The quality control of steel products by human vision remains tedious, fatiguing, somewhat fast, rather robust, sketchy, dangerous or impossible. For these reasons, the use of the artificial vision in the world of quality control has become more than necessary. However, these images are often large in terms of quantity and size, which becomes a problem in quality control centers, where engineers are unable to store these images. For this, efficient compression techniques are necessary for archiving and transmitting the images. The reduction in file size allows more images to be stored in a disk or memory space. The present paper proposes an effective technique for redundancy extraction using the Principal Component Analysis (PCA) approach. Furthermore, it aims to study the effects of the number of eigenvectors employed in the PCA compression technique on the quality of the compressed image. The results revealed that using only 25% of the eigenvectors provide very similar compressed images compared to the original ones, in terms of quality. These images are characterized by high compression ratios and a small storage space.
Seismic Data Quality Control and Interpolation Using Principal Component Analysis
