developed from laboratory tests and revised for better estimation in this study. The results show that high-resolution 3D images (point interval B1 mm) can reduce the JRC errors obtained from field photogrammetric surveys. Using the high-resolution images, the photogrammetric JRC values in the range of high oblique camera angles are highly consistent with the revised error models. Therefore, the analysis indicates that the revised error models facilitate the verification of the accuracy of photogrammetric JRC values.

show abstract

Section: Accuracy and Precision Of Photogrammetric Jrc Valuescontrasting

confidence: 86%

Section: Profile Agreementsupporting

confidence: 37%