Evaluación del marmoleado en carne bovina (m. longissimus dorsi) mediante imágenes hiperespectrales

• Autor: Velásquez Castillo, Lía Ethel
• Formato: tesis de grado
• Fecha de Publicación: 2015
• Institución: Universidad Nacional de Trujillo
• Repositorio: UNITRU-Tesis
• Lenguaje: español
• OAI Identifier: oai:dspace.unitru.edu.pe:20.500.14414/4415
• Enlace del recurso: https://hdl.handle.net/20.500.14414/4415
• Nivel de acceso: acceso abierto
• Materia: Binarización, Parámetros de distribución espacial, Imagen hiperespectral, Marmoleado, Árbol de decisiones

The aim of this study was to develop a system for evaluating the marbling of beef using hyperspectral imaging technology. The Japanese standard classification of the degree of marbling of beef was used as reference and twelve standards were digitized to obtain the parameters of shape and spatial distribution of marbling of each class. A total of 35 samples M. longissmus dorsi muscle were scanned by the hyperspectral imaging system of 400-1000 nm in reflectance mode. The wavelength of 528 nm was selected for segmentation of the sample and background. Processing algorithms of the image based on decision tree were trained and implemented for the binarization in the region of interest. The decision tree classified successfully 99.92% of the spectral data during training system. The classes were determined for all samples and these results were compared with the results obtained in a traditional evaluation with 23 judges semi-trained by the nonparametric Mann-Whitney test, it determined that there is no difference between the two methods of evaluation (p=0.324>0.05). Furthermore, the characterization of samples was performed to obtain average values of pH = 5.74 ± 0.13, an L * = 30.65 ± 1.92 with a * = 11.55 b * = ± 1.55 5.62 ± 1.12 for color, 0049 ± 00 009% lactic acid and 73.95 ± 1.54% moisture in the samples studied.