| | Document Type | : | Latin Dissertation | Language of Document | : | English | Record Number | : | 53301 | Doc. No | : | TL23255 | Call number | : | MR50596 | Main Entry | : | Dipali Shridhar Narvankar | Title & Author | : | Assessment of soft X-rays for detection of fungal infection in stored wheatDipali Shridhar Narvankar | College | : | University of Manitoba (Canada) | Date | : | 2008 | Degree | : | M.Sc. | student score | : | 2008 | Page No | : | 92 | Abstract | : | Fungal infection is responsible for 5 to 10% of global food losses which can be reduced by early detection of fungal infection. Conventional methods currently being used for fungal detection are time consuming and tedious. Therefore, a fast, reliable, user friendly and easily upgradeable fungal detection method is necessary. In this study, the potential of a soft X-ray method for detection of fungal infection in stored wheat was explored. X-ray images of healthy wheat kernels and wheat kernels infected with Aspergillus niger , Aspergillus glaucus, and Penicillium spp. were acquired at 184 μA current and 13.6 kV voltage. A total of 34 features extracted from X-ray images were used to discriminate healthy and fungal-infected kernels. Statistical classifiers (linear, quadratic, and Mahalanobis) were applied to develop two-class, and four-class models. The maximum classification accuracy of 98.9% was obtained by the two-class model. The Mahalanobis discriminant classifier correctly identified on average 94.4% infected kernels. Four-class linear and quadratic classifiers could identify Penicillium with accuracy greater than 85%. Conversely, A. niger, A. glaucus, and healthy kernels were poorly classified by all statistical classifiers. | Subject | : | Applied sciences; Biological sciences; Aspergillus glaucus; Aspergillus niger; Penicillium; Agronomy; Agricultural engineering; 0359:Agronomy; 0539:Agricultural engineering | Added Entry | : | University of Manitoba (Canada) |
| | | | | |
| | |  |
http://lib.clisel.com/site/catalogue/53301
| | |
| | |  |
|