Abstract:  In order to extract crop organ from images accurately in precision agriculture, an image-based method based on low-rank matrix recovery is proposed. A crop image is considered to be compose of two factors: background and organ. In a certain feature space, the image is represented as a low-rank matrix plus sparse noises. The organ is then extracted by identifying the sparse noises when using low-rank matrix recovery algorithm. To ensure the rank of background is low, a linear transform for the feature space is introduced and that historical data is ured . Dynamic threshold segmentation followed by vegetation removing techniques are ultimately adopted in the final step. The experimental results on the benchmark farmland show that our method achieves competitive performance, compared to the other well-established methods, yielding the highest performance of 93. 9% with the lowest standard deviation of 2. 86%, which means our method is more robust and not sensitive to the complex environmental elements and different cultivars.

Key words:  crop organ, automatic extraction, low-rank matrix recovery, feature space transformation