Home » KB » Innovations Blog » Research publication: The Effect of Camera Calibration on Multichannel Texture Classification

The Effect of Camera Calibration on Multichannel Texture Classification

March 22. 2021

Research publication in Journal of Imaging Science and Technology from Society for Imaging Science and Technology from our employee Michele Conni together with Peter Nussbaum and Phil Green from Norwegian University of Science and Technology :

The Effect of Camera Calibration on Multichannel Texture Classification

 The efficiency of a texture classification procedure depends on the color space in which it is performed. Classification in a perceptually meaningful space requires chromatic coordinates obtained from a calibrated acquisition setup. The authors assess the impact of camera calibration, within a generic color picture acquisition workflow, on the performance of a number of texture classification techniques. An image calibration pipeline is established and applied to a texture database, and the accuracy of the classification algorithms is evaluated for each step.

 

The results show that the most significant step of the workflow is color rendering although the effect is relatively small. Hence precise scene-referred characterization of the raw data from an acquisition camera is not essential for most texture classification tasks. In addition, working with output-referred RGB data is likely to be adequate for the majority of classification tasks.

 

https://www.ingentaconnect.com/content/ist/jist/2021/00000065/00000001/art00005