Does Linus' Law Apply to Data Quality in Volunteered Street View Imagery (VSVI)? Mechanisms Contributing to VSVI Data Quality Improvement for Environmental Audits

ABSTRACT Street View Imagery (SVI) is crucial for urban environmental audits. Although Volunteered Street View Imagery (VSVI) has the potential to provide higher spatial and temporal resolution than traditional SVI, the use of such user‐generated data is currently limited by the inherent spatial heterogeneity in contributions and unclear quality control. To explore the underlying mechanisms, this study focused on the validity of Linus’ law in VSVI quality improvement in terms of spatial and temporal qualities related to environmental audits, using Mapillary as an example. It is assumed that as the number of street revisits for a road increases, so does the level of data quality. Results from regression and correlation analyses show that Linus’ law applies to VSVI quality improvement with different relationships depending on quality elements and road types. Furthermore, this study determines the number of revisits required to exceed the traditional SVI or achieve a higher quality level.