Peng Lyu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Peng Lyu has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Jeongik Min
Graduate School of Artificial Intelligence, Jeonju University, Jeonju-si 55069, Republic of Korea
3 shared publications - Juwhan Song
Graduate School of Artificial Intelligence, Jeonju University, Jeonju-si 55069, Republic of Korea
2 shared publications - Jia Jia
Graduate School of Artificial Intelligence, Artificial Intelligence Research Center, Jeonju University, Jeonju-si 55069, Republic of Korea
1 shared publication - Sunwoo Ko
Graduate School of Artificial Intelligence, Jeonju University, Jeonju-si 55069, Republic of Korea
1 shared publication - Yumi Oh
Graduate School of Artificial Intelligence, Jeonju University, Jeonju-si 55069, Republic of Korea
1 shared publication - Zhibo Wang
School of Information Science and Technology, Zhengzhou Normal University, Zhengzhou, China
1 shared publication
Latest Publications
- Open Access
- 23 May 2024
- Published by MDPI AG in Agriculture
- Vol. 14 (6) , 809
Enhancing Broiler Weight Estimation through Gaussian Kernel Density Estimation Modeling
The management of individual weights in broiler farming is not only crucial for increasing farm income but also directly linked to the revenue growth of integrated broiler companies, necessitating prompt resolution. This paper proposes a model to estimate daily average broiler weights using time and weight data collected through scales. In the proposed model, a method of self-adjusting weights in the bandwidth calculation formula is employed, and the daily average weight representative value is estimated using KDE. The focus of this study is to contribute to the individual weight management of broilers by intensively researching daily fluctuations in average broiler weight. To address this, weight and time data are collected and preprocessed through scales. The Gaussian kernel density estimation model proposed in this paper aims to estimate the representative value of the daily average weight of a single broiler using statistical estimation methods, allowing for self-adjustment of bandwidth values. When applied to the dataset collected through scales, the proposed Gaussian kernel density estimation model with self-adjustable bandwidth values confirmed that the estimated daily weight did not deviate beyond the error range of ±50 g compared with the actual measured values. The next step of this study is to systematically understand the impact of the broiler environment on weight for sustainable management strategies for broiler demand, derive optimal rearing conditions for each farm by combining location and weight data, and develop a model for predicting daily average weight values. The ultimate goal is to develop an artificial intelligence model suitable for weight management systems by utilizing the estimated daily average weight of a single broiler even in the presence of error data collected from multiple weight measurements, enabling more efficient automatic measurement of broiler weight and supporting both farms and broiler demand.
- 23 May 2024
- Published by MDPI AG in Agriculture
- Vol. 14 (6) , 809
Open AccessEnhancing Broiler Weight Estimation through Gaussian Kernel Density Estimation Modeling
The management of individual weights in broiler farming is not only crucial for increasing farm income but also directly linked to the revenue growth of integrated broiler companies, necessitating prompt resolution. This paper proposes a model to estimate daily average broiler weights using time and weight data collected through scales. In the proposed model, a method of self-adjusting weights in the bandwidth calculation formula is employed, and the daily average weight representative value is estimated using KDE. The focus of this study is to contribute to the individual weight management of broilers by intensively researching daily fluctuations in average broiler weight. To address this, weight and time data are collected and preprocessed through scales. The Gaussian kernel density estimation model proposed in this paper aims to estimate the representative value of the daily average weight of a single broiler using statistical estimation methods, allowing for self-adjustment of bandwidth values. When applied to the dataset collected through scales, the proposed Gaussian kernel density estimation model with self-adjustable bandwidth values confirmed that the estimated daily weight did not deviate beyond the error range of ±50 g compared with the actual measured values. The next step of this study is to systematically understand the impact of the broiler environment on weight for sustainable management strategies for broiler demand, derive optimal rearing conditions for each farm by combining location and weight data, and develop a model for predicting daily average weight values. The ultimate goal is to develop an artificial intelligence model suitable for weight management systems by utilizing the estimated daily average weight of a single broiler even in the presence of error data collected from multiple weight measurements, enabling more efficient automatic measurement of broiler weight and supporting both farms and broiler demand.
- Open Access
- 23 November 2023
- Published by MDPI AG in Agriculture
- Vol. 13 (12) , 2193
Application of Machine Learning Algorithms for On-Farm Monitoring and Prediction of Broilers’ Live Weight: A Quantitative Study Based on Body Weight Data
A non-invasive automatic broiler weight estimation and prediction method based on a machine learning algorithm was developed to address the issue of high labor costs and stress responses caused by the traditional broiler weighing method in large-scale broiler production. Machine learning algorithms are a data-driven strategy that enables computer systems to make predictions and judgments based on patterns and regularities that they have learned. To estimate the current weight of individual live broilers on farms, machine learning algorithms such as the Gaussian mixture model, Isolation Forest, and Ordering Points To Identify the Clustering Structure (OPTICS) are used to filter and extract data features using a two-stage clustering and noise reduction process. Real-time weight prediction was also achieved by combining polynomial fitting and the gray models and adjusting the model parameters based on prediction accuracy feedback. The symmetric mean absolute percentage error (SMAPE) value is a metric that is commonly used to evaluate the predictive performance of a model by comparing the degree of error between the model’s predicted value on the day of slaughter and the true value measured manually, and the results of the experiments on 111 datasets showed that 7.21% were less than or equal to 0.03, 28.83% were less than or equal to 0.1 and greater than 0.03, and 31.53% were less than or equal to 0.2 and greater than 0.1. This method can be used as a prediction scheme for broiler weight monitoring in a large-scale rearing environment, considering the cost of implementation and the accuracy of estimation.
- 23 November 2023
- Published by MDPI AG in Agriculture
- Vol. 13 (12) , 2193
Open AccessApplication of Machine Learning Algorithms for On-Farm Monitoring and Prediction of Broilers’ Live Weight: A Quantitative Study Based on Body Weight Data
A non-invasive automatic broiler weight estimation and prediction method based on a machine learning algorithm was developed to address the issue of high labor costs and stress responses caused by the traditional broiler weighing method in large-scale broiler production. Machine learning algorithms are a data-driven strategy that enables computer systems to make predictions and judgments based on patterns and regularities that they have learned. To estimate the current weight of individual live broilers on farms, machine learning algorithms such as the Gaussian mixture model, Isolation Forest, and Ordering Points To Identify the Clustering Structure (OPTICS) are used to filter and extract data features using a two-stage clustering and noise reduction process. Real-time weight prediction was also achieved by combining polynomial fitting and the gray models and adjusting the model parameters based on prediction accuracy feedback. The symmetric mean absolute percentage error (SMAPE) value is a metric that is commonly used to evaluate the predictive performance of a model by comparing the degree of error between the model’s predicted value on the day of slaughter and the true value measured manually, and the results of the experiments on 111 datasets showed that 7.21% were less than or equal to 0.03, 28.83% were less than or equal to 0.1 and greater than 0.03, and 31.53% were less than or equal to 0.2 and greater than 0.1. This method can be used as a prediction scheme for broiler weight monitoring in a large-scale rearing environment, considering the cost of implementation and the accuracy of estimation.
- Open Access
- 11 September 2023
- Published by MDPI AG in Journal of Imaging
- Vol. 9 (9) , 183
Efficient Dehazing with Recursive Gated Convolution in U-Net: A Novel Approach for Image Dehazing
Image dehazing, a fundamental problem in computer vision, involves the recovery of clear visual cues from images marred by haze. Over recent years, deploying deep learning paradigms has spurred significant strides in image dehazing tasks. However, many dehazing networks aim to enhance performance by adopting intricate network architectures, complicating training, inference, and deployment procedures. This study proposes an end-to-end U-Net dehazing network model with recursive gated convolution and attention mechanisms to improve performance while maintaining a lean network structure. In our approach, we leverage an improved recursive gated convolution mechanism to substitute the original U-Net’s convolution blocks with residual blocks and apply the SK fusion module to revamp the skip connection method. We designate this novel U-Net variant as the Dehaze Recursive Gated U-Net (DRGNet). Comprehensive testing across public datasets demonstrates the DRGNet’s superior performance in dehazing quality, detail retrieval, and objective evaluation metrics. Ablation studies further confirm the effectiveness of the key design elements.
- 11 September 2023
- Published by MDPI AG in Journal of Imaging
- Vol. 9 (9) , 183
Open AccessEfficient Dehazing with Recursive Gated Convolution in U-Net: A Novel Approach for Image Dehazing
Image dehazing, a fundamental problem in computer vision, involves the recovery of clear visual cues from images marred by haze. Over recent years, deploying deep learning paradigms has spurred significant strides in image dehazing tasks. However, many dehazing networks aim to enhance performance by adopting intricate network architectures, complicating training, inference, and deployment procedures. This study proposes an end-to-end U-Net dehazing network model with recursive gated convolution and attention mechanisms to improve performance while maintaining a lean network structure. In our approach, we leverage an improved recursive gated convolution mechanism to substitute the original U-Net’s convolution blocks with residual blocks and apply the SK fusion module to revamp the skip connection method. We designate this novel U-Net variant as the Dehaze Recursive Gated U-Net (DRGNet). Comprehensive testing across public datasets demonstrates the DRGNet’s superior performance in dehazing quality, detail retrieval, and objective evaluation metrics. Ablation studies further confirm the effectiveness of the key design elements.