Integration of visible branch sections and cherry clusters for detecting cherry tree branches in dense foliage canopies

To minimise the demand for seasonal workers in sweet cherry production, there is a need to develop automated harvesting systems. The first step in automating a shake-and-catch type harvesting system is to develop a machine vision system for detecting tree branches and localising shaking points in those branches. In this study, an image processing algorithm was developed to detect branches of cherry trees using segmentation of branch and cherry pixels. Firstly, partially visible branch segments within the tree canopies were connected using morphological features of the segments to form whole branches. Then, the positions of cherry clusters in the canopy were used as an indication to detect branch sections that were occluded by cherries and leaves. Different cherry clusters were grouped together based on their spatial location and distance between them. Branch equations were then defined through those cherry clusters using minimum residual criteria. Overall, 93.8% branches were detected in a Y-trellis fruiting wall cherry orchard, with 55.0% of branches detected using only branch pixels and 38.8% additional branches detected using cherry clusters. The method resulted in a total of 12.4% of false positive detection. The results showed that branch detection accuracy can be substantially improved by integrating cherry location information with the location of segments of partially visible branches. This study has shown the potential of machine vision systems to detect cherry tree branches in full foliage season, which is highly promising for the development of automated sweet cherry harvesting systems.