Narrow Results

This paper presents the design process of an embedded stereo vision system, which investigates the most relevant criteria for developing the hardware and software architectures for plant phenotyping. In other words, this paper is the result of a preliminary study in which the main motivation was the evaluation of the viability of a low-cost visual system for such field of knowledge. In addition, the implications of the adversities in an actual agricultural scenario under the system design are presented, since the system should not only meet the portability requirements but also the quality and precision for the measurements carried out by cameras. After the use of such method, the systems obtained may present a high chance of satisfying a set of constraints, and meeting their possibility to be used for machine vision applied in agricultural decision-making processes related to plant architecture and in situ recognition.

Total green leaf area (GLA) is an important trait for agronomic studies. However, existing methods for estimating the GLA of individual rice plants are destructive and labor-intensive. A nondestructive method for estimating the total GLA of individual rice plants based on multi-angle color images is presented. Using projected areas of the plant in images, linear, quadratic, exponential and power regression models for estimating total GLA were evaluated. Tests demonstrated that the side-view projected area had a stronger relationship with the actual total leaf area than the top-projected area. And power models fit better than other models. In addition, the use of multiple side-view images was an efficient method for reducing the estimation error. The inclusion of the top-view projected area as a second predictor provided only a slight improvement of the total leaf area estimation. When the projected areas from multi-angle images were used, the estimated leaf area (ELA) using the power model and the actual leaf area had a high correlation coefficient (R² > 0.98), and the mean absolute percentage error (MAPE) was about 6%. The method was capable of estimating the total leaf area in a nondestructive, accurate and efficient manner, and it may be used for monitoring rice plant growth.