X-ray Computed Tomography Image Processing &amp; Segmentation: A Case Study Applying Machine Learning and Deep Learning-Based Strategies

Ferreira, Talita R.; Cássaro, Fábio Augusto Meira; Zhou, Hu; Pires, Luiz F.

doi:10.1007/978-3-031-12176-0_5

Cited by 4 publications

(5 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Currently, deep learning and machine learning strategies are used for fast and efficient image processing. This type of data analysis allows extracting information from complex datasets [80][81][82]. Important detailed information on the use of computed tomography applied to soil physics can be found in Mooney et al [83].…”

Section: Perspectives For Studies Related To Movement Of Water and So...mentioning

confidence: 99%

Movement of Water and Solutes in Agricultural Soils

Nascimento Pessoa,

Martineli Batista,

Nogueira de Sousa

et al. 2024

Agricultural Sciences

View full text Add to dashboard Cite

Intensive agricultural systems have increased in recent decades. These systems usually provide high soil disturbance that promotes negative consequences on soil physical, chemical, and biological processes. Soil compaction degrades soil structure causing changes in soil porosity, pore size distribution, and pore space geometry. Pore space dynamics affects the movement of water and the transport of solutes in soils because both are highly dependent on the properties of this space. Agricultural activities, which include the use of mechanization, soil fertilizers, herbicides, and pesticides with the aim of creating a favorable condition for plant growth, can favor the soil erosion processes and increase the transport of contaminants to groundwater. Models for predicting water movement and transport of solute in soils are important to understand different scenarios of groundwater contamination and to improve appropriate soil management. Therefore, negative consequences and environmental impacts related to agricultural practices should be studied.

show abstract

Section: Perspectives For Studies Related To Movement Of Water and So...mentioning

confidence: 99%

Movement of Water and Solutes in Agricultural Soils

Nascimento Pessoa,

Martineli Batista,

Nogueira de Sousa

et al. 2024

Agricultural Sciences

View full text Add to dashboard Cite

show abstract

“…4 Recent advancements in X-ray CT application X-ray computed tomography is a well-established technique to characterize different aspects of soil and plant in 3D and spatially. Ferreira et al (2022) intended to look at the impacts of soil surface liming on the intraaggregate pore structure of soil aggregates using synchrotron-based computed microtomography. The IMX Beamline of the Brazilian Synchrotron Light Source Facility (LNLS -CNPEM) was used for the scanning, with a pink beam, 550 μm Silicon filter, and 1,024 projections over an angle range of 180 °, with a voxel size of 1.64 × 1.64×1.64.…”

Section: Habitat Architecturementioning

confidence: 99%

“…5 Machine learning and deep learning in X-ray CT image processing CT imaging of soil samples (after reconstruction) produces grayscale images with a series of grey values corresponding to soil components with varying densities. The majority of current techniques for quantifying the soil pore system rely on binary images, with target objects or foreground pixels (e.g., pores) labeled as 1 and background pixels (e.g., soil) labeled as 0 (Ferreira et al, 2022). This means that before further quantification, the grayscale images must be segmented into binary images.…”

Section: Habitat Architecturementioning

confidence: 99%

Application of X-ray computed tomography in soil and plant -a review

Ghosh,

Maity,

Rabbi

et al. 2023

Front. Environ. Sci.

View full text Add to dashboard Cite

X-ray computed tomography (X-ray CT) is a non-destructive method of soil analysis which can provide three-dimensional (3D) view, quantitative information of the internal organization of the soil. In this paper, we discuss the potential application of X-ray CT in characterization of soil properties like porosity and pore size distribution (PSD), root architecture, soil phase classification, water and solute transport in soil, and highlight the research during last 10–15 years. Here, we review the recent development of X-ray CT in soil science, use of artificial intelligence and machine learning in image analysis, point out the major challenges associated with its use, discuss few improvements to overcome these difficulties and elaborate the possible future technological developments for non-invasive/destructive soil characterization by integrating X-ray CT with recently available complementary techniques.

show abstract

“…This is important since a classical ground truth via root excavation and washing, results in a loss of fine root structure and interconnectivity of the whole root system. To this end, the 3D segmentation of the root structures was mainly performed by analytical algorithms based on classical image processing and image analysis methods [e.g., Mairhofer et al., 2011 ; Flavel et al., 2012 ; Mairhofer et al., 2015 ; Flavel et al., 2017 ; Gao et al., 2019 ; Soltaninejad et al., 2020 ; Gerth et al., 2021 ; Phalempin et al., 2021 ; Ferreira et al, 2022 ; Lucas & Vetterlein, 2022 ), which, however, are not able to detect roots on all scales equally. See Figure 1 for an example which contains challenging small and fine roots.…”

Section: Introductionmentioning

confidence: 99%

3D segmentation of plant root systems using spatial pyramid pooling and locally adaptive field-of-view inference

et al. 2023

View full text Add to dashboard Cite

BackgroundThe non-invasive 3D-imaging and successive 3D-segmentation of plant root systems has gained interest within fundamental plant research and selectively breeding resilient crops. Currently the state of the art consists of computed tomography (CT) scans and reconstruction followed by an adequate 3D-segmentation process.ChallengeGenerating an exact 3D-segmentation of the roots becomes challenging due to inhomogeneous soil composition, as well as high scale variance in the root structures themselves.Approach(1) We address the challenge by combining deep convolutional neural networks (DCNNs) with a weakly supervised learning paradigm. Furthermore, (2) we apply a spatial pyramid pooling (SPP) layer to cope with the scale variance of roots. (3) We generate a fine-tuned training data set with a specialized sub-labeling technique. (4) Finally, to yield fast and high-quality segmentations, we propose a specialized iterative inference algorithm, which locally adapts the field of view (FoV) for the network.ExperimentsWe compare our segmentation results against an analytical reference algorithm for root segmentation (RootForce) on a set of roots from Cassava plants and show qualitatively that an increased amount of root voxels and root branches can be segmented.ResultsOur findings show that with the proposed DCNN approach combined with the dynamic inference, much more, and especially fine, root structures can be detected than with a classical analytical reference method.ConclusionWe show that the application of the proposed DCNN approach leads to better and more robust root segmentation, especially for very small and thin roots.

show abstract

X-ray Computed Tomography Image Processing & Segmentation: A Case Study Applying Machine Learning and Deep Learning-Based Strategies

Cited by 4 publications

References 69 publications

Movement of Water and Solutes in Agricultural Soils

Movement of Water and Solutes in Agricultural Soils

Application of X-ray computed tomography in soil and plant -a review

3D segmentation of plant root systems using spatial pyramid pooling and locally adaptive field-of-view inference

Contact Info

Product

Resources

About