An image analysis environment for species identification of food contaminating beetles

作者: Leihong Wu , Weida Tong , Zhichao Liu , Su Inn Park , Hongjian Ding

DOI:

关键词:

摘要: Food safety is vital to the well-being of society; therefore, it important inspect food products ensure minimal health risks are present. The presence certain species insects, especially storage beetles, a reliable indicator possible contamination during and processing. However, current approach identifying by visual examination insect fragments rather subjective time-consuming. To aid this inspection process, we have developed in collaboration with FDA analysts some image analysis-based machine intelligence achieve identification up 90% accuracy. project continuation development effort. Here present an analysis environment that allows practical deployment on computers limited processing power memory. Using environment, users can prepare input sets selecting images for analysis, these through integrated panning zooming capabilities. After results panel user compare analyzed reference proposed species. Further additions should include log previously images, eventually extend interaction central cloud repository web-based interface.

参考文章(4)
Alan R. Olsen, John S. Gecan, George C. Ziobro, John R. Bryce, Regulatory action criteria for filth and other extraneous materials v. strategy for evaluating hazardous and nonhazardous filth. Regulatory Toxicology and Pharmacology. ,vol. 33, pp. 363- 392 ,(2001) , 10.1006/RTPH.2001.1472
Su Inn Park, Halil Bisgin, Hongjian Ding, Howard G. Semey, Darryl A. Langley, Weida Tong, Joshua Xu, Species Identification of Food Contaminating Beetles by Recognizing Patterns in Microscopic Images of Elytra Fragments PLOS ONE. ,vol. 11, pp. e0157940- ,(2016) , 10.1371/JOURNAL.PONE.0157940
Ameet Talwalkar, Mehryar Mohri, Afshin Rostamizadeh, The Foundations of Machine Learning ,(2012)