Analyzing magnetic resonance images of Iberian pork loin to predict its sensorial characteristics
详细信息 查看全文
- 作者：Cernadas ; E. ; Carri??n ; P. ; Rodriguez ; P.G. ; Muriel ; E. ; Antequera ; T.
- 关键词：Meat product ; Iberian pork loin ; Texture analysis ; Image classification ; Feature selection ; Pattern recognition
- 刊名：Computer Vision and Image Understanding
- 出版年：2005
- 出版时间：May, 2005
- 年：2005
- 卷：98
- 期：2
- 页码：344-360
- 全文大小：597 K
文摘
Iberian pork comes from genuinely bred Southwest Iberian Peninsula pigs traditionally fattened with acorns and pasture in an extensive production system. Dry-cured loins and hams constitute the main uncooked pork products with high sensorial quality and a first rate consumer acceptance, leading to high prices in the market. Several aspects related to quality in Iberian products have been examined by using chemical and sensorial procedures to provide quality. However, all these approaches are tedious and destroy the item. In addition, food science has shown little interest in MRI to explore meat products in a non-invasive way. Therefore, this paper introduce an objective and non-destructive methodology to classify Iberian loins consistently. It is based on texture analysis of MRI images displaying dry-cured pork loins. A statistical evaluation is provided for a set of 47 loins to predict three levels of different sensorial characteristics.