Abstract
In this paper, we present a challenging task of 3D segmentation of individual plant leaves from occlusions in the complicated natural scene. Depth data of plant leaves is introduced to improve the robustness of plant leaf segmentation. The low cost RGB-D camera is utilized to capture depth and color image in fields. Mean shift clustering is applied to segment plant leaves in depth image. Plant leaves are extracted from the natural background by examining vegetation of the candidate segments produced by mean shift. Subsequently, individual leaves are segmented from occlusions by active contour models. Automatic initialization of the active contour models is implemented by calculating the center of divergence from the gradient vector field of depth image. The proposed segmentation scheme is tested through experiments under greenhouse conditions. The overall segmentation rate is 87.97% while segmentation rates for single and occluded leaves are 92.10% and 86.67%, respectively. Approximately half of the experimental results show segmentation rates of individual leaves higher than 90%. Nevertheless, the proposed method is able to segment individual leaves from heavy occlusions.
Highlights
Precision agriculture is a synthesis technology that enhancing crop production with minimal energy costs and environmental pollution [1,2]
These images contain 474 target leaves that 24.05% (114 leaves) of them are single leaves while the rest 75.95% (360 leaves) of leaves are with occlusions (Table 1)
Plant leaf segmentation from a natural background is implemented based on mean shift clustering
Summary
Precision agriculture is a synthesis technology that enhancing crop production with minimal energy costs and environmental pollution [1,2]. Precision agriculture has been rapidly developed by introducing advanced technologies, such as intelligent sensors and robotics. Due to the success of computer vision techniques, imaging sensors have become the most common sensing devices in agricultural automation systems for collecting information of plants. Image analysis of plant leaves is one of the essential tasks for agricultural automation since a plant leaf contains abundant information of plants. Automatic detection of individual leaves is a fundamental task for achieving precision operations in agricultural practices. Genetic algorithms are introduced to extract individual leaves from canopy images [8]. Watershed based leaf segmentation is reported in [9], which efficiently extracts plant leaves from the images taken from tomato fields. Neural network is demonstrated high performance in detecting vegetation pixels and extract leaves from the ground [10]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have