An image stitching algorithm for the mineralogical analysis

• Published in: Minerals engineering Vol. 169; p. 106968
• Main Authors: Ro, Sung-Hyok, Kim, Se-Hun
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.08.2021
• Subjects: Image stitching; Microscopic mineralogical analysis; Mineral processing; Microscopic mineralogical analysis; Image stitching; Mineral processing
• ISSN: 0892-6875; 1872-9444
• DOI: 10.1016/j.mineng.2021.106968

•In the mineralogical analysis, stitching of the handheld microscope images with a small field of view has been described.•When detecting feature points for image matching, the size of the search region was fixed to improve stitching speed.•Stitching was performed using the ferruginous quartzite images and compared with three type of panorama software. Mineralogical analysis performed by automatic techniques, it includes qualitative and quantitative data about composition, grain size and mineral liberation, is a very important process while being the first step in the mineral processing and is one of the most useful parameters for metallurgists. Currently, there are many mineralogical analysis methods, but the problem is that analysis equipment is expensive and in-situ analysis is not possible. Therefore, in this paper, we have proposed an in-situ mineralogical analysis method using a relatively inexpensive handheld microscope. At this time, image stitching was applied to solve the inconvenience of observation due to the small field of view of the microscope. Experiments using a ferruginous quartzite ore confirmed that this method is very useful for immediate mineralogical analysis and detection of interesting areas for more detailed analysis. The algorithm suggested in this paper can be applied to the development of a dedicated mineralogical analysis system in future.