Tomography analysis tool: an application for image analysis based on unsupervised machine learning

Abstract We developed a graphical user interface (GUI) to analyse tomographic images of superconducting Nb 3 Sn wires designed for the next generation accelerator magnets. The Tomography Analysis Tool (TAT) relies on k -means algorithm, an unsupervised machine learning technique which is widely used partition into separated clusters. GUI compatible with both Linux and Windows operating systems. software reliability was tested by optical inspecting superimposed clustered image obtained k-means algorithm. TAT proven correctly segment various components single pixel precision. Finally, this can be useful tool scientific community quickly reproducibly images.