Abstract
As machine learning becomes more popular in the precision engineering sector, the need for large datasets of measurement data increases. Due to the often manual, user dependent and labour-intensive measurement processes, collecting a large amount of data is often infeasible. It would, therefore, be desirable to collect a small amount of data on which to train a model to generate synthetic data that is representative of the real measurement data. To this end, we present an approach to numerical surface texture generation based on a progressively growing generative adversarial network. We show that by encoding height data into grayscale values within an image, the network can create realistic synthetic surface data both qualitatively and quantitatively. The proposed approach is general to any encoded surface; we demonstrate the model trained on two example datasets consisting of surfaces from different manufacturing processes and measured with different techniques. We finally present an extension to the generator model which automatically categorises the produced surfaces, allowing a surface of a desired category to be generated. Finally, we calculate the distributions of areal surface texture parameters for each type of surface and show that there is good agreement between the synthetic and real data.
Highlights
The ability to generate synthetic surface texture data which convincingly represents the result of a real measurement has many ap plications [1,2,3]
In this paper we present a new method to produce surface texture data based on an approach initially developed for generating high res olution synthetic images: a progressively growing adversarial network (PGGAN) [10]
Many of the features of the PGGAN are designed to increase variation in the output but, at least for our application, there is still some work to do in this area
Summary
The ability to generate synthetic surface texture data which convincingly represents the result of a real measurement has many ap plications [1,2,3]. The synthetic texture was simulated by analysing the real surface data of a part made with the same AM process, extracting the dominant spatial frequencies and amplitudes, and layer ing various pseudo-random noise functions at these frequencies and amplitudes. While this approach produces a good estimation of the surface parameters, it does not capture properties related to the surface features, such as feature shape and surface anisotropy. In this paper we present a new method to produce surface texture data based on an approach initially developed for generating high res olution synthetic images: a progressively growing adversarial network (PGGAN) [10]. We perform a quantitative comparison of the cat egorised generated surfaces with their real counterparts and show the model provides a sound representation of the surface types
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
