Fast inline inspection by Neural Network Based Filtered Backprojection: Application to apple inspection. (November 2016)

• Record Type: Journal Article
• Authors: Janssens, Eline; Alves Pereira, Luis F.; De Beenhouwer, Jan; Tsang, Ing Ren; Van Dael, Mattias; Verboven, Pieter; Nicolaï, Bart; Sijbers, Jan
• Is Part Of: Case studies in nondestructive testing and evaluation. Volume 6:Part B(2016:Oct.)
• Page Start: 14
• Page End: 20
• Publication Date: 2016-11
• Subjects: Nondestructive testing -- Case studies -- Periodicals; Electronic journals; 620.112705
• DOI: 10.1016/j.csndt.2016.03.003
• Languages: English
• ISSNs: 2214-6571

Abstract

Abstract: Speed is an important parameter of an inspection system. Inline computed tomography systems exist but are generally expensive. Moreover, their throughput is limited by the speed of the reconstruction algorithm. In this work, we propose a Neural Network-based Hilbert transform Filtered Backprojection (NN-hFBP) method to reconstruct objects in an inline scanning environment in a fast and accurate way. Experiments based on apple X-ray scans show that the NN-hFBP method allows to reconstruct images with a substantially better tradeoff between image quality and reconstruction time.