Camera-projector calibration for near infrared imaging system

Junita Mohamad-Saleh,Marlina Yakno,W. N. A. W. Samsudin,Mohd Zamri Ibrahim

doi:10.11591/eei.v9i1.1697

Junita Mohamad-Saleh, Marlina Yakno + Show 2 more

Open Access

https://doi.org/10.11591/eei.v9i1.1697

Copy DOI

Abstract

Advanced biomedical engineering technologies are continuously changing the medical practices to improve medical care for patients. Needle insertion navigation during intravenous catheterization process via Near infrared (NIR) and camera-projector is one solution. However, the central point of the problem is the image captured by camera misaligns with the image projected back on the object of interest. This causes the projected image not to be overlaid perfectly in the real-world. In this paper, a camera-projector calibration method is presented. Polynomial algorithm was used to remove the barrel distortion in captured images. Scaling and translation transformations are used to correct the geometric distortions introduced in the image acquisition process. Discrepancies in the captured and projected images are assessed. The accuracy of the image and the projected image is 90.643%. This indicates the feasibility of the captured approach to eliminate discrepancies in the projection and navigation images.

Highlights

Intravenous injection is difficult and painful especially for newborns, children, and obese people because appropriate veins maybe deep and not visible
It can be seen that the projected chessboard pattern is the one which looks like a shadow and the black is the real chessboard pattern. This is due to the central point of the image being captured by the camera is different than that being projected back from the projector
After I=-0.12, all corrected lines do not go through the center of the image are bowed inwards, towards the center of the image like pincushion (refer to Figure 4 (e) and (f))

Summary

INTRODUCTION

Intravenous injection is difficult and painful especially for newborns, children, and obese people because appropriate veins maybe deep and not visible. Some researchers and companies have developed devices which can show an enhanced image on the screen to help locate vein [5,6,7,8,9,10]. To be different from others, [16] proposed a combination of NIR imaging technique and augmented reality (AR) technology in head mount device (HMD). This HMD system created a virtual image and overlay onto real-world object to form a composite view in the HMD see-through lenses.

SYSTEM STRUCTURE

RESULTS AND DISCUSSION

CONCLUSION