Paulino Cerqueira
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Paulino Cerqueira has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Jorge Ribeiro
Instituto Politécnico de Viana do Castelo
3 shared publications - João Ferreira
ADiT-LAB–Applied Digital Transformation Laboratory, Instituto Politécnico de Viana do Castelo
2 shared publications - André Costa
Instituto Politécnico de Viana do Castelo, Rua Escola Industrial e Comercial Nun’Álvares, 4900-347 Viana do Castelo, Portugal
1 shared publication - Luis Dias
Instituto Politécnico de Viana do Castelo, Rua Escola Industrial e Comercial Nun’Álvares, 4900-347 Viana do Castelo, Portugal
1 shared publication - Luis Pinto
Instituto Politécnico de Viana do Castelo, Rua Escola Industrial e Comercial Nun’Álvares, 4900-347 Viana do Castelo, Portugal
1 shared publication - Miguel Cruz
University of Minho
1 shared publication
Latest Publications
- 16 September 2025
- Published by Springer Nature
- p. 29-40
Food Waste Detection in Canteen Plates with Visual Large Language Models
- 16 September 2025
- Published by Springer Nature
- p. 29-40
Food Waste Detection in Canteen Plates with Visual Large Language Models
- Open Access
- 25 July 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (15) , 8273
A Contribution of Shortest Paths Algorithms to the NetworkX Python Library
NetworkX is a free Python library for graphs and networks and is used in many applications and projects to find the shortest path in path planning scenarios. For dense graphs, the library provides the Floyd–Warshall algorithm for shortest paths and the A* (“A-Star”) algorithm for shortest paths and path lengths. However, several extensions have been proposed to improve the A*, but they are not included in the library. In this context, this paper presents a set of implementations improving the A*, such as the IDA*, D* Lite, SMA*, Bidirectional A* and RTA*. The goal or challenge is to address the limitations of the A* in specific scenarios, such as searching for an optimal path repeatedly or when confronted with memory limitations, as exemplified by the NetworkX library. To do this, we first review the literature of the usage and general application of NetworkX in different domains of applicability and then explore their usage in a shortest path context. By reviewing and validating the usage of A* and extensions in Python using the NetworkX framework, the implementations were submitted to the network environment validation and passed the tests. We have also done the benchmarking of the A*, comparing it with the new ones, and concluded the better efficiency of the A* extensions in tri-objective scenario parameters (length, cost and toll). Despite the extensive utilisation of A* and its notable efficacy in identifying optimal paths, its performance is suboptimal in specific scenarios, such as when confronted with memory constraints and dynamic environments. Almost every algorithm outperformed or matched the A* in the fields that were developed to have an advantage, demonstrating the quality and robustness of the implemented algorithms. As a contribution and to foster further research in this shortest path specific context field, the dataset and Python code of the algorithms are available in a GitHub opensource repository.
- 25 July 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (15) , 8273
Open AccessA Contribution of Shortest Paths Algorithms to the NetworkX Python Library
NetworkX is a free Python library for graphs and networks and is used in many applications and projects to find the shortest path in path planning scenarios. For dense graphs, the library provides the Floyd–Warshall algorithm for shortest paths and the A* (“A-Star”) algorithm for shortest paths and path lengths. However, several extensions have been proposed to improve the A*, but they are not included in the library. In this context, this paper presents a set of implementations improving the A*, such as the IDA*, D* Lite, SMA*, Bidirectional A* and RTA*. The goal or challenge is to address the limitations of the A* in specific scenarios, such as searching for an optimal path repeatedly or when confronted with memory limitations, as exemplified by the NetworkX library. To do this, we first review the literature of the usage and general application of NetworkX in different domains of applicability and then explore their usage in a shortest path context. By reviewing and validating the usage of A* and extensions in Python using the NetworkX framework, the implementations were submitted to the network environment validation and passed the tests. We have also done the benchmarking of the A*, comparing it with the new ones, and concluded the better efficiency of the A* extensions in tri-objective scenario parameters (length, cost and toll). Despite the extensive utilisation of A* and its notable efficacy in identifying optimal paths, its performance is suboptimal in specific scenarios, such as when confronted with memory constraints and dynamic environments. Almost every algorithm outperformed or matched the A* in the fields that were developed to have an advantage, demonstrating the quality and robustness of the implemented algorithms. As a contribution and to foster further research in this shortest path specific context field, the dataset and Python code of the algorithms are available in a GitHub opensource repository.
- Open Access
- 25 June 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (13) , 7137
Food Waste Detection in Canteen Plates Using YOLOv11
This work presents a Computer Vision (CV) platform for Food Waste (FW) detection in canteen plates exploring a research gap in automated FW detection using CV models. A machine learning methodology was followed, starting with the creation of a custom dataset of canteen plates images before and after lunch or dinner, and data augmentation techniques were applied to enhance the model’s robustness. Subsequently, a CV model was developed using YOLOv11 to classify the percentage of FW on a plate, distinguishing between edible food items and non-edible discarded material. To evaluate the performance of the model, we used a real dataset as well as three benchmarking datasets with food plates, in which it could be detected waste. For the real dataset, the system achieved a mean average precision (mAP) of 0.343, a precision of 0.62, and a recall of 0.322 on the test set as well as demonstrating high accuracy in classifying waste considering the traditional evaluation metrics on the benchmarking datasets. Given these promising results and the provision of open-source code on a GitHub repository, the platform can be readily utilized by the research community and educational institutions to monitor FW in student meals and proactively implement reduction strategies.
- 25 June 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (13) , 7137
Open AccessFood Waste Detection in Canteen Plates Using YOLOv11
This work presents a Computer Vision (CV) platform for Food Waste (FW) detection in canteen plates exploring a research gap in automated FW detection using CV models. A machine learning methodology was followed, starting with the creation of a custom dataset of canteen plates images before and after lunch or dinner, and data augmentation techniques were applied to enhance the model’s robustness. Subsequently, a CV model was developed using YOLOv11 to classify the percentage of FW on a plate, distinguishing between edible food items and non-edible discarded material. To evaluate the performance of the model, we used a real dataset as well as three benchmarking datasets with food plates, in which it could be detected waste. For the real dataset, the system achieved a mean average precision (mAP) of 0.343, a precision of 0.62, and a recall of 0.322 on the test set as well as demonstrating high accuracy in classifying waste considering the traditional evaluation metrics on the benchmarking datasets. Given these promising results and the provision of open-source code on a GitHub repository, the platform can be readily utilized by the research community and educational institutions to monitor FW in student meals and proactively implement reduction strategies.