An empirical investigation into the effects of JPEG compression on the accuracy of photogrammetric point determination (PPD) is described. A pair of black-and-white aerial photographs of a city, taken at a scale of 1:8000, was selected and scanned at a resolution of 25 μm. Eighteen image points were measured with the ISDM module of an Intergraph digital photogrammetric workstation (DPW), and the bundle adjustment of a single model was performed using WuCAPSSGPS (Wuhan GPSsupported bundle block adjustment software). In processing various JPEG compressed images with Q-factors from 1 to 100, the accuracy of the 3D coordinates of the pass points was assessed and compared with that obtained from the original images (i.e., without compression). The empirical results show that, when the compression ratios are under 10, the compressed image is near-lossless. In other words, the visual quality of JPEG compressed images is still excellent and the accuracy of manual image mensuration is essentially not influenced. However, no indication can be found from the results that a compression of 10 is the critical value or the optimum compression level for PPD. Indeed, it is clear that the degradation of accuracy in PPD is almost linear.
CITATION STYLE
Li, Z., Yuan, X., & Lam, K. W. K. (2002). Effects of JPEG compression on the accuracy of photogrammetric point determination. Photogrammetric Engineering and Remote Sensing, 68(8), 847–853. https://doi.org/10.11834/jrs.20010307
Mendeley helps you to discover research relevant for your work.