Automated Dead Chicken Detection in Poultry Farms Using Knowledge Distillation and Vision Transformers DOI Creative Commons

Ridip Khanal,

Wenqin Wu, Joonwhoan Lee

et al.

Applied Sciences, Journal Year: 2024, Volume and Issue: 15(1), P. 136 - 136

Published: Dec. 27, 2024

Detecting dead chickens in broiler farms is critical for maintaining animal welfare and preventing disease outbreaks. This study presents an automated system that leverages CCTV footage to detect chickens, utilizing a two-step approach improve detection accuracy efficiency. First, stationary regions the footage—likely representing chickens—are identified. Then, deep learning classifier, enhanced through knowledge distillation, confirms whether detected object indeed chicken. EfficientNet-B0 employed as teacher model, while DeiT-Tiny functions student balancing high computational A dynamic frame selection strategy optimizes resource usage by adjusting monitoring intervals based on chickens’ age, ensuring real-time performance resource-constrained environments. method addresses key challenges such lack of explicit annotations along with common farm issues like lighting variations, occlusions, cluttered backgrounds, chicken growth, camera distortions. The experimental results demonstrate validation accuracies 99.3% model 98.7% significant reductions demands. system’s robustness scalability make it suitable large-scale deployment, minimizing need labor-intensive manual inspections. Future work will explore integrating methods incorporate temporal attention mechanisms removal processes.

Language: Английский

Machine and Deep Learning Methods for Concrete Strength Prediction: A Bibliometric and Content Analysis Review of Research Trends and Future Directions DOI
Raman Kumar, Essam Althaqafi, S. Gopal Krishna Patro

et al.

Applied Soft Computing, Journal Year: 2024, Volume and Issue: 164, P. 111956 - 111956

Published: July 8, 2024

Language: Английский

Citations

14
Prediction of Compressive Strength of Fly Ash-Recycled Mortar Based on Grey Wolf Optimizer–Backpropagation Neural Network DOI Open Access

Jingjing Shao,

Lin-Bin Li,

Guang-Ji Yin

et al.

Materials, Journal Year: 2025, Volume and Issue: 18(1), P. 139 - 139

Published: Jan. 1, 2025

The evaluation of the mechanical performance fly ash-recycled mortar (FARM) is a necessary condition to ensure efficient utilization recycled fine aggregates. This article describes design nine mix proportions FARMs with low water/cement ratio and screens six reasonable flowability. compressive strengths were tested, influence (w/c) age on strength was analyzed. Meanwhile, backpropagation neural network (BPNN) model optimized by grey wolf optimizer (GWO), namely GWO-BPNN model, established predict FARM. input layer consisted w/c, cement/sand ratio, water reducer, age, ash content, while output strength. data set 150 sets from this existing research in literature, which 70% used for training 30% validation. results show that compared traditional BPNN, coefficient determination (R2) increases 0.85 0.93, mean squared error (MSE) decreases 0.018 0.015. convergence iterations validation decrease 108 65. indicates GWO improved prediction accuracy computational efficiency BPNN. characteristic heat, kernel density estimation, scatter matrix, SHAP value all indicated w/c strongly negatively correlated strength, sand/cement positively However, relationship between contents ash, not obvious.

Language: Английский

Citations

0
Detection of high-risk diseases in poultry feces through transfer learning DOI
Abdulkadir Taşdelen,

Yenal Arslan

Engineering Science and Technology an International Journal, Journal Year: 2025, Volume and Issue: 64, P. 102002 - 102002

Published: Feb. 21, 2025

Language: Английский

Citations

0
Separating Chickens’ Heads and Legs in Thermal Images via Object Detection and Machine Learning Models to Predict Avian Influenza and Newcastle Disease DOI Creative Commons

Alireza Ansarimovahed,

Ahmad Banakar, Guoming Li

et al.

Animals, Journal Year: 2025, Volume and Issue: 15(8), P. 1114 - 1114

Published: April 11, 2025

Poultry body temperature is closely related to their metabolism and vital activities, which can indicate physiological status health. Therefore, monitoring these changes by analyzing thermal images help in the early accurate diagnosis of diseases using a non-destructive method. On other hand, it very important state part bird has greatest effect on disease. This not only speeds up process but also determines an index for animal pathologists. In this study, intelligent algorithm was presented with aim classification two diseases, Avian influenza Newcastle disease, hours disease transmission. For purpose, three different models were developed based images, including: original background removal, head legs chicken separated YOLO-v8 model. Then, features extracted from including texture color, evaluated all support vector machine (SVM) classifier. Also, most effective introduced researchers Relief feature selection algorithm. The results without chickens 75.89, 83.93, 92.48%, respectively, 83.04, 91.52, 94.20% respectively. model showed ability diagnose at 8 h after infection accuracy more than 90%. show that contribution texture-related greater poultry diseases. focusing feet areas will increase accuracy, allows real time stages

Language: Английский

Citations

0
A comprehensive review of machine learning for heart disease prediction: challenges, trends, ethical considerations, and future directions DOI Creative Commons
Raman Kumar, S. K. Garg, Rupinder Kaur

et al.

Frontiers in Artificial Intelligence, Journal Year: 2025, Volume and Issue: 8

Published: May 13, 2025

This review provides a thorough and organized overview of machine learning (ML) applications in predicting heart disease, covering technological advancements, challenges, future prospects. As cardiovascular diseases (CVDs) are the leading cause global mortality, there is an urgent demand for early precise diagnostic tools. ML models hold considerable potential by utilizing large-scale healthcare data to enhance predictive diagnostics. To systematically investigate this field, literature into five thematic categories such as “Heart Disease Detection Diagnostics,” “Machine Learning Models Algorithms Healthcare,” “Feature Engineering Optimization Techniques,” “Emerging Technologies “Applications AI Across Diseases Conditions.” The incorporates performance benchmarking various models, highlighting that hybrid deep (DL) frameworks, e.g., convolutional neural network-long short-term memory (CNN-LSTM) consistently outperform traditional terms sensitivity, specificity, area under curve (AUC). Several real-world case studies presented demonstrate successful deployment clinical wearable settings. showcases progression approaches from classifiers DL structures federated (FL) frameworks. It also discusses ethical issues, dataset limitations, model transparency. conclusions provide important insights development artificial intelligence (AI) powered, clinically applicable disease prediction systems.

Language: Английский

Citations

0
Enhancing Poultry Multi-Behavior Detection with Semi-Supervised Auto-Labeling and Prompt-Driven Zero-Shot Recognition DOI
Ramesh Bahadur Bist, Lilong Chai, Sachin Subedi

et al.

Published: Jan. 1, 2025

Language: Английский

Citations

0
Revolutionizing agricultural productivity with automated early leaf disease detection system for smart agriculture applications using IoT platform DOI

R. Karthickmanoj,

T. Sasilatha,

D. Lakshmi

et al.

Environment Development and Sustainability, Journal Year: 2024, Volume and Issue: unknown

Published: July 12, 2024

Language: Английский

Citations

1
The Efficiency in Controlling and Monitoring a Poultry Farm based on Internet of Things (IoT) DOI
Md Gapar Md Johar,

Fatria Jumara Adha,

Asif Iqbal Hajamydeen

et al.

Published: June 29, 2024

Language: Английский

Citations

0
Environmental impact of leachate pollution of ground water supplies DOI Creative Commons

A. Krishna Kumar Athithan,

K. Sivalingam,

Annapoorna

et al.

E3S Web of Conferences, Journal Year: 2024, Volume and Issue: 564, P. 11010 - 11010

Published: Jan. 1, 2024

The population living in cities needs a variety of urban services, such as solid waste management, sewage, and water supply. majority communities dispose their open dumps that are not properly lined, which has an impact on the land, water, air quality. Out fifteen largest states india, Tamil Nadu elevated rate urbanisation. is now state nation with greatest urbanization nearly forty-four percent according to 2011 Census. Nonetheless, influence leachate percolation was main focus this investigation. Samples were gathered from city’s environs disposal site. then divided age. It noted drinking had adverse effect health those close dumpsite. determined ground tainted unsuitable for residential usage, including drinking.

Language: Английский

Citations

0
Deep learning methods for poultry disease prediction using images DOI
George Hope Chidziwisano, Eric Samikwa,

Chisomo Daka

et al.

Computers and Electronics in Agriculture, Journal Year: 2024, Volume and Issue: 230, P. 109765 - 109765

Published: Dec. 24, 2024

Language: Английский

Citations

0