Phenotype maps of coffee plantation

• Grant number: 16/15511-9
• Support Opportunities: Research Grants - Innovative Research in Small Business - PIPE
• Start date: February 01, 2018
• End date: October 31, 2018
• Field of knowledge: Engineering - Electrical Engineering
• Principal Investigator: Jaime Armando Delgado Vargas
• Grantee: Jaime Armando Delgado Vargas
• Company: FITOMAP - Mapeamento Agrícola Ltda. - ME
• Associated scholarship(s): 18/03308-0 - Phenotype maps of coffee plantation, BP.TT; 17/26677-8 - Phenotype maps of coffee plantation, BP.PIPE

Abstract

The automation of agricultural processes aimed increase the efficiency, accuracy of production at lowest cost. The monitoring of plantation allows the forecasting crop of a property. Also, it has an impact on crop quality, and consequently the monetary return to the grower. In the phytosanitary planning plantations, specifically the coffee crop forecast depend on the monitoring of pests, diseases and phenological index coffee production. In crop forecast the level of maturity of the fruit in the orchard indicates the best time to harvest, having direct impact on the quality, taste, flavor and price front coffee to the world market. Traditional monitoring techniques to estimate the next harvest are manual processes, susceptible to distractions and subjective inference. Even with these techniques they can estimate the current state of the coffee plant, remains a laborious, inaccurate and slow task because the measurements made by human workers. This process is not viable for permanent monitoring of large plantations. In order to know the quality and quantity of pre-harvest crop, a solution composed by: automatic mapping and localization system, image processing algorithms, sorting algorithms and pattern recognition are proposed. (AU)