Image Quality Assessment for Wood Images
Heshalini Rajagopal, Norrima Binti Mokhtar, Anis Salwa Mohd Khairuddin · 2022 IEEE International Conference on Artificial Intelligence in Engineering and Technology (IICAIET) · 2022
This work proposed the implementation of subjective and objective assessment on wood images to analyse the quality of wood images for wood species recognition purposes. Several distorted images are generated from the reference images by applying Gaussian White Noise (GWN) and Motion Blur (MB) at various levels of distortions for comparison purposes. Ten subjects from Negeri Sembilan Forestry Department were selected to assess the distorted images for the subjective evaluation. In the objective evaluation, five Full Reference-IQAs (FR-IQAs) were used to evaluate the distorted images. The subjective scores were used as the benchmark to determine the most suitable objective FR-IQA to assess wood images. The relationship between the subjective scores and objective FR-IQAs are examined using performance metrics, namely PLCC and RMSE. It was found that FSIM is the most suitable FR-IQA to assess wood images distorted with GWN and MB.