Unsupervised Domain Adaptation for Plant Organ Counting

Ayalew, Tewodros W.; Ubbens, Jordan; Stavness, Ian

doi:10.1007/978-3-030-65414-6_23

Cited by 18 publications

(17 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In computer vision, attempting to alter training images to more closely match a target domain is common, as seen in Zhang et al, where they present a method for using domain adaptation for the detection of different fruits using a CycleGAN [18] demonstrating transformation between different fruits for a detection task. Domain adaptation has proven popular across most of the main phenotyping tasks such as leaf counting via regression as shown in [19] and wheat head detection as shown in [3] and in our own work.…”

Section: Domain Adaptationmentioning

confidence: 92%

“…In problem spaces such as plant phenotyping, this kind of approach is highly relevant owing to the high diversity of different plants and plant varieties; even within the same species, it is common to see a domain shift between different datasets of the same species due to imaging location and setup, time of year, the age of the plant, etc. In the past 5 years especially, synthetic to real domain adaptation has become popular [2,3] since automatically generated data are cheaper than the collection of real images and do not need manual annotation. Increased popularization of adversarial approaches has contributed to this increase in popularity, with many GAN-based models being used for domain adaptation.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Domain Adaptation of Synthetic Images for Wheat Head Detection

Hartley

French

2021

Plants

View full text Add to dashboard Cite

Wheat head detection is a core computer vision problem related to plant phenotyping that in recent years has seen increased interest as large-scale datasets have been made available for use in research. In deep learning problems with limited training data, synthetic data have been shown to improve performance by increasing the number of training examples available but have had limited effectiveness due to domain shift. To overcome this, many adversarial approaches such as Generative Adversarial Networks (GANs) have been proposed as a solution by better aligning the distribution of synthetic data to that of real images through domain augmentation. In this paper, we examine the impacts of performing wheat head detection on the global wheat head challenge dataset using synthetic data to supplement the original dataset. Through our experimentation, we demonstrate the challenges of performing domain augmentation where the target domain is large and diverse. We then present a novel approach to improving scores through using heatmap regression as a support network, and clustering to combat high variation of the target domain.

show abstract

Section: Domain Adaptationmentioning

confidence: 92%

Section: Introductionmentioning

confidence: 99%

Domain Adaptation of Synthetic Images for Wheat Head Detection

Hartley

French

2021

Plants

View full text Add to dashboard Cite

show abstract

“…And the image-level methods (Zhu et al, 2017;Wang et al, 2019; manipulate the styles of images, e.g., hues, illuminations, textures to make the images in two different domains closer. Some of the UDA methods are proposed to address the domain gap for plant counting (Giuffrida et al, 2019;Ayalew et al, 2020). However, existing UDA methods for plant counting directly adopt the generic feature-level UDA methods.…”

Section: Related Workmentioning

confidence: 99%

“…By contrast, UDA for object counting, especially for plant counting, has been less studied. To our knowledge, existing UDA methods (Giuffrida et al, 2019;Ayalew et al, 2020) applied to plant counting are often direct adoptions of generic UDA ideas without considering the particularities of domain gaps in plant counting. In fact, different from crowd counting or car counting, domain gaps in plant counting are much more diverse.…”

Section: Introductionmentioning

confidence: 99%

Background-Aware Domain Adaptation for Plant Counting

Shi

Lü

et al. 2022

Front. Plant Sci.

View full text Add to dashboard Cite

Deep learning-based object counting models have recently been considered preferable choices for plant counting. However, the performance of these data-driven methods would probably deteriorate when a discrepancy exists between the training and testing data. Such a discrepancy is also known as the domain gap. One way to mitigate the performance drop is to use unlabeled data sampled from the testing environment to correct the model behavior. This problem setting is also called unsupervised domain adaptation (UDA). Despite UDA has been a long-standing topic in machine learning society, UDA methods are less studied for plant counting. In this paper, we first evaluate some frequently-used UDA methods on the plant counting task, including feature-level and image-level methods. By analyzing the failure patterns of these methods, we propose a novel background-aware domain adaptation (BADA) module to address the drawbacks. We show that BADA can easily fit into object counting models to improve the cross-domain plant counting performance, especially on background areas. Benefiting from learning where to count, background counting errors are reduced. We also show that BADA can work with adversarial training strategies to further enhance the robustness of counting models against the domain gap. We evaluated our method on 7 different domain adaptation settings, including different camera views, cultivars, locations, and image acquisition devices. Results demonstrate that our method achieved the lowest Mean Absolute Error on 6 out of the 7 settings. The usefulness of BADA is also supported by controlled ablation studies and visualizations.

show abstract

“…Phenotyping is an important task for monitoring plants, similar to object tracking and identification. Ayalew et al [16] present a method to use an unsupervised domain adaptation network to adapt the meticulously pre-labeled Computer Vision Problems in Plant Phenotyping (CVPPP) dataset [17], [18] to other plant and image domains. The data consists of single plants, their leaves, and a point map of leaf centers.…”

Section: Related Workmentioning

confidence: 99%

AlphaGarden: Learning to Autonomously Tend a Polyculture Garden

Presten¹,

Avigal²,

Theis³

et al. 2021

Preprint

View full text Add to dashboard Cite

This paper presents AlphaGarden: an autonomous polyculture garden that prunes and irrigates living plants in a 1.5m × 3.0m physical testbed. AlphaGarden uses an overhead camera and sensors to track the plant distribution and soil moisture. We model individual plant growth and interplant dynamics to train a policy that chooses actions to maximize leaf coverage and diversity. For autonomous pruning, AlphaGarden uses two custom-designed pruning tools and a trained neural network to detect prune points. We present results for four 60-day garden cycles. Results suggest AlphaGarden can autonomously achieve 0.96 normalized diversity with pruning shears while maintaining an average canopy coverage of 0.86 during the peak of the cycle. Code, datasets, and supplemental material can be found at https://github.com/BerkeleyAutomation/AlphaGarden.

show abstract

Unsupervised Domain Adaptation for Plant Organ Counting

Cited by 18 publications

References 39 publications

Domain Adaptation of Synthetic Images for Wheat Head Detection

Domain Adaptation of Synthetic Images for Wheat Head Detection

Background-Aware Domain Adaptation for Plant Counting

AlphaGarden: Learning to Autonomously Tend a Polyculture Garden

Contact Info

Product

Resources

About