Joshua Peeples
Research Subjects
Joshua Peeples has a total of 50 co-authors. Here are the top co-authors by number of shared publications:
- Alina Zare
Department of Electrical and Computer Engineering University of Florida Gainesville Florida USA
6 shared publications Weihuang Xu
3 shared publications- Weihuang Xu
Electrical and Computer Engineering University of Florida Gainesville FL USA
3 shared publications Alina Zare
2 shared publications- Andrew J. McFarland
Texas A&M AgriLife Research Texas A&M University College Station Texas USA
2 shared publications - Romain Gloaguen
INTERACT - Innovation, Territoire, Agriculture et Agro-industrie, Connaissance et Technologie (19, rue Pierre WAGUET – BP 30313 – 60026 Beauvais Cedex et 3, rue du Tronquet - CS40118 - 76130 Mont-Saint-Aignan - France)
2 shared publications
Latest Publications
- Open Access
- 18 October 2026
- Published by Wiley in The Plant Phenome Journal
- Vol. 9 (1)
A data‐driven image extraction and analysis pipeline for plant phenotyping in controlled environments
Advances in automation, imaging, and artificial intelligence have enabled large‐scale plant phenotyping, but image analysis remains a critical bottleneck for crop improvement and biological discovery. We developed an integrated multispectral phenotyping framework using imagery from the Texas A&M AgriLife Automated Precision Phenotyping Greenhouse and expanded plant growth and phenotyping (PGP v2) data across maize ( Zea mays L.), cotton ( Gossypium hirsutum ), rice ( Oryza sativa ), and sorghum ( Sorghum bicolor L.). The pipeline integrates pseudo‐red‐green‐blue composite image generation, plant detection and segmentation, image stitching, vegetation‐index analysis, texture analysis, morphological trait extraction, and temporal comparison of image‐derived features to quantify changes in plant structure, spectral reflectance, and texture over time. Among the evaluated segmentation approaches, segment anything model (SAM) v3 provided the highest and most consistent accuracy across diverse crop structures, although it required greater computational time than classical methods. Segment anything model for long sequences (SAM2Long)maintained plant‐instance associations across vertically stacked frames, while scale‐invariant feature transform–based stitching reconstructed plant mosaics when individual plants extended beyond a single field of view. For each plant and imaging date, the pipeline generated an 863‐dimensional feature vector spanning vegetation indices, spectral statistics, texture descriptors, and morphological traits. The framework was evaluated through two case studies: treatment‐level temporal analysis of mutagenized sorghum lines and cold‐stress phenotyping of maize using a separate imaging system. In both studies, the extracted features supported statistical and multivariate analyses of phenotypic variation and enabled separation of plants based on treatment‐ or stress‐related responses. The combined dataset and workflow provide structured, automated, and well‐documented phenotypic analysis across multiple crops, experimental settings, and imaging systems for controlled‐environment plant science and crop improvement.
- 18 October 2026
- Published by Wiley in The Plant Phenome Journal
- Vol. 9 (1)
Open AccessA data‐driven image extraction and analysis pipeline for plant phenotyping in controlled environments
Advances in automation, imaging, and artificial intelligence have enabled large‐scale plant phenotyping, but image analysis remains a critical bottleneck for crop improvement and biological discovery. We developed an integrated multispectral phenotyping framework using imagery from the Texas A&M AgriLife Automated Precision Phenotyping Greenhouse and expanded plant growth and phenotyping (PGP v2) data across maize ( Zea mays L.), cotton ( Gossypium hirsutum ), rice ( Oryza sativa ), and sorghum ( Sorghum bicolor L.). The pipeline integrates pseudo‐red‐green‐blue composite image generation, plant detection and segmentation, image stitching, vegetation‐index analysis, texture analysis, morphological trait extraction, and temporal comparison of image‐derived features to quantify changes in plant structure, spectral reflectance, and texture over time. Among the evaluated segmentation approaches, segment anything model (SAM) v3 provided the highest and most consistent accuracy across diverse crop structures, although it required greater computational time than classical methods. Segment anything model for long sequences (SAM2Long)maintained plant‐instance associations across vertically stacked frames, while scale‐invariant feature transform–based stitching reconstructed plant mosaics when individual plants extended beyond a single field of view. For each plant and imaging date, the pipeline generated an 863‐dimensional feature vector spanning vegetation indices, spectral statistics, texture descriptors, and morphological traits. The framework was evaluated through two case studies: treatment‐level temporal analysis of mutagenized sorghum lines and cold‐stress phenotyping of maize using a separate imaging system. In both studies, the extracted features supported statistical and multivariate analyses of phenotypic variation and enabled separation of plants based on treatment‐ or stress‐related responses. The combined dataset and workflow provide structured, automated, and well‐documented phenotypic analysis across multiple crops, experimental settings, and imaging systems for controlled‐environment plant science and crop improvement.
- Open Access
- 8 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 72746-72756
LGDWT-GS: Local and Global Discrete Wavelet-Regularized 3-D Gaussian Splatting for Sparse-View Scene Reconstruction
We propose a new method for few-shot 3D reconstruction that integrates global and local frequency regularization to stabilize geometry and preserve fine details under sparse-view conditions, addressing a key limitation of existing 3D Gaussian Splatting (3DGS) models.We also introduce a multispectral greenhouse dataset containing four spectral bands captured from diverse plant species under controlled conditions. Alongside the dataset, we release an open-source benchmarking package that defines standardized few-shot reconstruction protocols for evaluating 3DGS-based methods. Experiments on our multispectral dataset, as well as standard benchmarks, demonstrate that the proposed method achieves sharper, more stable, and spectrally consistent reconstructions than existing baselines. The dataset and code for this work are publicly available at https://github.com/Advanced-Vision-and-Learning-Lab/sparse-view-3dgs-pack.
- 8 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 72746-72756
Open AccessLGDWT-GS: Local and Global Discrete Wavelet-Regularized 3-D Gaussian Splatting for Sparse-View Scene Reconstruction
We propose a new method for few-shot 3D reconstruction that integrates global and local frequency regularization to stabilize geometry and preserve fine details under sparse-view conditions, addressing a key limitation of existing 3D Gaussian Splatting (3DGS) models.We also introduce a multispectral greenhouse dataset containing four spectral bands captured from diverse plant species under controlled conditions. Alongside the dataset, we release an open-source benchmarking package that defines standardized few-shot reconstruction protocols for evaluating 3DGS-based methods. Experiments on our multispectral dataset, as well as standard benchmarks, demonstrate that the proposed method achieves sharper, more stable, and spectrally consistent reconstructions than existing baselines. The dataset and code for this work are publicly available at https://github.com/Advanced-Vision-and-Learning-Lab/sparse-view-3dgs-pack.
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 145754-145771
Structural and Statistical Audio Texture Knowledge Distillation for Acoustic Classification
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 145754-145771
Open AccessStructural and Statistical Audio Texture Knowledge Distillation for Acoustic Classification
- Open Access
- 29 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Artificial Intelligence
- Vol. 7 (3) , 1261-1272
Histogram Layers for Neural “Engineered” Features
In the computer vision literature, many effective histogram-based features have been developed. These “engineered” features include local binary patterns and edge histogram descriptors among others and they have been shown to be informative features for a variety of computer vision tasks. In this paper, we explore whether these features can be learned through histogram layers embedded in a neural network and, therefore, be leveraged within deep learning frameworks. By using histogram features, local statistics of the feature maps from the convolution neural networks can be used to better represent the data. We present neural versions of local binary pattern and edge histogram descriptors that jointly improve feature representation for image classification. Experiments are presented on benchmark and real-world datasets. Our code is publicly available 11https://github.com/Advanced-Vision-and-Learning-Lab/NEHD NLBP.
- 29 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Artificial Intelligence
- Vol. 7 (3) , 1261-1272
Open AccessHistogram Layers for Neural “Engineered” Features
In the computer vision literature, many effective histogram-based features have been developed. These “engineered” features include local binary patterns and edge histogram descriptors among others and they have been shown to be informative features for a variety of computer vision tasks. In this paper, we explore whether these features can be learned through histogram layers embedded in a neural network and, therefore, be leveraged within deep learning frameworks. By using histogram features, local statistics of the feature maps from the convolution neural networks can be used to better represent the data. We present neural versions of local binary pattern and edge histogram descriptors that jointly improve feature representation for image classification. Experiments are presented on benchmark and real-world datasets. Our code is publicly available 11https://github.com/Advanced-Vision-and-Learning-Lab/NEHD NLBP.
- Open Access
Divergence Regulated Encoder Network for Joint Dimensionality Reduction and Classification
Feature representation is an important aspect of remote-sensing based image classification. While deep convolutional neural networks are able to effectively amalgamate information, large numbers of parameters often make learned features inscrutable and difficult to transfer to alternative models. In order to better represent statistical texture information for remote-sensing image classification, in this paper, we investigate performing joint dimensionality reduction and classification using a novel histogram neural network. Motivated by a popular dimensionality reduction approach, t-Distributed Stochastic Neighbor Embedding (t-SNE), our proposed method incorporates a classification loss computed on samples in a low-dimensional embedding space. We compare the learned sample embeddings against coordinates found by t-SNE in terms of classification accuracy and qualitative assessment. We also explore use of various divergence measures in the t-SNE objective. The proposed method has several advantages such as readily embedding out-of-sample points and reducing feature dimensionality while retaining class discriminability. Our results show that the proposed approach maintains and/or improves classification performance and reveals characteristics of features produced by neural networks that may be helpful for other applications.
Open AccessDivergence Regulated Encoder Network for Joint Dimensionality Reduction and Classification
Feature representation is an important aspect of remote-sensing based image classification. While deep convolutional neural networks are able to effectively amalgamate information, large numbers of parameters often make learned features inscrutable and difficult to transfer to alternative models. In order to better represent statistical texture information for remote-sensing image classification, in this paper, we investigate performing joint dimensionality reduction and classification using a novel histogram neural network. Motivated by a popular dimensionality reduction approach, t-Distributed Stochastic Neighbor Embedding (t-SNE), our proposed method incorporates a classification loss computed on samples in a low-dimensional embedding space. We compare the learned sample embeddings against coordinates found by t-SNE in terms of classification accuracy and qualitative assessment. We also explore use of various divergence measures in the t-SNE objective. The proposed method has several advantages such as readily embedding out-of-sample points and reducing feature dimensionality while retaining class discriminability. Our results show that the proposed approach maintains and/or improves classification performance and reveals characteristics of features produced by neural networks that may be helpful for other applications.
- 2 March 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 19 (1545598X) , 1-5
Divergence Regulated Encoder Network for Joint Dimensionality Reduction and Classification
Feature representation is an important aspect of remote-sensing-based image classification. While deep convolutional neural networks (DCNNs) are able to effectively amalgamate information, large numbers of parameters often make learned features inscrutable and difficult to transfer to alternative models. In order to better represent statistical texture information for remote-sensing image classification, in this letter, we investigate performing joint dimensionality reduction (DR) and classification using a novel histogram neural network. Motivated by a popular DR approach, t-distributed stochastic neighbor embedding (t-SNE), our proposed method incorporates a classification loss computed on samples in a low-dimensional embedding space. We compare the learned sample embeddings against coordinates found by t-SNE in terms of classification accuracy and qualitative assessment. We also explore the use of various divergence measures in the t-SNE objective. The proposed method has several advantages such as readily embedding out-of-sample points and reducing feature dimensionality while retaining class discriminability. Our results show that the proposed approach maintains and/or improves classification performance and reveals characteristics of features produced by neural networks that may be helpful for other applications.
- 2 March 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 19 (1545598X) , 1-5
Divergence Regulated Encoder Network for Joint Dimensionality Reduction and Classification
Feature representation is an important aspect of remote-sensing-based image classification. While deep convolutional neural networks (DCNNs) are able to effectively amalgamate information, large numbers of parameters often make learned features inscrutable and difficult to transfer to alternative models. In order to better represent statistical texture information for remote-sensing image classification, in this letter, we investigate performing joint dimensionality reduction (DR) and classification using a novel histogram neural network. Motivated by a popular DR approach, t-distributed stochastic neighbor embedding (t-SNE), our proposed method incorporates a classification loss computed on samples in a low-dimensional embedding space. We compare the learned sample embeddings against coordinates found by t-SNE in terms of classification accuracy and qualitative assessment. We also explore the use of various divergence measures in the t-SNE objective. The proposed method has several advantages such as readily embedding out-of-sample points and reducing feature dimensionality while retaining class discriminability. Our results show that the proposed approach maintains and/or improves classification performance and reveals characteristics of features produced by neural networks that may be helpful for other applications.
- 1 March 2022
- Published by Elsevier in Rhizosphere
- Vol. 21
The plasticity of early root development in Sesamum indicum L. as influenced by genotype, water, and nutrient availability
- 1 March 2022
- Published by Elsevier in Rhizosphere
- Vol. 21
The plasticity of early root development in Sesamum indicum L. as influenced by genotype, water, and nutrient availability
- 15 December 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Artificial Intelligence
- Vol. 3 (4) , 541-552
Histogram Layers for Texture Analysis
An essential aspect of texture analysis is the extraction of features that describe the distribution of values in local, spatial regions. We present a localized histogram layer for artificial neural networks. Instead of computing global histograms as done previously, the proposed histogram layer directly computes the local, spatial distribution of features for texture analysis, and parameters for the layer are estimated during backpropagation. We compare our method to state-of-the-art texture encoding methods such as: The deep encoding pooling network, deep texture encoding network, Fisher vector convolutional neural network, and multilevel texture encoding and representation. We used three material/texture datasets: 1) The describable texture dataset; 2) an extension of the ground terrain in outdoor scenes dataset; and 3) a subset of the materials in context dataset. Results indicate that the inclusion of the proposed histogram layer improves performance.
- 15 December 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Artificial Intelligence
- Vol. 3 (4) , 541-552
Histogram Layers for Texture Analysis
An essential aspect of texture analysis is the extraction of features that describe the distribution of values in local, spatial regions. We present a localized histogram layer for artificial neural networks. Instead of computing global histograms as done previously, the proposed histogram layer directly computes the local, spatial distribution of features for texture analysis, and parameters for the layer are estimated during backpropagation. We compare our method to state-of-the-art texture encoding methods such as: The deep encoding pooling network, deep texture encoding network, Fisher vector convolutional neural network, and multilevel texture encoding and representation. We used three material/texture datasets: 1) The describable texture dataset; 2) an extension of the ground terrain in outdoor scenes dataset; and 3) a subset of the materials in context dataset. Results indicate that the inclusion of the proposed histogram layer improves performance.
- Open Access
Learnable Adaptive Cosine Estimator (LACE) for Image Classification
In this work, we propose a new loss to improve feature discriminability and classification performance. Motivated by the adaptive cosine/coherence estimator (ACE), our proposed method incorporates angular information that is inherently learned by artificial neural networks. Our learnable ACE (LACE) transforms the data into a new "whitened" space that improves the inter-class separability and intra-class compactness. We compare our LACE to alternative state-of-the art softmax-based and feature regularization approaches. Our results show that the proposed method can serve as a viable alternative to cross entropy and angular softmax approaches. Our code is publicly available: https://github.com/GatorSense/LACE.
Open AccessLearnable Adaptive Cosine Estimator (LACE) for Image Classification
In this work, we propose a new loss to improve feature discriminability and classification performance. Motivated by the adaptive cosine/coherence estimator (ACE), our proposed method incorporates angular information that is inherently learned by artificial neural networks. Our learnable ACE (LACE) transforms the data into a new "whitened" space that improves the inter-class separability and intra-class compactness. We compare our LACE to alternative state-of-the art softmax-based and feature regularization approaches. Our results show that the proposed method can serve as a viable alternative to cross entropy and angular softmax approaches. Our code is publicly available: https://github.com/GatorSense/LACE.
- Open Access
Jointly Optimized Spatial Histogram UNET Architecture (JOSHUA) for Adipose Tissue Segmentation
Objective We quantify adipose tissue deposition at surgical sites as a function of biomaterial implantation. Impact Statement To our knowledge, this study is the first investigation to apply convolutional neural network (CNN) models to identify and segment adipose tissue in histological images from silk fibroin biomaterial implants. Introduction When designing biomaterials for the treatment of various soft tissue injuries and diseases, one must consider the extent of adipose tissue deposition. In this work, we implant silk fibroin biomaterials in a rodent subcutaneous injury model. Current strategies for quantifying adipose tissue after biomaterial implantation are often tedious and prone to human bias during analysis. Methods We used CNN models with novel spatial histogram layer(s) that can more accurately identify and segment regions of adipose tissue in hematoxylin and eosin (H&E) and Masson’s Trichrome stained images, allowing for determination of the optimal biomaterial formulation. We compared the method, Jointly Optimized Spatial Histogram UNET Architecture (JOSHUA), to the baseline UNET model and an extension of the baseline model, Attention UNET, as well as to versions of the models with a supplemental “attention”-inspired mechanism (JOSHUA+ and UNET+). Results The inclusion of histogram layer(s) in our models shows improved performance through qualitative and quantitative evaluation. Conclusion Our results demonstrate that the proposed methods, JOSHUA and JOSHUA+, are highly beneficial for adipose tissue identification and localization. The new histological dataset and code for our experiments are publicly available.
Open AccessJointly Optimized Spatial Histogram UNET Architecture (JOSHUA) for Adipose Tissue Segmentation
Objective We quantify adipose tissue deposition at surgical sites as a function of biomaterial implantation. Impact Statement To our knowledge, this study is the first investigation to apply convolutional neural network (CNN) models to identify and segment adipose tissue in histological images from silk fibroin biomaterial implants. Introduction When designing biomaterials for the treatment of various soft tissue injuries and diseases, one must consider the extent of adipose tissue deposition. In this work, we implant silk fibroin biomaterials in a rodent subcutaneous injury model. Current strategies for quantifying adipose tissue after biomaterial implantation are often tedious and prone to human bias during analysis. Methods We used CNN models with novel spatial histogram layer(s) that can more accurately identify and segment regions of adipose tissue in hematoxylin and eosin (H&E) and Masson’s Trichrome stained images, allowing for determination of the optimal biomaterial formulation. We compared the method, Jointly Optimized Spatial Histogram UNET Architecture (JOSHUA), to the baseline UNET model and an extension of the baseline model, Attention UNET, as well as to versions of the models with a supplemental “attention”-inspired mechanism (JOSHUA+ and UNET+). Results The inclusion of histogram layer(s) in our models shows improved performance through qualitative and quantitative evaluation. Conclusion Our results demonstrate that the proposed methods, JOSHUA and JOSHUA+, are highly beneficial for adipose tissue identification and localization. The new histological dataset and code for our experiments are publicly available.