Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jun 27, 2026

Automated Analysis of C. elegans Fluorescence Images using SegElegans
06:27

Automated Analysis of C. elegans Fluorescence Images using SegElegans

Published on: October 10, 2025

Light Attention Encoder-Decoder for Cattle Body Segmentation and Body Weight Estimation.

Sahilpreet Singh Mann1, Halah K Shehada1, Sabrina T Amorim2

  • 1Bradley Department of Electrical and Computer Engineering, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061, USA.

Animals : an Open Access Journal From MDPI
|June 26, 2026
PubMed
Summary

Related Concept Videos

Light Acquisition02:16

Light Acquisition

In order to produce glucose, plants need to capture sufficient light energy. Many modern plants have evolved leaves specialized for light acquisition. Leaves can be only millimeters in width or tens of meters wide, depending on the environment. Due to competition for sunlight, evolution has driven the evolution of increasingly larger leaves and taller plants, to avoid shading by their neighbors with contaminant elaboration of root architecture and mechanisms to transport water and nutrients.

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Genomic prediction via reaction norm models for pregnancy loss using temperature humidity index in brahman cattle.

Translational animal science·2026
Same author

Impact of Trait Measurement Error on Quantitative Genetic Analysis of Computer Vision-Derived Traits.

Genes·2026
Same author

Artificial intelligence in animal breeding and genetics: applications, opportunities, and challenges.

Animal frontiers : the review magazine of animal agriculture·2026
Same author

Reciprocal BLUP: A Predictability-Guided Multi-Omics Framework for Plant Phenotype Prediction.

Plants (Basel, Switzerland)·2026
Same author

Using Structural Equation Models to Interpret Genome-Wide Association Studies for Morphological and Productive Traits in Soybean [<i>Glycine max</i> (L.) Merr.].

Plants (Basel, Switzerland)·2025
Same author

Genomic prediction of stalk lodging resistance and the associated intermediate phenotypes in maize using whole-genome resequence and multi-environmental data.

The plant genome·2025
Same journal

Correction: Gernhardt et al. Ex Vivo Computed Tomographic Morphometry and Motion of the Native and Fractured Equine Accessory Carpal Bone. <i>Animals</i> 2026, <i>16</i>, 1132.

Animals : an open access journal from MDPI·2026
Same journal

Camera-Trap Assessment of Terrestrial Mammals and Ground-Dwelling Birds in the Zhangjiajie Chinese Giant Salamander National Nature Reserve, China.

Animals : an open access journal from MDPI·2026
Same journal

Beyond the Mission: Long-Term Endocrine Dynamics in Search and Rescue Dog-Handler Teams.

Animals : an open access journal from MDPI·2026
Same journal

Phenotypic Characterisation of the Abruzzo Donkey (<i>Equus asinus</i>), an Endangered Italian Genetic Resource: Body Measurements.

Animals : an open access journal from MDPI·2026
Same journal

Assessment of Maternal Genetic Diversity and Mitochondrial Population Structure of Endangered Indigenous Chicken Breeds in China.

Animals : an open access journal from MDPI·2026
Same journal

Effects of Expected Progeny Difference and Feeding Systems on Carcass Characteristics in Hanwoo Steers.

Animals : an open access journal from MDPI·2026
See all related articles
This summary is machine-generated.

This study introduces a non-invasive method for estimating beef cattle body weight using overhead depth imaging and a Light Attention Encoder-Decoder (LAED) model. The system accurately segments cattle and predicts weight, reducing the need for manual handling and infrastructure.

Area of Science:

  • Agricultural Engineering
  • Computer Vision
  • Animal Science

Background:

  • Accurate body weight estimation is crucial for beef cattle management and performance monitoring.
  • Traditional methods like scales and manual measurements are labor-intensive and require specialized infrastructure.
  • Non-invasive techniques are needed to streamline cattle assessment and improve system efficiency.

Purpose of the Study:

  • To develop and evaluate an integrated pipeline for non-invasive cattle body segmentation and weight estimation using overhead depth imaging.
  • To assess the performance of a novel Light Attention Encoder-Decoder (LAED) model with Gaussian Context Transformer (GCT) attention for cattle segmentation.
  • To compare different machine learning models for body weight prediction based on extracted image features and biometric traits.
Keywords:
U-Netboundary-aware segmentationconvolutional networksdepth mapsgaussian context transformer

Related Experiment Videos

Last Updated: Jun 27, 2026

Automated Analysis of C. elegans Fluorescence Images using SegElegans
06:27

Automated Analysis of C. elegans Fluorescence Images using SegElegans

Published on: October 10, 2025

Main Methods:

  • Collected depth videos of 60 beef heifers using an Intel RealSense D435 RGB-D camera.
  • Employed a Light Attention Encoder-Decoder (LAED) model with Gaussian Context Transformer (GCT) attention for cattle segmentation, using leave-one-animal-out cross-validation.
  • Evaluated weight prediction using random forest, support vector regression (SVR), and fully connected neural networks (FCNNs) with biometric traits and deep features.

Main Results:

  • The LAED + GCT model achieved high segmentation accuracy with 96.91% Dice and 94.22% IoU, operating at 33.08 frames per second.
  • The best weight prediction model utilized biometric traits with SVR, yielding a Mean Absolute Percentage Error (MAPE) of 6.75%, pooled R² of 0.68, MAE of 23.92 kg, and RMSE of 31.79 kg.
  • FCNN models using ResNet50 features achieved a MAPE of 7.76% and a pooled R² of 0.56.

Conclusions:

  • Overhead depth imaging offers a promising non-invasive approach for cattle body segmentation and weight estimation.
  • The developed LAED model demonstrates high performance in segmenting cattle from depth images.
  • Further external validation is necessary to confirm the generalizability and robustness of the proposed system for real-world applications.