Convolutional Neural Network Regression Research Articles

Deep learning with plasma plume image sequences for anomaly detection and prediction of growth kinetics during pulsed laser deposition

Materials synthesis platforms that are designed for autonomous experimentation are capable of collecting multimodal diagnostic data that can be utilized for feedback to optimize material properties. Pulsed laser deposition (PLD) is emerging as a viable autonomous synthesis tool, and so the need arises to develop machine learning (ML) techniques that are capable of extracting information from in situ diagnostics. Here, we demonstrate that intensified-CCD image sequences of the plasma plume generated during PLD can be used for anomaly detection and the prediction of thin film growth kinetics. We develop multi-output (2 + 1)D convolutional neural network regression models that extract deep features from plume dynamics that not only correlate with the measured chamber pressure and incident laser energy, but more importantly, predict parameters of an auto-catalytic film growth model derived from in situ laser reflectivity experiments. Our results demonstrate how ML with in situ plume diagnostics data in PLD can be utilized to maintain deposition conditions in an optimal regime. Further, the predictive capabilities of plume dynamics on the kinetics of film growth or other film properties prior to deposition provides a means for rapid pre-screening of growth conditions for the non-expert, which promises to accelerate materials optimization with PLD.

Estimation of non-uniform motion blur using a patch-based regression convolutional neural network

The non-uniform blur of atmospheric turbulence can be modeled as a superposition of linear motion blur kernels at a patch level. We propose a regression convolutional neural network (CNN) to predict angle and length of a linear motion blur kernel for varying sized patches. We analyze the robustness of the network for different patch sizes and the performance of the network in regions where the characteristics of the blur are transitioning. Alternating patch sizes per epoch in training, we find coefficient of determination scores across a range of patch sizes of R2>0.78 for length and R2>0.94 for angle prediction. We find that blur predictions in regions overlapping two blur characteristics transition between the two characteristics as overlap changes. These results validate the use of such a network for prediction of non-uniform blur characteristics at a patch level.

Individual tree detection in large-scale urban environments using high-resolution multispectral imagery

Systematic maps of urban forests are useful for regional planners and ecologists to understand the spatial distribution of trees in cities. However, manually-created urban forest inventories are expensive and time-consuming to create and typically do not provide coverage of private land. Toward the goal of automating urban forest inventory through machine learning techniques, we performed a comparative study of methods for automatically detecting and localizing trees in multispectral aerial imagery of urban environments, and introduce a novel method based on convolutional neural network regression. Our evaluation is supported by a new dataset of over 1,500 images and almost 100,000 tree annotations, covering eight cities, six climate zones, and three image capture years. Our method outperforms previous methods, achieving 73.6% precision and 73.3% recall when trained and tested in Southern California, and 76.5% precision 72.0% recall when trained and tested across the entire state. To demonstrate the scalability of the technique, we produced the first map of trees across the entire urban forest of California. The map we produced provides important data for the planning and management of California’s urban forest, and establishes a proven methodology for potentially producing similar maps nationally and globally in the future.

Improved CNN-based model for shape parameter determination of sand particles considering imbalanced dataset

This study presents an artificial intelligence prediction model for determining the roundness and sphericity of sand particles using a convolutional neural network (CNN) regression model, which combines a CNN and a regression model. The dataset used for the model development consisted of 27,000 binary images of coarse grains with various shapes. To develop the prediction model, the validated algorithms, ResNet101 and ResNet152, were utilized and modified into a CNN regression model to predict the shape parameters. The performance of the developed model was evaluated using 3,000 randomly generated binary images, and the mean squared error and root mean squared error were calculated as evaluation metrics. The evaluation results indicated that the model effectively predicted the sphericity and roundness of sand particles, with a relatively low prediction performance observed for roundness.

Deep learned features selection algorithm: Removal operation of anomaly feature maps (RO-AFM)

Class/Regression Activation Maps (CAMs/RAMs; AMs) are often embedded into Convolutional Neural Networks (CNNs) for checking activated regions on input images at estimation. CNNs sometime generate unreliable AMs, such as activated regions, are inappropriate. Because AM is calculated by stacking many feature maps generated by the final convolutional layer, when there are Anomaly Feature Maps (AFMs), unreliable AMs can be generated. For example, suppose we have a CNN that evaluates the heart. In this case, the feature maps that focus on regions unrelated to the heart (e.g., shoulders and esophagus) are AFMs. Additionally, we have a hypothesis that the estimation accuracy of CNNs is increased by removing AFMs. However, methods for automatically detecting and removing AFMs have not been sufficiently studied in previous research to improve the performance of CNNs. Therefore, we propose a method named “Removal Operation of Anomaly Feature Maps (RO-AFMs)” to automatically detect and remove AFMs. When applying an RO-AFM to the Global Average Pooling (GAP) feature vectors of a CNN, dimensions of the GAP vector are reduced. Therefore, an RO-AFM is regarded as a deep-feature selection algorithm. From the results of adopting an RO-AFM to a Regression CNN (R-CNN) for estimating pulmonary artery wedge pressure, which is one of the measurement score for representing cardiac anomaly state, improved reliability of AM and estimation accuracy were verified. A comparison of RO-AFM and the existing methods, i.e., Lasso and the Feature Selection Layer (FSL), indicated that RO-AFM performed slightly better on the estimation accuracy. The computation time required for RO-AFM to evaluate all features was 1.833 s on average, confirming that RO-AFM is a lightweight process. Therefore, RO-AFM is useful for constructing a medical CNN that emphasizes explainability (e.g., CNNs for estimating the risk of a disease or a test value from chest X-ray or computed tomography images).

Physics-Informed Approximation of Internal Thermal History for Surface Deformation Predictions in Wire Arc Directed Energy Deposition

Abstract This work presents a physics-informed fusion methodology for deformation detection using multi-sensor thermal data. A challenge with additive manufacturing (AM) is that abnormalities commonly occur due to rapid changes in the thermal gradient. Different non-destructive in-situ thermal sensors capture parts of the thermal history but are limited by the visible temperature spectrum and sensor field of view of the fabrication process. Various sensors mitigate problems with the loss of thermal history information; however, it brings forth challenges with integrating different data streams and the need to interpolate the internal thermal history. This study develops a thermal data-informed heat flux methodology that fills the gap in fusing numerical temperature approximation with data-driven knowledge of the surface of additive manufactured (AM) components. First, this study fuses infrared (IR) thermal data complexities during the AM process with the Goldak double ellipsoidal heat flux to model the energy input into the component. Second, a thermal physics-informed model input (PIMI) is created with thermal data-informed heat flux to capture internal thermal history. Lastly, regression convolutional neural network (CNN) captures the relationship between the three-dimensional thermal gradient and surface deformation. The rapid thermal gradient formation and identification of deformation is a key step toward using thermal history data and machine learning to improve quality control in AM. The proposed surface deformation detection model achieved an MSE of 1.14 mm and an R2 of 0.89.

In-Season Cotton Yield Prediction with Scale-Aware Convolutional Neural Network Models and Unmanned Aerial Vehicle RGB Imagery.

In the pursuit of sustainable agriculture, efficient water management remains crucial, with growers relying on advanced techniques for informed decision-making. Cotton yield prediction, a critical aspect of agricultural planning, benefits from cutting-edge technologies. However, traditional methods often struggle to capture the nuanced complexities of crop health and growth. This study introduces a novel approach to cotton yield prediction, leveraging the synergy between Unmanned Aerial Vehicles (UAVs) and scale-aware convolutional neural networks (CNNs). The proposed model seeks to harness the spatiotemporal dynamics inherent in high-resolution UAV imagery to improve the accuracy of the cotton yield prediction. The CNN component adeptly extracts spatial features from UAV-derived imagery, capturing intricate details related to crop health and growth, modeling temporal dependencies, and facilitating the recognition of trends and patterns over time. Research experiments were carried out in a cotton field at the USDA-ARS Cropping Systems Research Laboratory (CSRL) in Lubbock, Texas, with three replications evaluating four irrigation treatments (rainfed, full irrigation, percent deficit of full irrigation, and time delay of full irrigation) on cotton yield. The prediction revealed that the proposed CNN regression models outperformed conventional CNN models, such as AlexNet, CNN-3D, CNN-LSTM, ResNet. The proposed CNN model showed state-of-art performance at different image scales, with the R2 exceeding 0.9. At the cotton row level, the mean absolute error (MAE) and mean absolute percentage error (MAPE) were 3.08 pounds per row and 7.76%, respectively. At the cotton grid level, the MAE and MAPE were 0.05 pounds and 10%, respectively. This shows the proposed model's adaptability to the dynamic interplay between spatial and temporal factors that affect cotton yield. The authors conclude that integrating UAV-derived imagery and CNN regression models is a potent strategy for advancing precision agriculture, providing growers with a powerful tool to optimize cultivation practices and enhance overall cotton productivity.

In-field and non-destructive determination of comprehensive maturity index and maturity stages of Camellia oleifera fruits using a portable hyperspectral imager

To efficiently detect the maturity stages of Camellia oleifera fruits, this study proposed a non-invasive method based on hyperspectral imaging technology. First, a portable hyperspectral imager was used for the in-field image acquisition of Camellia oleifera fruits at three maturity stages, and ten quality indexes were measured as reference standards. Then, factor analysis was performed to obtain the comprehensive maturity index (CMI) by analyzing the change trends and correlations of different indexes. To reduce the high dimensionality of spectral data, the successive projection algorithm (SPA) was employed to select effective feature wavelengths. The prediction models for CMI, including partial least squares regression (PLSR), support vector regression (SVR), extreme learning machine (ELM), and convolutional neural network regression (CNNR), were constructed based on full spectra and feature wavelengths; for CNNR, only the raw spectra were used as input. The SPA-CNNR model exhibited more promising performance (RP = 0.839, RMSEP = 0.261, and RPD = 1.849). Furthermore, PLS-DA models for maturity discrimination of Camellia oleifera fruits were developed using full wavelength, characteristic wavelengths and their fusion CMI, respectively. The PLS-DA model using the fused dataset achieved the highest maturity classification accuracy, with the best simplified model achieving 88.6 % accuracy in prediction set. This study indicated that a portable hyperspectral imager can be used for in-field determination of the internal quality and maturity stages of Camellia oleifera fruits. It provides strong support for non-destructive quality inspection and timely harvesting of Camellia oleifera fruits in the field.

Deep point cloud regression for above-ground forest biomass estimation from airborne LiDAR

Quantifying forest biomass stocks and their dynamics is important for implementing effective climate change mitigation measures by aiding local forest management, studying processes driving af-, re-, and deforestation, and improving the accuracy of carbon accounting. Owing to the 3-dimensional nature of forest structure, remote sensing using airborne LiDAR can be used to perform these measurements of vegetation structure at large scale. Harnessing the full dimensionality of the data, we present deep learning systems predicting wood volume and above ground biomass (AGB) directly from the full LiDAR point cloud and compare results to state-of-the-art approaches operating on basic statistics of the point clouds. For this purpose, we devise different neural network architectures for point cloud regression and evaluate them on remote sensing data of areas for which AGB estimates have been obtained from field measurements in the Danish national forest inventory. Our adaptation of Minkowski convolutional neural networks for regression give the best results. The deep neural networks produce significantly more accurate wood volume, AGB, and carbon stock estimates compared to state-of-the-art approaches. In contrast to other methods, the proposed deep learning approach does not require a digital terrain model and is robust to artifacts along the boundaries of the evaluated areas, which we demonstrate for the case where trees protrude into the area from the outside. We expect this finding to have a strong impact on LiDAR-based analyses of biomass dynamics.

Stochastic Computing Convolutional Neural Network Architecture Reinvented for Highly Efficient Artificial Intelligence Workload on Field-Programmable Gate Array.

Stochastic computing (SC) has a substantial amount of study on application-specific integrated circuit (ASIC) design for artificial intelligence (AI) edge computing, especially the convolutional neural network (CNN) algorithm. However, SC has little to no optimization on field-programmable gate array (FPGA). Scaling up the ASIC logic without FPGA-oriented designs is inefficient, while aggregating thousands of bitstreams is still challenging in the conventional SC. This research has reinvented several FPGA-efficient 8-bit SC CNN computing architectures, i.e., SC multiplexer multiply-accumulate, multiply-accumulate function generator, and binary rectified linear unit, and successfully scaled and implemented a fully parallel CNN model on Kintex7 FPGA. The proposed SC hardware only compromises 0.14% accuracy compared to binary computing on the handwriting Modified National Institute of Standards and Technology classification task and achieved at least 99.72% energy saving per image feedforward and 31× more data throughput than modern hardware. Unique to SC, early decision termination pushed the performance baseline exponentially with minimum accuracy loss, making SC CNN extremely lucrative for AI edge computing but limited to classification tasks. The SC's inherent noise heavily penalizes CNN regression performance, rendering SC unsuitable for regression tasks.

REPRODUCIBILITY OF BRAIN AGE SALIENCIES ACROSS DEEP NEURAL NETWORK ARCHITECTURES

Abstract The brain’s biological age (BA) reflects structural changes related to neuroanatomic senescence. BA is estimated from magnetic resonance images (MRIs) using convolutional neural networks (CNNs), which provide interpretability through saliency maps. By conveying the importance of brain regions to BA estimation, these maps reveal sex differences in brain-aging. We compared two CNN architectures to assess the reproducibility of such differences. Both regression CNNs had T1-weighted MRIs as inputs; outputs were estimated BAs. One CNN was trained on males and females separately, whereas the other was trained on both sexes. The dataset used consisted of T1-weighted MRIs from 5,851 cognitively normal individuals (3,142 females) aged 22 to 95 years, sampled from the Alzheimer’s Disease Neuroimaging Initiative (N = 510), Human Connectome Project Aging (N = 508) and Young Adult (N = 1,112), and UK Biobank (N = 3,721). For both models, compared to the average saliency for a null distribution, males’ BA estimation relied significantly (p < 0.05) more on the right lateral temporal lobe and superior frontal gyrus. Females’ BA estimation depended significantly (p <; 0.05) more on the right posterior and bilateral occipital regions, and medial parietal lobe. Simulated alterations in brain morphometry indicated saliencies correctly revealed regions with aging-related dynamics, which confirm prior findings on sex differences in brain-aging. Variations in CNN models’ saliencies did not affect overall anatomic patterns, suggesting that CNNs can capture brain-aging patterns robustly despite architecture differences. Novel statistical models for formal comparison of CNN saliencies should be developed to accommodate their nonlinear nature.

Urban Air-Quality Estimation Using Visual Cues and a Deep Convolutional Neural Network in Bengaluru (Bangalore), India.

Mobile monitoring provides robust measurements of air pollution. However, resource constraints often limit the number of measurements so that assessments cannot be obtained in all locations of interest. In response, surrogate measurement methodologies, such as videos and images, have been suggested. Previous studies of air pollution and images have used static images (e.g., satellite images or Google Street View images). The current study was designed to develop deep learning methodologies to infer on-road pollutant concentrations from videos acquired with dashboard cameras. Fifty hours of on-road measurements of four pollutants (black carbon, particle number concentration, PM2.5 mass concentration, carbon dioxide) in Bengaluru, India, were analyzed. The analysis of each video frame involved identifying objects and determining motion (by segmentation and optical flow). Based on these visual cues, a regression convolutional neural network (CNN) was used to deduce pollution concentrations. The findings showed that the CNN approach outperformed several other machine learning (ML) techniques and more conventional analyses (e.g., linear regression). The CO2 prediction model achieved a normalized root-mean-square error of 10-13.7% for the different train-validation division methods. The results here thus contribute to the literature by using video and the relative motion of on-screen objects rather than static images and by implementing a rapid-analysis approach enabling analysis of the video in real time. These methods can be applied to other mobile-monitoring campaigns since the only additional equipment they require is an inexpensive dashboard camera.

Deciphering the effect of grain boundary characteristics on fracture toughness of silicon nitride ceramics through a CNN regression model

Fracture toughness of silicon nitride (Si3N4) ceramics with different grain boundary characteristics was determined from their microstructural images via a convolutional neural network (CNN) model. For this purpose, the CNN model was trained using Si3N4 ceramics with sintering additives of Y2O3 and MgO, and then the fracture toughness for those with various lanthanoid oxides and MgO was determined from the microstructural images through the trained model. Herein, the difference between the determined and measured values indicates the effect of the grain boundary characteristics because they little appear in the microstructures. The measured and AI-determined fracture toughness showed that the large ionic radius of Sm2O3 enhanced the fracture toughness of the Si3N4 ceramics sintered at 1850 °C for 24 h. The trained model successfully determined fracture toughness values that closely matched the measured ones from the microstructural images, even for samples with pores and different grain boundary characteristics, except for the Sm2O3 sample sintered for 24 h.

Magnetocaloric properties predicted by deep learning with compositional features for bulk metallic glasses

A convolutional neural network (CNN) is a widely recognized deep-learning model that performs regression or classification from domain knowledge using a kernel function that extracts specific features of the input data. This study demonstrates the prediction performance of the magnetic entropy changes (−ΔSmag) and transition temperatures (Tr) of bulk metallic glasses (BMGs) with magnetocaloric effects using a CNN regression model. Compositional descriptors were generated using the “Magpie” method and were used as input data. The prediction performance was evaluated using the determination coefficient (R2 score) and root mean square error (RMSE), resulting in R2 score of 0.920 and 0.917 and RMSE of 1.285 J/kgK and 45.619 K for −ΔSmag and Tr, respectively. In the case study, we further applied the CNN model to unseen datasets for extrapolated −ΔSmag and Tr prediction and compared the results with the experimental values, resulting in excellent agreement with the reported experimental results.

Monitoring the growth of insect larvae using a regression convolutional neural network and knowledge transfer

Recently, there has been an increase in the popularity of breeding insect larvae (Tenebrio Molitor and Hermetia Illucens). Dimensioning larvae and observing their growth over time is a key component of monitoring insect larvae breeding. Due to the high number of larvae in the analysed images (dense scenes) and their overlap, determining the size distribution of larvae in real-time is a research challenge. In this work, we proposed an efficient method for determining the size distribution of larvae based on a regression convolutional neural network (RegCNN) and knowledge transfer. Larval width was chosen as the main measured larval parameter due to its ease of registration in dense scenes. The larval length L and its volume V were determined indirectly using determined regression models L(width) and V(width). RegCNN training was performed using knowledge transfer to omit the time-consuming labelling of multiple images containing larvae at different growth stages. Training used quartiles (lower quartile, median, upper quartile) of larval widths determined using improved multistage larvae phenotyping based on classical computer vision methods and larvae segmentation model. Finally, our approach required labelling only a few images for calibration purposes. The study evaluated different RegCNN architectures: pre-trained on ImageNet (ResNet, EfficientNet) and custom with a reduced number of model parameters. The proposed method was validated for the distribution of larvae characterised by width quartiles taking values from 1.7 mm to 3.1 mm, corresponding to an average larval length of 16 mm to 28 mm. For the best evaluated model (ResNet18) in larval width estimation, we obtained RMSE = 0.131 mm (average RMSE = 1.12 mm for larval length estimation) and R2=0.870 (coefficient of determination) with an average inference time of 0.30 s/box. The best proposed custom architecture (TenebrioRegCNN_v3) achieved slightly lower accuracy (RMSE = 0.134 mm, R2=0.864) with about five times lower inference time per image than ResNet18. The quantitative results confirmed the proposed method’s potential to be applied in real breeding conditions.

Phenotyping and predicting wheat spike characteristics using image analysis and machine learning

AbstractImprovements in trait phenotyping are needed to increase the quantity and quality of data available for genetic improvement of crops. In this study, we used moderate throughput image analysis and machine learning as a pipeline for phenotyping a key wheat spike characteristic: spikelet number per spike. A population of 594 soft red winter wheat inbred lines was evaluated in the field for 2 years and images of wheat spikes were taken and used to train deep‐learning algorithms to predict spikelet number. A total of 12,717 images were used to train, test, and validate a basic regression convolutional neural network (CNN), a visual geometry group application regression model, VGG16, the ResNet152V2 model, and the EfficientNetV2L model. The EfficientNetV2L model was the most accurate, having the lowest mean absolute error, second lowest root mean square error, and highest coefficient of determination (mean absolute error [MAE] = 0.60, root mean square error [RMSE] = 0.79, and R2 = 0.90). The ResNet152V2 model was slightly less accurate with a slightly better fit (MAE = 0.61,m RMSE = 0.78, and R2 = 0.87), followed by the basic CNN (MAE = 0.75, RMSE = 1.00, and R2 = 0.74) and finally by the VGG16 (MAE = 1.51, RMSE = 1.29, and R2 = 0.076). With an average error of just above one half of a spikelet, utilizing image analysis and machine learning counting methods could be used for multiple breeding applications, including direct selection of spikelet number, to provide data to identify quantitative trait loci, or for training whole genome selection models.

PCD reconstruction, object classification and pose estimation for underwater vehicles using orthogonal multibeam forward looking sonar fusion

Orthogonal Multibeam Sonar Fusion (OMSF) is a recent fusion based method capable of producing accurate underwater 3D Point Cloud Data (PCD) from Multibeam Forward Looking Sonars (MFLS), enabling accurate seabed mapping and object scanning. This article provides methodical testing of OMSF reconstruction for MFLS accounting for operating frequency effects and object shapes. The article then proposes novel perception based applications for OMSF, consisting of classification technique integrating OMSF PCDs with a PCD based Convolutional Neural Network (CNN), and pose estimation method combining Orthogonal Feature Matching (OFM) bounding box regression with a pose regression CNN. Reconstruction test results show that OMSF is more accurate and efficient using higher frequency MFLS and achieves up to 53% higher accuracy on solid surfaces compared to hollow frames. Application tests based on an underwater garage dock show classification using OMSF PCDs can achieve 25% and 37% higher success rate and confidence while being more efficient, compared to using raw 3D sonar data. OFM bounding box regression produces 4.28% higher mean Intersection over Union (IoU), and 10% increase in ¿25% IoU metric compared to methods based on more traditional MFLS filtering. Similarly, end-to-end pose estimation achieves 6.25% higher success rate with OFM bounding box samples compared to ones obtained using traditional MFLS filtering.

Study on the economic benefits of carbon-neutral digital platforms for sustainable development based on the GPT-QRCNN model

IntroductionThis article proposes a method for assessing the economic benefits of carbon-neutral digital platforms, which promote sustainable development by reducing carbon emissions through digital technology and data platforms.MethodsThe proposed method combines the GPT (Generative Pre-trained Transformer) and QRCNN (Quantile Regression Convolutional Neural Network) models. Firstly, the GPT model is utilized to extract features from platform data. Then, these features are combined with the QRCNN model for sequence modeling, enhancing prediction accuracy and generalization ability.ResultsThe method's effectiveness is demonstrated through experimental verification using actual platform data. The results highlight the practical significance and application value of the proposed method in evaluating the economic benefits of carbon-neutral digital platforms.DiscussionBy leveraging digital technology and data platforms, carbon-neutral digital platforms aim to reduce carbon emissions and promote sustainable development. The proposed method provides a means to accurately predict and analyze the economic benefits associated with these platforms. The combination of the GPT and QRCNN models enhances the accuracy and generalization ability of economic benefit predictions, enabling informed decision-making and policy formulation.

Prediction of soluble solids content using near-infrared spectra and optical properties of intact apple and pulp applying PLSR and CNN

Soluble solids content (SSC) is one of the most important internal quality attributes of fruit and could be predicted using near-infrared (NIR) spectra and optical properties. Partial least squares regression (PLSR) is a conventional regression method in SSC prediction. In recent years, deep learning methods represented by convolutional neural network (CNN) was suggested to be implied in spectral analysis. However, researchers are inevitably facing problems with regard to the selection of spectral pretreatment methods and the evaluation of the performance of the chosen regression. This study employed PLSR and CNN regression to predict SSC of apple based on the collected diffuse reflectance spectra of intact apple, total reflectance and total transmittance spectra of apple pulp, and the calculated optical property spectra, i.e., absorption coefficient and reduced scattering coefficient spectra of apple pulp. Five different spectral pretreatment methods were exerted on these spectra. Results showed that at a given regression (PLSR or CNN), the built models based on the diffuse reflectance spectra of intact apple had the best SSC prediction, and the built models based on pulp’s reduced scattering coefficient spectra had the poorest prediction performance. The best prediction performance was achieved by PLSR models using Savitzky-Golay with multiple scattering correction (Rp = 0.96, RMSEP = 0.54 %) and CNN regressions using Savitzky-Golay with standard normal variational transformation (Rp = 0.95, RMSEP = 0.59 %), respectively. Additionally, when the unknown original spectra were used for modeling, CNN had a better performance compared to PLSR, indicating the outstanding preponderance of CNN in spectral analysis. This study provides an effective reference for the selection of chemometric method based on NIR spectra.

Buckwheat Plant Height Estimation Based on Stereo Vision and a Regression Convolutional Neural Network under Field Conditions

Buckwheat plant height is an important indicator for producers. Due to the decline in agricultural labor, the automatic and real-time acquisition of crop growth information will become a prominent issue for farms in the future. To address this problem, we focused on stereo vision and a regression convolutional neural network (CNN) in order to estimate buckwheat plant height. MobileNet V3 Small, NasNet Mobile, RegNet Y002, EfficientNet V2 B0, MobileNet V3 Large, NasNet Large, RegNet Y008, and EfficientNet V2 L were modified into regression CNNs. Through a five-fold cross-validation of the modeling data, the modified RegNet Y008 was selected as the optimal estimation model. Based on the depth and contour information of buckwheat depth image, the mean absolute error (MAE), root mean square error (RMSE), mean square error (MSE), and mean relative error (MRE) when estimating plant height were 0.56 cm, 0.73 cm, 0.54 cm, and 1.7%, respectively. The coefficient of determination (R2) value between the estimated and measured results was 0.9994. Combined with the LabVIEW software development platform, this method can estimate buckwheat accurately, quickly, and automatically. This work contributes to the automatic management of farms.