Biology subjects

Kazic, T.

Publications and source records attributed to Kazic, T..

3 recordsLinked to original sources

bioRxiv2025

MaiZaic: a robust end-to-end pipeline for mosaicking freely flown aerial video of agricultural fields

Unmanned aerial vehicles (UAVs) are increasingly used for high throughput phenotyping. In principle, freely flown vehicles would permit real-time flexibility in identifying and scouting regions of interest. Mosaicking multiple images provides a high resolution global image and consumer-grade UAVs offer low cost, ease of flying, and excellent RGB cameras. The vehicles inaccurate telemetry complicates estimating the homographies between pairs of frames, the standard mosaicking approach. Moreover, errors accumulate during computation, distorting later portions of the mosaic. Finally, crop fields are particularly challenging to mosaic because their planting is so regular and the plants are so similar, eliminating distinctive features that could guide mosaicking. We propose MaiZaic, an end-to-end pipeline that dynamically samples video frames using optical flow, automates camera and gimbal calibration, estimates homographies with an unsupervised convolutional neural network, detects shots among frames, and generates mini-mosaics. Together, these techniques significantly reduce errors in the output mosaics. Our deep learning model is trained on a comprehensive video dataset comprising different flight trajectories, maize lines, growth stages, and augmented illumination data. MaiZaic is more accurate and faster than ASIFT and more robust than our earlier CorNet and CorNetv2. We demonstrate MaiZaics effectiveness in generating accurate mosaics of imagery captured by freely-flown UAVs and explore its generalizability. Core ideasO_LIMaiZaic is an end-to-end pipeline to mosaic freely flown agricultural imagery captured with consumer-grade UAVs. C_LIO_LIMaiZaic introduces novel algorithms that efficiently choose video frames, calibrate, and mosaic the imagery. C_LIO_LIOur unsupervised deep homography estimator, CorNetv3, is 14 times faster and 8.59% more accurare than ASIFT. C_LIO_LIMaiZaic generalizes well and mosaicks maize at different growth stages, objects, trajectories, cameras, and pilots. C_LIO_LIThe mini-mosaicking algorithm improves mosaic accuracy by interrupting error accumulation. C_LI

plant biology↗

bioRxiv2020

Chloe: Flexible, Efficient Data Provenance and Management

1Reproducible and sharable research requires robust data provenance during and after the experimental process. Each laboratory and experiment has its own goals and methods, and these change frequently. Planning, managing, and collecting data from research crops are particularly labor-intensive tasks, given the tightly compressed time schedule and the operating environments. Moving from a labs present record-keeping approach to an electronic ecosystem that improves provenance is an additional burden for groups without dedicated, consistent computational support to make that transition and then to adapt the system as needed. This high barrier to entry and the press of field work makes it easy to postpone "computerizing". I have developed Chloe to reduce manual effort during experiments and maintain data provenance. A flexible, modular system, Chloe integrates simple equipment, data collection strategies, and software into workflows. The design lets one use parts without deploying the whole. This reduces the barriers to entry while still improving workflow efficiency and making Chloe accessible to a wide range of users. I offer guidance on ways to adapt Chloe to ones own experimental situation. Chloe has been tested and refined with many changes of students, hardware, and experimental goals over the last fourteen years. Though originally designed for maize genetics and computational experiments, Chloe can accommodate other types of experiments, wetbench work, and other crops.

bioinformatics↗

bioRxiv2017

Dose-response surface fits to drought and nitrogen limitation applied together allow mapping of loci that exhibit nonlinear responses

Crop improvement must accelerate to feed an increasing human population in the face of environmental changes. Breeding programs can include anticipated climatic changes and genetic architecture to optimize improvement strategies. We analyzed the genetic architecture underlying the response of Zea mays to combinations of water and nitrogen stresses. Recombinant inbreds were subjected to nine combinations of the two stresses using an optimized response surface design, and their growth was measured. Three-dimensional dose response surfaces were fit globally and to each polymorphic allele to determine which genetic markers were associated with different response surfaces. Three quantitative trait loci that produced nonlinear surfaces were mapped. Alleles that performed better in combinations of mid-range stresses were typically not the alleles that performed best under combinations of extreme stresses. To develop physiologically relevant models for future genetic analyses, we modeled the network that explains the response surfaces. The network contains two components, an elliptical paraboloid and a plane, that each combine the nitrogen and water inputs. The relative weighting of the two components and the inputs is governed by five parameters. We estimated parameter values for the smoothed surfaces from the experimental lines using a set of points that covered the most distinctive regions of the three-dimensional surfaces. Surfaces computed using these values reproduced the smoothed experimental surfaces well, especially in the neighborhood of the peaks, as judged by three different criteria. The parameters exaggerated the amplitudes of the simulated surfaces. Experiments using single stresses could misestimate responses to their combinations and disguise loci that respond nonlinearly. The three-dimensional shape evaluation strategy used here more thoroughly compares nonlinear, nonplanar responses. We encourage the application of our findings and methods to experiments that mix crop protection measures, stresses, or both, on elite and landrace germplasm.

genetics↗