Siavash Mahmoudi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Siavash Mahmoudi has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Chaitanya Pallerla
Department of Food Science, University of Arkansas, Fayetteville, AR 72701, USA
3 shared publications - Dongyi Wang
Department of Biological and Agricultural Engineering, University of Arkansas, Fayetteville, AR 72701, USA
3 shared publications - Philip Glen Crandall
Department of Food Science, University of Arkansas, Fayetteville, AR 72701, USA
3 shared publications - Pouya Sohrabipour
Department of Biological and Agricultural Engineering, University of Arkansas, Fayetteville, AR 72701, USA
3 shared publications - Ammirreza Davar
Department of Mechanical Engineering, University of Arkansas, Fayetteville, AR 72701, USA
2 shared publications - Wan Shou
Department of Mechanical Engineering, University of Arkansas , Fayetteville, Arkansas 72701
2 shared publications
Latest Publications
- Open Access
Evaluation of Robotic Swabbing and Fluorescent Sensing to Monitor the Hygiene of Food Contact Surfaces
Effective environmental monitoring is critical for preventing microbial and allergenic cross-contamination. However, manual swabbing methods, commonly used to verify hygienic conditions, are prone to inconsistent results because of variability in pressure, coverage, and techniques. Two novel solutions will be explored to address these challenges: a robotic swabbing system with tactile sensing control, and a fluorescence/absorbance spectrometer for non-contact, protein-based residue detection. The robotic system was evaluated against trained and untrained humans, measuring water pickup, surface coverage, and pressure consistency. Concurrently, the fluorescence system analyzed model poultry protein soil to correlate spectral patterns with contamination levels. The robotic system demonstrated statistically superior performance, achieving consistent force application and near-complete surface coverage, overcoming key limitations of manual sampling. The fluorescence system distinguished contamination with high sensitivity, validating its use as a rapid, non-contact assessment tool. Together, the robotic sample acquisition and the spectrometer’s sensitive analysis provide a dual-modality framework for enhancing hygiene monitoring in manufacturing facilities.
Open AccessEvaluation of Robotic Swabbing and Fluorescent Sensing to Monitor the Hygiene of Food Contact Surfaces
Effective environmental monitoring is critical for preventing microbial and allergenic cross-contamination. However, manual swabbing methods, commonly used to verify hygienic conditions, are prone to inconsistent results because of variability in pressure, coverage, and techniques. Two novel solutions will be explored to address these challenges: a robotic swabbing system with tactile sensing control, and a fluorescence/absorbance spectrometer for non-contact, protein-based residue detection. The robotic system was evaluated against trained and untrained humans, measuring water pickup, surface coverage, and pressure consistency. Concurrently, the fluorescence system analyzed model poultry protein soil to correlate spectral patterns with contamination levels. The robotic system demonstrated statistically superior performance, achieving consistent force application and near-complete surface coverage, overcoming key limitations of manual sampling. The fluorescence system distinguished contamination with high sensitivity, validating its use as a rapid, non-contact assessment tool. Together, the robotic sample acquisition and the spectrometer’s sensitive analysis provide a dual-modality framework for enhancing hygiene monitoring in manufacturing facilities.
- Open Access
- 12 March 2025
- Published by MDPI AG in Agriengineering
- Vol. 7 (3) , 77
Cost-Effective Active Laser Scanning System for Depth-Aware Deep-Learning-Based Instance Segmentation in Poultry Processing
The poultry industry plays a pivotal role in global agriculture, with poultry serving as a major source of protein and contributing significantly to economic growth. However, the sector faces challenges associated with labor-intensive tasks that are repetitive and physically demanding. Automation has emerged as a critical solution to enhance operational efficiency and improve working conditions. Specifically, robotic manipulation and handling of objects is becoming ubiquitous in factories. However, challenges exist to precisely identify and guide a robot to handle a pile of objects with similar textures and colors. This paper focuses on the development of a vision system for a robotic solution aimed at automating the chicken rehanging process, a fundamental yet physically strenuous activity in poultry processing. To address the limitation of the generic instance segmentation model in identifying overlapped objects, a cost-effective, dual-active laser scanning system was developed to generate precise depth data on objects. The well-registered depth data generated were integrated with the RGB images and sent to the instance segmentation model for individual chicken detection and identification. This enhanced approach significantly improved the model’s performance in handling complex scenarios involving overlapping chickens. Specifically, the integration of RGB-D data increased the model’s mean average precision (mAP) detection accuracy by 4.9% and significantly improved the center offset—a customized metric introduced in this study to quantify the distance between the ground truth mask center and the predicted mask center. Precise center detection is crucial for the development of future robotic control solutions, as it ensures accurate grasping during the chicken rehanging process. The center offset was reduced from 22.09 pixels (7.30 mm) to 8.09 pixels (2.65 mm), demonstrating the approach’s effectiveness in mitigating occlusion challenges and enhancing the reliability of the vision system.
- 12 March 2025
- Published by MDPI AG in Agriengineering
- Vol. 7 (3) , 77
Open AccessCost-Effective Active Laser Scanning System for Depth-Aware Deep-Learning-Based Instance Segmentation in Poultry Processing
The poultry industry plays a pivotal role in global agriculture, with poultry serving as a major source of protein and contributing significantly to economic growth. However, the sector faces challenges associated with labor-intensive tasks that are repetitive and physically demanding. Automation has emerged as a critical solution to enhance operational efficiency and improve working conditions. Specifically, robotic manipulation and handling of objects is becoming ubiquitous in factories. However, challenges exist to precisely identify and guide a robot to handle a pile of objects with similar textures and colors. This paper focuses on the development of a vision system for a robotic solution aimed at automating the chicken rehanging process, a fundamental yet physically strenuous activity in poultry processing. To address the limitation of the generic instance segmentation model in identifying overlapped objects, a cost-effective, dual-active laser scanning system was developed to generate precise depth data on objects. The well-registered depth data generated were integrated with the RGB images and sent to the instance segmentation model for individual chicken detection and identification. This enhanced approach significantly improved the model’s performance in handling complex scenarios involving overlapping chickens. Specifically, the integration of RGB-D data increased the model’s mean average precision (mAP) detection accuracy by 4.9% and significantly improved the center offset—a customized metric introduced in this study to quantify the distance between the ground truth mask center and the predicted mask center. Precise center detection is crucial for the development of future robotic control solutions, as it ensures accurate grasping during the chicken rehanging process. The center offset was reduced from 22.09 pixels (7.30 mm) to 8.09 pixels (2.65 mm), demonstrating the approach’s effectiveness in mitigating occlusion challenges and enhancing the reliability of the vision system.
- Open Access
- 26 January 2025
- Published by MDPI AG in Preprints.org
Cost Effective Active Laser Scanning System for Depth-Aware Deep-Learning-Based Instance Segmentation for Poultry Processing
The poultry industry plays a pivotal role in global agriculture, serving as a major source of protein and contributing significantly to economic growth. However, the sector faces challenges associated with labor-intensive tasks that are repetitive and physically demanding. Automation has emerged as a critical solution to enhance operational efficiency and improve working conditions. Specifically, robot manipulation and handling of objects is becoming ubiquitous in factories. However, challenges exist to precisely identify and guide the robot to handle a pile of objects with similar textures and colors. This paper focuses on the development of a vision system for a robotic solution aimed at automating the chicken rehanging process, a fundamental yet physically strenuous activity in poultry processing. To address the limitation of the generic instance segmentation model in identifying overlapped objects, a cost-effective dual-active laser scanning system was developed to generate precise depth data of objects. The generated well-registered depth data integrated with the RGB images were sent into the instance segmentation model for individual chicken detection and identification. This enhanced approach significantly improved the model’s performance in handling complex scenarios involving overlapping chickens. Specifically, the integration of RGB-D data increased the model’s mean Average Precision (mAP) detection accuracy by 4.9% and significantly improved the center offset—a customized metric introduced in this study to quantify the distance between the ground truth mask center and the predicted mask center. Precise center detection is crucial for the development of future robotic control solutions, as it ensures accurate grasping during the chicken rehanging process. The center offset was reduced from 22.09 pixels (7.30 mm) to 8.09 pixels (2.65 mm), demonstrating the approach’s effectiveness in mitigating occlusion challenges and enhancing the reliability of the vision system.
- 26 January 2025
- Published by MDPI AG in Preprints.org
Open AccessCost Effective Active Laser Scanning System for Depth-Aware Deep-Learning-Based Instance Segmentation for Poultry Processing
The poultry industry plays a pivotal role in global agriculture, serving as a major source of protein and contributing significantly to economic growth. However, the sector faces challenges associated with labor-intensive tasks that are repetitive and physically demanding. Automation has emerged as a critical solution to enhance operational efficiency and improve working conditions. Specifically, robot manipulation and handling of objects is becoming ubiquitous in factories. However, challenges exist to precisely identify and guide the robot to handle a pile of objects with similar textures and colors. This paper focuses on the development of a vision system for a robotic solution aimed at automating the chicken rehanging process, a fundamental yet physically strenuous activity in poultry processing. To address the limitation of the generic instance segmentation model in identifying overlapped objects, a cost-effective dual-active laser scanning system was developed to generate precise depth data of objects. The generated well-registered depth data integrated with the RGB images were sent into the instance segmentation model for individual chicken detection and identification. This enhanced approach significantly improved the model’s performance in handling complex scenarios involving overlapping chickens. Specifically, the integration of RGB-D data increased the model’s mean Average Precision (mAP) detection accuracy by 4.9% and significantly improved the center offset—a customized metric introduced in this study to quantify the distance between the ground truth mask center and the predicted mask center. Precise center detection is crucial for the development of future robotic control solutions, as it ensures accurate grasping during the chicken rehanging process. The center offset was reduced from 22.09 pixels (7.30 mm) to 8.09 pixels (2.65 mm), demonstrating the approach’s effectiveness in mitigating occlusion challenges and enhancing the reliability of the vision system.