Abstract:

We present a multiple instance learning class activation map (MIL-CAM) approach for pixel-level minirhizotron image segmentation given weak image-level labels. Minirhizotrons are used to image plant roots in situ. Minirhizotron imagery is often composed of soil containing a few long and thin root objects of small diameter. The roots prove to be challenging for existing semantic image segmentation methods to discriminate. In addition to learning from weak labels, our proposed MILCAM approach re-weights the root versus soil pixels during analysis for improved performance due to the heavy imbalance between soil and root pixels. The proposed approach outperforms other attention map and multiple instance learning methods for localization of root objects in minirhizotron imagery.

Links:

Citation:

G. Yu, A. Zare, W. Xu, R. Matamala, J. Reyes-Cabrera, F. B. Fritschi and T. E. Juenger, "Weakly Supervised Minirhizotron Image Segmentation with MIL-CAM," in the 16th European Conference on Computer Vision (ECCV) Workshop on Computer Vision Problems in Plant Phenotyping (CVPPP 2020),  Lecture Notes in Computer Science, vol 12540. Springer, Cham. https://doi.org/10.1007/978-3-030-65414-6_30, 2020.

@Article{Yu2020MILCAM,
Title = {Weakly Supervised Minirhizotron Image Segmentation with MIL-CAM}, 
Author = {Guohao Yu and Alina Zare and Weihuang Xu and Roser Matamala and Joel Reyes-Cabrera and Felix B. Fritschi and Thomas E. Juenger},  
Journal = {16th European Conference on Computer Vision (ECCV) Workshop on Computer Vision Problems in Plant Phenotyping (CVPPP 2020)}, 
Volume = {12540},  
DOI = {https://doi.org/10.1007/978-3-030-65414-6_30}, 
ISBN={978-3-030-65414-6},
Year = {2020}
}