Skip to main content
ARS Home » Research » Publications at this Location » Publication #231698
Title: Rapid identification of rice samples using an electronic nose

Author
item ZHENG, XIANZHE - Northeast Agricultural University
item Lan, Yubin
item ZHU, JIANMIN - Fort Valley State University
item Westbrook, John
item Hoffmann, Wesley
item LACEY, RON - TEXAS A&M UNIV

Submitted to: Journal of Bionics
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/1/2009
Publication Date: 9/30/2009
Citation: Zheng, X., Lan, Y., Zhu, J., Westbrook, J.K., Hoffmann, W.C., Lacey, R. 2009. Rapid identification of rice samples using an electronic nose. Journal of Bionics. 6:290-497.

Interpretive Summary: Rice provides a large proportion of the total nourishment of the world’s population, but the nutritional and economic value differs among varieties of rice. Because humans are subject to fatigue and inconsistencies in human sensory panels of food testing, there is a need for an instrument that can differentiate rice varieties. This study was conducted to apply the electronic nose technology (computerized sensors that mimic the human sense of smell by detecting airborne chemical compounds) to differentiate rice varieties. Results from this laboratory research found that an electronic nose successfully differentiated varieties of rice. This method could be used by the rice industry to rapidly identify rice samples which have different market values.

Technical Abstract: Four rice samples of long grain type were tested using an electronic nose (Cyranose-320). Samples of 5 g of each variety of rice were placed individually in vials and were analyzed with the electronic nose unit consisting of 32 polymer sensors. The Cyranose-320 was able to differentiate between varieties of rice. The chemical composition of the rice odors for differentiating rice samples need to be investigated. The optimum parameter settings should be considered during the Cyranose-320 training process especially for multiple samples, which are helpful for obtaining an accurate training model to improve identification capability. Further, it is necessary to investigate the E-nose sensor selection for obtaining better classification accuracy. A reduced number of sensors could potentially shorten the data processing time, and could be used to establish an application procedure and reduce the cost for a specific electronic nose. Further research is needed for developing analytical procedures that adapt the Cyranose-320 as a tool for testing rice quality.