LOW-LOSS IMAGE COMPRESSION TECHNIQUES FOR ARTIFICIAL NEURAL NETWORK TRAINING: A STUDY IN CUTTING TOOL WEAR IMAGES
DOI:
https://doi.org/10.5585/exacta.v8i2.2000Keywords:
Cutting tool. Image compression. Lifting technique. Principal Components Analysis.Abstract
This work accomplishes a comparative study between two distinct image compression techniques, namely the Lifting technique and the Principal Components Analysis. Lifting and Principal Components Analysis were applied in original images of a cutting tool in order to produce a low resolution version, which will be used as input cases during the neural network training processes. The main objective of this study is to choose the most appropriate low-loss image compression technique for cutting tool images which are used for training an artificial neural network in the degeneration patterns recognition. As a consequence it will be possible to decide about the best moment to change the cutting tool.Downloads
Download data is not yet available.
Downloads
Published
2010-11-19
How to Cite
Pereira, F. H., Baptista, E. A., Coppini, N. L., Espírito-Santo, R. do, & Oliveira, A. J. de. (2010). LOW-LOSS IMAGE COMPRESSION TECHNIQUES FOR ARTIFICIAL NEURAL NETWORK TRAINING: A STUDY IN CUTTING TOOL WEAR IMAGES. Exacta, 8(2), 225–235. https://doi.org/10.5585/exacta.v8i2.2000
Issue
Section
Papers