Low-density foreign body detection in food products using single-shot grid-based dark-field X-ray imaging

• Published in: Journal of food engineering 2022-12, Vol.335, p.111189, Article 111189
• Main Authors: Lim, Hyunwoo, Lee, Jonghyeok, Lee, Soohyun, Cho, Hyosung, Lee, Hunwoo, Jeon, Duhee
• Format: Article
• Language: English
• Subjects: Food inspection; Small-angle X-ray scattering; Spatial harmonic imaging; X-ray grid
• ISSN: 0260-8774; 1873-5770
• DOI: 10.1016/j.jfoodeng.2022.111189

This study presents a novel approach for detecting low-density foreign bodies (FBs) in food products using single-shot grid-based dark-field X-ray imaging (SG-DFXI). SG-DFXI obtains information related to the sample's small-angle X-ray scattering using a conventional X-ray source and grid. It is specifically utilized to detect low-density materials that are invisible when using conventional techniques. To demonstrate its efficacy, an experiment was conducted with food samples containing low-density FBs such as wood, Styrofoam, pencil lead, and soft plastic. A quantitative evaluation was conducted based on the contrast-to-noise ratio (CNR) and relative contrast gain (RCG). The proposed method detected low-density FBs barely visible in absorption images. Additionally, the CNR values measured in dark-field images were approximately 2.8–10.5 times larger than those in the absorption images, indicating that the proposed approach significantly improved the ability to detect low-density FBs in food samples. The RCG characteristic was improved as the system's autocorrelation length (ξG) increased; the RCG value for ξG = 453 nm was approximately 2.5 times larger than that for ξG = 171 nm. •Method detecting low-density foreign bodies (FBs) in food products is proposed.•Single-shot grid-based dark-field X-ray imaging (SG-DFXI) method is used.•SG-DFXI obtains information related to samples' small-angle X-ray scattering.•The experiment used food samples with FBs like wood, Styrofoam, soft plastic, etc.•CNR and relative contrast gain (RCG) were used for quantitative evaluation.