Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Image Segmentation”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 415 records · Page 23

ASDFL: An adaptive super‐pixel discriminative feature‐selective learning for vehicle matching

Abstract There are a large number of cameras in modern transportation system that capture numerous vehicle images continuously. Therefore, automatic analysis of these vehicle images is helpful for traffic flow management, criminal investigations and vehicle inspections. Vehicle matching, which aims to determine whether two input images depict an identical vehicle, is one of the core tasks in vehicle analysis. Recent relevant studies have focused on local feature extraction instead of global extraction, since local details can provide crucial cues to distinguish between cars. However, these methods do not select local features; that is, they do not assign weights to local features. Therefore, in this research, we systematically study the vehicle matching task, and present a novel annotation‐free local‐based deep learning method called Adaptive super‐pixel discriminative feature‐selective learning (ASDFL) to address this issue. In ASDFL, vehicle images are segmented into clusters of super‐pixels of similar size by considering the location and colour similarities of pixels without using any component‐level annotation. These super‐pixels are deemed to be the virtual components of vehicles. Moreover, a convolutional neural network is used to extract the deep features of these virtual components. Thereafter, an instance‐specific mask generation module driven by the extracted global features is enhanced to produce a mask to select the most distinctive virtual components of each vehicle image pair in the feature space. Finally, the vehicle matching task is accomplished by classifying the selected virtual component features of each imaged vehicle pair. Extensive experiments on two popular vehicle identification benchmarks demonstrate that our method is 1.57% and 0.8% more accurate than the previous baselines in a vehicle matching task on the VeRi and VehicleID datasets, respectively, which demonstrates the effectiveness of our method.

Qin, Rong↗

Convolutional Neural Network for Segmenting Micro-X-ray Computed Tomography Images of Wood Cellular Structures

To further enhance the performance of wood products, improved tools are needed to study in situ cellular scale phenomena like mechanical deformations and moisture swelling. Micro-X-ray computed tomography (μXCT) using brilliant synchrotron light sources now has the spatial and temporal resolution for real-time visualization of phenomena in three-dimensional cellular structures. However, the tradeoff for speed includes the loss of intensity contrast between different types of materials within the imaged structure, such as cell wall and air in wood. This loss of contrast prevents traditional histogram-based segmentation methods from being used effectively. A new convolutional neural network (CNN) approach was therefore developed to segment fast μXCT images of wood into cell wall and air volumes. The fast μXCT and segmentation were demonstrated in the study of moisture swelling in loblolly pine (Pinus taeda) earlywood and latewood cellular structures conditioned at 0%, 33%, 75%, and 95% relative humidity (RH). The CNN segmentation results had a mean intersection over union (IoU) metric accuracy of 96%. Initial analysis of the swelling in the latewood revealed cell walls swelled about 25% when conditioned from 0% to 95% RH. Additionally, the widths of ray cell lumina in the transverse plane of latewood could be observed to increase at higher RH. The segmentation method presented here will facilitate future quantitative analyses in in situ μXCT studies of wood and other similar cellular materials.

Arzola-Villegas, Xavier (ORCID:0000000305369766)↗

SU-E-T-13: A Feasibility Study of the Use of Hybrid Computational Phantoms for Improved Historical Dose Reconstruction in the Study of Late Radiation Effects for Hodgkin's Lymphoma

Purpose: Radiation-induced cancer effects are well-documented following radiotherapy. Further investigation is needed to more accurately determine a dose-response relationship for late radiation effects. Recent dosimetry studies tend to use representative patients (Taylor 2009) or anthropomorphic phantoms (Wirth 2008) for estimating organ mean doses. In this study, we compare hybrid computational phantoms to patient-specific voxel phantoms to test the accuracy of University of Florida Hybrid Phantom Library (UFHP Library) for historical dose reconstructions. Methods: A cohort of 10 patients with CT images was used to reproduce the data that was collected historically for Hodgkin's lymphoma patients (i.e. caliper measurements and photographs). Four types of phantoms were generated to show a range of refinement from reference hybrid-computational phantom to patient-specific phantoms. Each patient is matched to a reference phantom from the UFHP Library based on height and weight. The reference phantom is refined in the anterior/posterior direction to create a ‘caliper-scaled phantom’. A photograph is simulated using a surface rendering from segmented CT images. Further refinement in the lateral direction is performed using ratios from a simulated-photograph to create a ‘photograph and caliper-scaled phantom’; breast size and position is visually adjusted. Patient-specific hybrid phantoms, with matched organ volumes, are generated and show the capabilities of the UF Hybrid Phantom Library. Reference, caliper-scaled, photograph and caliper-scaled, and patient-specific hybrid phantoms are compared with patient-specific voxel phantoms to determine the accuracy of the study. Results: Progression from reference phantom to patient specific hybrid shows good agreement with the patient specific voxel phantoms. Each stage of refinement shows an overall trend of improvement in dose accuracy within the study, which suggests that computational phantoms can show improved accuracy in historical dose estimates. Conclusion: Computational hybrid phantoms show promise for improved accuracy within retrospective studies when CTs and other x-ray images are not available.

60 APPLIED LIFE SCIENCES↗

Factorization-based texture segmentation

This study introduces a factorization-based approach that efficiently segments textured images. We use local spectral histograms as features, and construct an M × N feature matrix using M-dimensional feature vectors in an N-pixel image. Based on the observation that each feature can be approximated by a linear combination of several representative features, we factor the feature matrix into two matrices-one consisting of the representative features and the other containing the weights of representative features at each pixel used for linear combination. The factorization method is based on singular value decomposition and nonnegative matrix factorization. The method uses local spectral histograms to discriminate region appearances in a computationally efficient way and at the same time accurately localizes region boundaries. Finally, the experiments conducted on public segmentation data sets show the promise of this simple yet powerful approach.

97 MATHEMATICS AND COMPUTING↗

Automated registration of multispectral MR vessel wall images of the carotid artery

Purpose: Atherosclerosis is the primary cause of heart disease and stroke. The detailed assessment of atherosclerosis of the carotid artery requires high resolution imaging of the vessel wall using multiple MR sequences with different contrast weightings. These images allow manual or automated classification of plaque components inside the vessel wall. Automated classification requires all sequences to be in alignment, which is hampered by patient motion. In clinical practice, correction of this motion is performed manually. Previous studies applied automated image registration to correct for motion using only nondeformable transformation models and did not perform a detailed quantitative validation. The purpose of this study is to develop an automated accurate 3D registration method, and to extensively validate this method on a large set of patient data. In addition, the authors quantified patient motion during scanning to investigate the need for correction. Methods: MR imaging studies (1.5T, dedicated carotid surface coil, Philips) from 55 TIA/stroke patients with ipsilateral <70% carotid artery stenosis were randomly selected from a larger cohort. Five MR pulse sequences were acquired around the carotid bifurcation, each containing nine transverse slices: T1-weighted turbo field echo, time of flight, T2-weighted turbo spin-echo, and pre- and postcontrast T1-weighted turbo spin-echo images (T1W TSE). The images were manually segmented by delineating the lumen contour in each vessel wall sequence and were manually aligned by applying throughplane and inplane translations to the images. To find the optimal automatic image registration method, different masks, choice of the fixed image, different types of the mutual information image similarity metric, and transformation models including 3D deformable transformation models, were evaluated. Evaluation of the automatic registration results was performed by comparing the lumen segmentations of the fixed image and moving image after registration. Results: The average required manual translation per image slice was 1.33 mm. Translations were larger as the patient was longer inside the scanner. Manual alignment took 187.5 s per patient resulting in a mean surface distance of 0.271 ± 0.127 mm. After minimal user interaction to generate the mask in the fixed image, the remaining sequences are automatically registered with a computation time of 52.0 s per patient. The optimal registration strategy used a circular mask with a diameter of 10 mm, a 3D B-spline transformation model with a control point spacing of 15 mm, mutual information as image similarity metric, and the precontrast T1W TSE as fixed image. A mean surface distance of 0.288 ± 0.128 mm was obtained with these settings, which is very close to the accuracy of the manual alignment procedure. The exact registration parameters and software were made publicly available. Conclusions: An automated registration method was developed and optimized, only needing two mouse clicks to mark the start and end point of the artery. Validation on a large group of patients showed that automated image registration has similar accuracy as the manual alignment procedure, substantially reducing the amount of user interactions needed, and is multiple times faster. In conclusion, the authors believe that the proposed automated method can replace the current manual procedure, thereby reducing the time to analyze the images.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

A machine learning pipeline for membrane segmentation of cryo-electron tomograms

We describe how to use several machine learning techniques organized in a learning pipeline to segment and identify cell membrane structures from cryo electron tomograms. These tomograms are difficult to analyze with traditional segmentation tools. The learning pipeline in our approach starts from supervised learning via a special convolutional neural network trained with simulated data. It continues with semi-supervised reinforcement learning and/or a region merging technique that tries to piece together disconnected components belonging to the same membrane structure. A parametric or non-parametric fitting procedure is then used to enhance the segmentation results and quantify uncertainties in the fitting. Domain knowledge is used in generating the training data for the neural network and in guiding the fitting procedure through the use of appropriately chosen priors and constraints. We demonstrate that the approach proposed here works well for extracting membrane surfaces in two real tomogram datasets.

97 MATHEMATICS AND COMPUTING↗

Model Assumptions and Data Characteristics: Impacts on Domain Adaptation in Building Segmentation

Studies on domain adaptation (DA) for remote sensing (RS) imagery analysis lack consistency in selection and description of evaluation scenarios. Without properly characterizing datasets, model assumptions, and evaluation scenarios, it is difficult to objectively compare DA methods and reach conclusions about their suitability across different applications. With this motivation, this work seeks to empirically assess to which extent the interaction between data characteristics and model assumptions influences the effectiveness of DA methods. Using the widely explored task of building footprint segmentation as a case study, we perform a large-scale study across over 200 DA scenarios that include variations across view angles, areas observed, and sensors used for data acquisition. Rather than adopting different model architectures or optimization criteria, we contrast the performances of two DA methods based on adversarial learning that differ only in their assumptions about source and target domains. Informed by metadata and data characteristics unveiled using traditional computer vision (CV) techniques as well as pretrained deep models, we provide a detailed meta-analysis of experiments highlighting the importance of accurately considering data assumptions for DA in RS segmentation tasks. As demonstrated by a “cherry-picking” exercise, different claims regarding which model is best could be made by selecting different subsets of evaluation scenarios. While well-calibrated assumptions can be beneficial, mismatching assumptions can lead to negative biases in DA applications. Furthermore, this study intends to motivate the community toward more consistent evaluation protocols while providing recommendations and insights toward creating novel benchmark datasets, documenting data characteristics, application-specific knowledge, and model assumptions.

42 ENGINEERING↗

Kilovoltage Intrafraction Monitoring for Prostate Intensity Modulated Arc Therapy: First Clinical Results

Purpose: Most linear accelerators purchased today are equipped with a gantry-mounted kilovoltage X-ray imager which is typically used for patient imaging prior to therapy. A novel application of the X-ray system is kilovoltage intrafraction monitoring (KIM), in which the 3-dimensional (3D) tumor position is determined during treatment. In this paper, we report on the first use of KIM in a prospective clinical study of prostate cancer patients undergoing intensity modulated arc therapy (IMAT). Methods and Materials: Ten prostate cancer patients with implanted fiducial markers undergoing conventionally fractionated IMAT (RapidArc) were enrolled in an ethics-approved study of KIM. KIM involves acquiring kV images as the gantry rotates around the patient during treatment. Post-treatment, markers in these images were segmented to obtain 2D positions. From the 2D positions, a maximum likelihood estimation of a probability density function was used to obtain 3D prostate trajectories. The trajectories were analyzed to determine the motion type and the percentage of time the prostate was displaced {>=}3, 5, 7, and 10 mm. Independent verification of KIM positional accuracy was performed using kV/MV triangulation. Results: KIM was performed for 268 fractions. Various prostate trajectories were observed (ie, continuous target drift, transient excursion, stable target position, persistent excursion, high-frequency excursions, and erratic behavior). For all patients, 3D displacements of {>=}3, 5, 7, and 10 mm were observed 5.6%, 2.2%, 0.7% and 0.4% of the time, respectively. The average systematic accuracy of KIM was measured at 0.46 mm. Conclusions: KIM for prostate IMAT was successfully implemented clinically for the first time. Key advantages of this method are (1) submillimeter accuracy, (2) widespread applicability, and (3) a low barrier to clinical implementation. A disadvantage is that KIM delivers additional imaging dose to the patient.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Machine learning enhanced characterization and optimization of photonic cured MAPbI 3 for efficient perovskite solar cells

Photonic curing (PC) can facilitate high-speed perovskite solar cell (PSC) manufacturing because it uses high-intensity light pulses to crystallize perovskite films in milliseconds. However, optimizing PC conditions is challenging due to its many variables, and using power conversion efficiency (PCE) as the optimization metric is both time-consuming and labor-intensive. This work presents a machine learning (ML) approach to optimize PC conditions for fabricating methylammonium lead iodide (MAPbI 3 ) films by quantitatively comparing their ultraviolet-visible (UV-vis) absorbance spectra to thermal annealed (TA) films using four similarity metrics. We perform Bayesian optimization coupled with Gaussian process regression (BO-GP) to minimize the similarity metrics. Refining PC conditions using active learning based on BO-GP models, we achieve a PC MAPbI3 film with an absorbance spectrum closely matching a TA reference film, which is further verified by its crystalline and morphological properties. Thus, we demonstrate that the UV-vis absorption spectrum can accurately proxy film quality. Additionally, we use an AI-based segmentation model for a more efficient grain size analysis. However, when we use the optimized PC condition to fabricate PSCs, we find that interaction between MAPbI 3 and the hole transport layer (HTL) during PC critically degrades the PSC performance. By adding a buffer layer between the HTL and MAPbI 3 , the optimized PC PSCs produce a champion PCE of 11.8%, comparable to the TA reference of 11.7%. Using UV-vis similarity metrics instead of device PCE as the objective in our BO-GP method accelerates the optimization of PC processing conditions for MAPbI 3 films.

14 SOLAR ENERGY↗

Bimodal Visualization of Industrial X-Ray and Neutron Computed Tomography Data

Advanced manufacturing creates increasingly complex objects with material compositions that are often difficult to characterize by a single modality. Our collaborating domain scientists are going beyond traditional methods by employing both X-ray and neutron computed tomography to obtain complementary representations expected to better resolve material boundaries. However, the use of two modalities creates its own challenges for visualization, requiring either complex adjustments of bimodal transfer functions or the need for multiple views. Together with experts in nondestructive evaluation, we designed a novel interactive bimodal visualization approach to create a combined view of the co-registered X-ray and neutron acquisitions of industrial objects. Using an automatic topological segmentation of the bivariate histogram of X-ray and neutron values as a starting point, the system provides a simple yet effective interface to easily create, explore, and adjust a bimodal visualization. Here, we propose a widget with simple brushing interactions that enables the user to quickly correct the segmented histogram results. Our semiautomated system enables domain experts to intuitively explore large bimodal datasets without the need for either advanced segmentation algorithms or knowledge of visualization techniques. We demonstrate our approach using synthetic examples, industrial phantom objects created to stress bimodal scanning techniques, and real-world objects, and we discuss expert feedback.

image segmentation↗

Superpixels for improved structure and terrain classification using multiple synthetic aperture radar image products

Various embodiments presented herein relate to assigning labels to segments of a synthetic aperture radar (SAR) image, where the segments are based upon a speckle-reduced SAR image product. A plurality of SAR images of a scene are co-registered to form a registered stack of SAR images. A speckle-reduced SAR image product is generated based upon at least one registered SAR image in the registered stack of SAR images. The speckle-reduced SAR image product is segmented into a plurality of superpixels, and boundaries of the superpixels are applied to the at least one registered SAR image to form a segmented SAR image. A segment of the SAR image is then labeled as including or not including a feature, wherein the label is assigned based upon values of pixels in the segment.

97 MATHEMATICS AND COMPUTING↗

Contour interpolated radial basis functions with spline boundary correction for fast 3D reconstruction of the human articular cartilage from MR images

Purpose: The aim of this work is to demonstrate a new image processing technique that can provide a “near real-time” 3D reconstruction of the articular cartilage of the human knee from MR images which is user friendly. This would serve as a point-of-care 3D visualization tool which would benefit a consultant radiologist in the visualization of the human articular cartilage. Methods: The authors introduce a novel fusion of an adaptation of the contour method known as “contour interpolation (CI)” with radial basis functions (RBFs) which they describe as “CI-RBFs.” The authors also present a spline boundary correction which further enhances volume estimation of the method. A subject cohort consisting of 17 right nonpathological knees (ten female and seven male) is assessed to validate the quality of the proposed method. The authors demonstrate how the CI-RBF method dramatically reduces the number of data points required for fitting an implicit surface to the entire cartilage, thus, significantly improving the speed of reconstruction over the comparable RBF reconstruction method of Carr. The authors compare the CI-RBF method volume estimation to a typical commercial package (3D DOCTOR), Carr’s RBF method, and a benchmark manual method for the reconstruction of the femoral, tibial, and patellar cartilages. Results: The authors demonstrate how the CI-RBF method significantly reduces the number of data points (p-value < 0.0001) required for fitting an implicit surface to the cartilage, by 48%, 31%, and 44% for the patellar, tibial, and femoral cartilages, respectively. Thus, significantly improving the speed of reconstruction (p-value < 0.0001) by 39%, 40%, and 44% for the patellar, tibial, and femoral cartilages over the comparable RBF model of Carr providing a near real-time reconstruction of 6.49, 8.88, and 9.43 min for the patellar, tibial, and femoral cartilages, respectively. In addition, it is demonstrated how the CI-RBF method matches the volume estimation of a typical commercial package (3D DOCTOR), Carr’s RBF method, and a benchmark manual method for the reconstruction of the femoral, tibial, and patellar cartilages. Furthermore, the performance of the segmentation method used for the extraction of the femoral, tibial, and patellar cartilages is assessed with a Dice similarity coefficient, sensitivity, and specificity measure providing high agreement to manual segmentation. Conclusions: The CI-RBF method provides a fast, accurate, and robust 3D model reconstruction that matches Carr’s RBF method, 3D DOCTOR, and a manual benchmark method in accuracy and significantly improves upon Carr’s RBF method in data requirement and computational speed. In addition, the visualization tool has been designed to quickly segment MR images requiring only four mouse clicks per MR image slice.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Development and testing of texture discriminators for the analysis of trabecular bone in proximal femur radiographs

Purpose: Texture analysis of femur radiographs may serve as a potential low cost technique to predict osteoporotic fracture risk and has received considerable attention in the past years. A further application of this technique may be the measurement of the quality of specific bone compartments to provide useful information for treatment of bone fractures. Two challenges of texture analysis are the selection of the best suitable texture measure and reproducible placement of regions of interest (ROIs). The goal of this in vitro study was to automatically place ROIs in radiographs of proximal femur specimens and to calculate correlations between various different texture analysis methods and the femurs' anchorage strength. Methods: Radiographs were obtained from 14 femoral specimens and bone mineral density (BMD) was measured in the femoral neck. Biomechanical testing was performed to assess the anchorage strength in terms of failure load, breakaway torque, and number of cycles. Images were segmented using a framework that is based on the usage of level sets and statistical in-shape models. Five ROIs were automatically placed in the head, upper and lower neck, trochanteric, and shaft compartment in an atlas subject. All other subjects were registered rigidly, affinely, and nonlinearly, and the resulting transformation was used to map the five ROIs onto the individual femora. Results: In each ROI, texture features were extracted using gray level co-occurence matrices (GLCM), third-order GLCM, morphological gradients (MGs), Minkowski dimensions (MDs), Minkowski functionals (MFs), Gaussian Markov random fields, and scaling index method (SIM). Coefficients of determination for each texture feature with parameters of anchorage strength were computed. In a stepwise multiregression analysis, the most predictive parameters were identified in different models. Texture features were highly correlated with anchorage strength estimated by the failure load of up to R{sup 2}=0.61 (MF and MG features, p<0.01) and were partially independent of BMD. The correlations were dependent on the choice of the ROI and the texture measure. The best predictive multiregression model for failure load R{sub adj}{sup 2}=0.86 (p<0.001) included a set of recently developed texture methods (MF and SIM) but excluded bone mineral density and commonly used texture measures. Conclusions: The results suggest that texture information contained in trabecular bone structure visualized on radiographs may predict whether an implant anchorage can be used and may determine the local bone quality from preoperative radiographs.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

TU-A-12A-10: Robust Radiomics Feature Quantification Using Semiautomatic Volumetric Segmentation

Purpose: Recent advances in medical imaging technologies provide opportunities to quantify the tumor phenotype throughout the course of treatment non-invasively. The emerging field of Radiomics addresses this by converting medical images into minable data by applying a large number of quantitative imaging algorithms. Accurate tumor segmentation is one of the main challenges of Radiomics. It has been shown that semiautomatic segmentation approaches efficiently reduce inter-observer variability as compared to the time consuming manual delineations. In this study, a semiautomatic volumetric segmentation algorithm, implemented in the free and publicly available 3D-Slicer platform, was investigated in terms of its robustness for Radiomics features quantification. Methods: Fifty-six 3D-Radiomics features, quantifying phenotypic differences based on the tumor intensity, shape and texture, were extracted from the computed tomography images of twenty lung cancer patients. These Radiomics features were derived from the 3D-tumor volumes defined by three independent observers twice using 3D-Slicer, and compared to manual slice-by-slice delineations of five independent physicians in terms of intra-class correlation coefficient (ICC) and feature range. Results: Radiomics features extracted from 3D-Slicer segmentations had significantly higher reproducibility (ICC= 0.85 0.15, p=0.0009) compared to the features extracted from the manual segmentations (ICC= 0.77 0.17). Furthermore, the features extracted from 3D-Slicer based segmentations, spread over significantly smaller range across observers as compared to those of the manual delineations (p= 3.819e-07). Moreover, the features derived from 3D-Slicer segmentations overlapped in range with those of the manual delineations, as the lower(higher) limit(s) being significantly higher(lower) for the 3D-Slicer features (p= 0.007, p= 5.863e-06). Conclusion: 3D-Slicer based semiautomatic segmentation significantly improves the robustness of radiomics features and thus could serve as a potential alternative to the time consuming manual segmentation process. So 3D-Slicer can have a large application in Radiomics, and be employed for high-throughput data mining research of medical imaging in clinical oncology.

60 APPLIED LIFE SCIENCES↗

L-PBF High-Throughput Data Pipeline Approach for Multi-modal Integration

Abstract Metal-based additive manufacturing requires active monitoring solutions for assessing part quality. Multiple sensors and data streams, however, generate large heterogeneous data sets that are impractical for manual assessment and characterization. In this work, an automated pipeline is developed that enables feature extraction from high-speed camera video and multi-modal data analysis. The framework removes the need for manual assessment through the utilization of deep learning techniques and training models in a weakly supervised paradigm. We demonstrate this pipeline’s capability over 700,000 high-speed camera frames. The pipeline successfully extracts melt pool and spatter geometries and links them to corresponding pyrometry, radiography, and processparameter information. 715 individual prints are examined to reveal melt pool areas that exceeds 0.07 mm 2 and pyrometry signal over a threshold (375 pyrometry units) were more likely to have defects. These automated processes enable massive throughput of characterization techniques.

36 MATERIALS SCIENCE↗

Unveiling the interaction of nanopatterned void superlattices with irradiation cascades

Nanopatterned microstructures in materials can have a profound impact on materials’ physical and chemical properties. While voids are typically considered as detrimental defects in irradiated materials, the patterning of nanoscale voids causes the formation of void superlattices and provides a highly efficient mechanism for gas storage. Despite the important applications of nanopatterned defect superlattices, how they degrade under irradiation remains unclear. Here we provide direct observation of the evolution of void superlattices under irradiation and elucidate the interaction of void superlattices with irradiation cascades. We reveal that the instability of void superlattices under irradiation is caused by heterogenous void shrinkage and demonstrate the imperative role of mixed 1D/3D diffusion of self-interstitial atoms and injected inert gas atoms on void shrinkage and void superlattice instability. Understanding the degradation mechanisms of nanopatterned microstructures is essential to designing damage-tolerant materials and broadening their applications in extreme environments.

36 MATERIALS SCIENCE↗

A machine learning pipeline for identifying infiltration managed aquifer recharge locations from satellite imagery in the San Joaquin Valley, California

This study focuses on an agricultural region in California’s Central Valley, USA, where Managed Aquifer Recharge (MAR) is widely implemented to mitigate groundwater depletion under increasing water demand and climate variability. A deep learning and machine learning framework was developed to identify infiltration-MAR locations using satellite imagery and environmental data. The framework integrates surface water detection from Sentinel-2 imagery, geospatial delineation of water bodies, spatiotemporal tracking of water body dynamics, and supervised classification using meteorological, environmental, and topographic variables. The framework was applied to a 2379 km² study area southwest of Fresno, where 765 water bodies were detected, including 139 identified MAR sites based on publicly available datasets and expert knowledge. The classification model achieved an accuracy of 0.94 and an F1 score of 0.85. Feature importance analysis indicates that cropland, normalized difference vegetation index (NDVI), and evaporation are among the most influential predictors for infiltration-MAR. Notably, the framework suggests that engineered water management in infiltration-MAR systems can disrupt or even reverse the expected positive correlation between surface water extent and precipitation. These findings provide physically interpretable insights into the characteristics of existing infiltration-MAR facilities and demonstrate the potential of the proposed framework as a reproducible, interpretable, and potentially transferable tool for data-driven infiltration-MAR identification and inventory development under growing climatic and hydrological uncertainty.

Classification↗