Introduction to Vehicle Documentation Using a Mobile Phone

Abstract

Introduction

Many current generations of mobile phones are equipped with high-resolution cameras and sensors that can detect the distance to objects. The computing power of these devices then allows the data thus obtained to be processed to create 3D models, with documentation possible even on larger objects such as vehicles. Many mobile applications take advantage of these capabilities and are presented as suitable for forensic documentation, for example, in the documentation of vehicle damage or accident sites. However, there is still a question about how accurately these apps can be used and what are their limitations.

Materials and methods

To assess the accuracy of the documentation with a mobile phone, the vehicle was scanned using a Leica RTC360 3D laser scanner. This vehicle scan served as a reference against which the documentation from the mobile device was compared. The mobile application Recon3D, which is one of the applications presented for forensic documentation and whose output is a point cloud, was chosen for comparison. The CoudCompare program was used to compare the measurements, in which the distance between the two selected points was performed for both documentation methods, as well as the comparison of the point cloud obtained by the mobile device (Apple iPhone 15 Pro) with the point cloud obtained by the laser scanner.

Results

The comparison of both measurement methods shows that the measurement using mobile phones or mobile applications achieves deviations of up to about 2 cm from the relatively accurate documentation of the vehicle with a 3D laser scanner (with a measurement accuracy of a few mm). It cannot be ruled out that part of the inaccuracy was due to user error in data acquisition and that the overall accuracy could be increased by using slicing targets that can be used for scale calibration.

Conclusions

Based on the testing performed on the Recon3D mobile application for 3D vehicle documentation, a relatively good measurement accuracy was achieved, which would be sufficient in the context of accident analysis when determining the deformation energy based on the extent of vehicle deformation or when documenting the final position of the vehicle. However, a number of further tests will be necessary in the research to investigate more applications (including mobile devices) and also to determine their measurement accuracy, limitations and the optimal documentation procedure to ensure that maximum data can be obtained with minimum error.