Desarrollo de una aplicación móvil para determinar el grado de fermentación de los granos de cacao (Theobroma Cacao L.) aplicando técnicas de visión artificial basadas en deep learning.
The fermentation of cocoa beans (Theobroma cacao L.) is a critical process for the manufacture of chocolate, since fermentation influences the development of flavor, improving components such as free amino acids, peptides, and sugars. The degree of fermentation is determined by visual inspection of...
Bewaard in:
| Hoofdauteur: | |
|---|---|
| Formaat: | bachelorThesis |
| Taal: | spa |
| Gepubliceerd in: |
2022
|
| Onderwerpen: | |
| Online toegang: | http://repositorio.utc.edu.ec/handle/27000/8436 |
| Tags: |
Voeg label toe
Geen labels, Wees de eerste die dit record labelt!
|
Voeg label toe | Samenvatting: | The fermentation of cocoa beans (Theobroma cacao L.) is a critical process for the manufacture of chocolate, since fermentation influences the development of flavor, improving components such as free amino acids, peptides, and sugars. The degree of fermentation is determined by visual inspection of the changes in the internal color and texture of the grains, through the Cut Test (cut-test). The sensory profile is the way that human beings have to interpret the information that an object of the environment or the environment itself has through the nervous system, and therefore, produce an adaptive response to said environment. This visual grading system is the traditional method used today to assess the quality and acceptability of marketable cocoa. However, this approach is qualitative, tedious, and quite subjective, since it depends on the perception of the evaluator, it is very limited in the evaluation of defects and color of the cocoa beans. That is why this research aims to develop a mobile application that allows, quickly, easily, accurately, and at low cost to determine the degree of fermentation of cocoa beans, classifying them into various quality categories. Machine learning methodologies are defined as a set of competent techniques capable of automatically detecting patterns in data. Therefore, this degree work aims to develop a mobile application that classifies cocoa beans according to the degree of fermentation quality using artificial vision as a fast and accurate method. In this way, this application will provide a contribution to the community of farmers to detect the fermentation quality of their products without knowing in detail the characteristics that a cocoa bean presents according to its fermentation quality, saving time and money. |
|---|