Aplicação de imagens digitais para a discriminação de mel de abelha.

• Autor(a) principal: Silva Júnior, Carlos Alberto da
• Data de Publicação: 2019
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UFPB
• Texto Completo: https://repositorio.ufpb.br/jspui/handle/123456789/19088
• topic: Abelha sem ferrão; Mel; Processamento digital de imagens; Jandaíra; Reconhecimento de padrões; Stingless bee; Digital image processing; Honey; Pattern recognition; CNPQ::CIENCIAS EXATAS E DA TERRA::QUIMICA

The honeybee is the third most adulterated food product in the world. In Brazil, this abundant and diversified spice has often been frauded or adulterated. Therefore, it is necessary the analysis for quality determination of this product. In this work, 100 honey samples of different bee species (Apis mellifera ligustica = 40, Melipona scutellaris = 40 and Melipona subnitida = 20) were analyzed from the municipalities of Alagoa Nova, Cabedelo, Itatuba and João Pessoa in the state of Paraíba (Brazil), as well as of the municipalities Passa e Fica and Macaíba in the state of Rio Grande do Norte (Brazil). The parameters evaluated were pH, moisture content, total soluble solids (BRIX) and reducing sugars (A.R). For the Apis, national and international legislations were regulated, respectively, by the Ministry of Agriculture from Brazil (MAPA) and the International Honey Commission (IHC). On the other hand, we opted for the legislation of the Malaysian Institute of Agricultural Research and Development for the Melipona. For honey of the species A. mellifera ligustica the average results were 18.1 ± 0.4% for moisture, 80.5 ± 0.3% for BRIX, 68.4 ± 2.0% for R.S and pH equal to 4.10 ± 0.07. On the other hand, the samples of M. scutellaris and M. subnitida presented, respectively, average values of 24.4 ± 1.9% and 26.3 ± 1.6% for moisture content, 74.2 ± 1,9% and 72.6 ± 1.7% for BRIX, 68.6 ± 0.9% and 67.7 ± 1.7% for R.S and 4.13 ± 0.27 and 3.80 ± 0.49 to pH. In accordance with the principles of Green Chemistry (GC), the Digital Image Processing (DIP) of these samples was also carried out, which allowed the discrimination of honey among the different species of bee, applying the methods of exploratory analysis (Principal Component Analysis - PCA). The results indicated the effectiveness of the methodology applied in DIP and pattern recognition and that, in general, the samples obey the limits established for the studied parameters.