The core of the method is a two-stage pipeline. First, photographs of rockfill material—whether spread on a laboratory bench or compacted in layers on a dam site—are fed into Cellpose, a generalist deep learning algorithm originally developed for segmenting individual cells in microscopy images. Cellpose’s neural network learns to delineate the boundaries of touching, overlapping, irregularly shaped objects, which makes it unusually well suited to the visual chaos of a pile of quarried rock. The algorithm produces a high-precision segmentation mask in which every individual particle is separated from its neighbors, solving the classic bottleneck that has limited automated rock image analysis: stones that touch or partially overlap are notoriously difficult for conventional thresholding techniques to separate.

Once Cellpose has produced the segmented image, the researchers turn to the MorphoLibJ plugin of ImageJ, a mature mathematical morphology library, for post-processing and quantitative computation. From each segmented particle, the software extracts geometric features including area, particle size, and aspect ratio. Aggregated over the hundreds or thousands of stones visible in a single image, these features yield the particle size distribution curve—one of the most fundamental descriptors of any granular construction material—without the sieving, weighing, and labor that traditional gradation testing demands. The team reports that the ImageJ-Cellpose method could accurately determine both the particle size distribution curve and the porosity of rockfill materials, results that were validated against conventional screening and porosity experiments conducted in the laboratory.

Beyond reproducing established metrics, the study introduces new shape parameters for rockfill material, including what the authors call the long-short ratio and the area ratio. Shape matters in geotechnical engineering more than casual observers might expect: angular, elongated, and flat particles interlock differently than rounded, equidimensional ones, influencing shear strength, particle breakage under load, and the volumetric behavior of the compacted mass. By statistically analyzing the distribution of these new shape descriptors across particle populations, the researchers offer a quantitative framework for describing rockfill morphology that goes beyond simple size gradation, opening a path toward correlating shape statistics directly with mechanical performance.

The critical question for any laboratory method is whether it survives contact with the real world. To answer it, the team partnered with a pumped-storage power station, where layered rolling experiments were carried out during actual rockfill placement. As the material was compacted by rollers in successive passes, images were collected at each stage and processed through the intelligent recognition pipeline. The results showed that the method could effectively trace how the particle size distribution and porosity of the rockfill evolved with each rolling pass—precisely the kind of real-time feedback that compaction quality control has historically lacked. Encouragingly, the trends identified by the image-based approach aligned well with those derived from the pit measuring method, the traditional but destructive and labor-intensive standard for evaluating compaction in the field.

The significance of this validation lies in what it could change about dam construction practice. Rockfill dams are among the largest man-made structures on Earth, and their long-term safety depends on how well the fill is compacted during construction; poor compaction leads to post-construction settlement, cracking of upstream concrete face slabs, and heightened vulnerability to overtopping failure and seismic loading. Current quality control typically relies on periodic pit sampling, in which crews excavate a hole, weigh the removed material, and measure density—a slow process that samples only a tiny fraction of the dam’s volume. A camera, a laptop running Cellpose and ImageJ, and the proposed analytical workflow could in principle assess compaction indicators continuously and across the entire working surface, converting construction monitoring from intermittent spot checks into comprehensive surveillance.

Technically, the study is notable for transplanting tools from an unexpected domain. Cellpose was created by biologists to solve the problem of segmenting cells in diverse image types without retraining a network for each new dataset, and its generalist design proved transferable to geological imagery. ImageJ, likewise, grew out of biomedical imaging but ships with general-purpose morphology operators that apply equally well to mineral particles. By connecting a state-of-the-art segmentation network to a battle-tested analysis library, the Xi’an team sidestepped the need to build bespoke deep learning infrastructure or large annotated rockfill datasets from scratch, lowering the barrier to adoption for engineering firms and construction supervisors who cannot maintain their own machine learning teams.

The method also complements a growing body of work on computer vision in geotechnical monitoring. Prior studies have used video image recognition to detect rockfill gradation, deep learning segmentation to calculate gradation from photographs, and instance segmentation to predict gradation for engineering projects. What distinguishes the new approach is its end-to-end coverage: rather than estimating only particle sizes, it simultaneously delivers size distribution, individual particle shape metrics through newly proposed parameters, and porosity, all from ordinary images, and it does so in a form validated against both sieving tests in the laboratory and pit measurements on an active construction site. The authors argue that this makes the technique suitable for real-world engineering applications and a robust means of ascertaining particle size information during the construction of rockfill dams.

Looking forward, the research points toward a swift and precise evaluation method for the compaction quality of rockfill material—a capability the authors describe as paramount for the comprehensive understanding and analysis of the safety conditions of rockfill dams and similar projects. If deployed at scale, the ImageJ-Cellpose workflow could give dam engineers a continuous, quantitative picture of how each roller pass is reshaping the granular skeleton of an embankment, catch gradation or compaction anomalies before they are buried under the next lift of rock, and feed high-quality geometric data into the numerical models used to predict settlement, seepage, and seismic response. In an era when aging dam infrastructure faces intensifying climate and seismic stresses, the idea that a deep learning algorithm designed for counting cells can help safeguard structures holding back billions of liters of water is a striking demonstration of how tools migrate across scientific boundaries—and of how much of modern construction safety may soon rest on image recognition pipelines that work quietly, frame by frame, on the rocks beneath the rollers.

Subject of Research: Intelligent image-based geometric characterization of rockfill materials using deep learning segmentation for dam compaction analysis

Article Title: Intelligent geometric characterization of rockfill materials using ImageJ-Cellpose: a novel approach for accurate morphological analysis

Article References: Ma, C., Zhou, C., Hou, Y., & Cheng, L. (2026). Intelligent geometric characterization of rockfill materials using ImageJ-Cellpose: a novel approach for accurate morphological analysis. Cluster Computing, 29(13), Article 779. https://doi.org/10.1007/s10586-026-06480-4

Image Credits: AI Generated

DOI: 10.1007/s10586-026-06480-4

Keywords: rockfill materials, deep learning, Cellpose, ImageJ, image segmentation, particle size distribution, porosity, particle shape, compaction quality, rockfill dams, image recognition, geotechnical engineering

At the heart of the new methodology is RONUM, an ImageJ-based disease quantification workflow that converts ordinary leaf photographs and scans into precise numerical disease metrics. Rather than relying on fully automated machine learning models that demand large training datasets and specialized computing infrastructure, RONUM uses a user-guided calibration approach. A researcher first teaches the system what infected and healthy tissue looks like within their specific experimental context, and the workflow then applies that calibration consistently across entire replicate datasets. This design choice matters enormously for practical plant pathology, where lighting conditions, leaf health, and disease presentation can vary dramatically between experiments, growth facilities, and pathogen isolates.

The team validated RONUM against manually corrected reference image sets derived from both whole-plant leaf scans and detached-leaf samples. The results were striking: disease estimates showed strong agreement with expert-verified measurements, and critically, the workflow proved reproducible when independent users performed their own calibrations. The agreement was further confirmed by pixel-level comparison against reference masks, demonstrating that different researchers could arrive at essentially the same quantitative answer from the same images. This reproducibility addresses one of the most persistent criticisms of visual disease scoring, where inter-rater variability can obscure genuine biological differences between wheat lines or treatments.

Benchmarking revealed another important advantage. When compared against another published ImageJ workflow, RONUM showed closer agreement with manually corrected measurements, particularly for images containing chlorosis, necrosis, or leaves in generally poor health. This distinction is far from trivial. Rust infections frequently trigger yellowing and tissue death that confounds simple color-based segmentation, causing automated pipelines to either overestimate disease by counting discolored tissue or underestimate it by missing subtle early symptoms. RONUM’s flexible calibration allows researchers to adapt the analysis to these challenging visual conditions rather than forcing their experiments to fit rigid algorithmic assumptions.

The researchers also took the unusual step of systematically evaluating the post-processing options available within the workflow. Their quantitative assessment showed that these choices had little influence on total pustule-area estimates, but substantially affected pustule count measurements. For researchers studying pathogen reproduction dynamics or the early stages of infection, where the number of individual lesions can be as informative as their total area, this finding provides practical guidance on how to standardize their analysis pipelines and avoid inadvertently introducing bias through arbitrary parameter selection.

To demonstrate real-world applications, the team paired RONUM with two complementary infection assay systems. The first is a soil-free whole-plant assay that enabled non-destructive tracking of wheat leaf rust development across an eighteen-day infection time course. By imaging the same plants repeatedly as the disease progressed, researchers could construct quantitative disease curves rather than relying on endpoint snapshots. This longitudinal approach captures the full trajectory of infection, revealing differences in disease onset, progression rate, and final severity that a single time-point measurement would miss entirely. The soil-free growth conditions, established with assistance from Dr Chris Brosnan, additionally reduce the space, cost, and waste associated with conventional soil-based plant culture.

The second assay is a benchtop-scale detached-leaf system that proved capable of delivering reproducible disease quantification across thirteen wheat lines with contrasting disease phenotypes. Detached-leaf assays dramatically increase screening throughput, allowing many genotypes to be evaluated simultaneously on a laboratory bench without the overhead of maintaining mature plants. In an exploratory analysis that showcases the biological depth of the approach, the researchers quantified the ratio of chlorosis to pustule area across these lines. This single derived metric substantially separated susceptible lines from those carrying pathotype-specific Lr resistance genes, suggesting that image-based phenotyping can capture biologically meaningful host responses, such as resistance-associated yellowing, that are typically invisible to conventional image quantification.

Perhaps the most immediately practical demonstration came in the form of a modified detached-leaf assay equipped with a nylon-mesh treatment reservoir, designed specifically for chemical screening. The team used this system to establish a dose-response curve for a commercial propiconazole fungicide formulation, deriving an estimated IC₅₀ of 20.49 ppm, with a 95 percent confidence interval spanning 14.64 to 28.11 ppm. This means the workflow can quantify the efficacy of disease control agents with proper statistical confidence using standard laboratory equipment. For agrochemical companies, academic screening programs, and breeding pipelines searching for novel control strategies, the ability to generate rigorous dose-response data at benchtop scale represents a significant democratization of what has traditionally required substantial infrastructure.

The principal value of RONUM, as the authors emphasize, lies not in any single technical breakthrough but in its flexible, accessible design philosophy. The workflow calibrates to individual experimental systems and then quantifies rapidly and consistently across replicate datasets, generating the annotations and output summaries needed for streamlined record keeping. Combined with the whole-plant and detached-leaf assays, it enables quantitative assessment of disease progression and screening of candidate treatments or putative resistance traits using freely available software and standard laboratory equipment. The research was supported by the Grains Research and Development Corporation and the Australian Research Council Research Hub for Sustainable Crop Protection, and Anne Sawyer was supported by an Advance Queensland Industry Research Fellowship.

The authors note that further validation will be required before these tools reach their full potential, and the preliminary nature of some analyses, particularly the chlorosis-based resistance discrimination, is honestly acknowledged. Yet the significance of the work is difficult to overstate in a field where wheat rust pathogens continuously evolve to overcome resistance genes, making rapid phenotyping of breeding material an ongoing arms race. By lowering the technical and financial barriers to quantitative disease assessment, this suite of methods could accelerate the identification of durable resistance and effective control agents. With rust outbreaks capable of destroying entire harvests in a single season, tools that transform subjective visual scores into reproducible numbers may prove as important to food security as any single resistance gene.

Subject of Research: A semi-automated image-based method for quantifying wheat leaf rust disease progression, resistance, and fungicide response

Article Title: A rapid, semi-automated image-based method for quantitative assessment of rust disease progression in wheat

Article References: Mann, C. W. G., Periyannan, S., Carroll, B. J., Gardiner, D. M., & Sawyer, A. (2026). A rapid, semi-automated image-based method for quantitative assessment of rust disease progression in wheat. Plant Methods. https://doi.org/10.1186/s13007-026-01590-x

Image Credits: AI Generated

DOI: 10.1186/s13007-026-01590-x

Keywords: wheat, leaf rust, RONUM, ImageJ, image analysis, plant pathology, disease quantification, phenotyping, fungicide screening, detached leaf assay, Puccinia triticina, crop protection