Jose Antonio Perez Garcia
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Jose Antonio Perez Garcia has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Jose Luis Martinez Lastra
FAST-Laboratory, Faculty of Engineering and Natural Sciences, Tampere University, Tampere, Finland
4 shared publications - Wael M. Mohammed
Tampere University
3 shared publications - Mahboob Elahi
Tampere University: Tampereen Yliopisto
2 shared publications - Pablo Malvido Fresnillo
FAST-Laboratory, Faculty of Engineering and Natural Sciences, Tampere University, Tampere, Finland
2 shared publications - Saigopal Vasudevan
FAST-Laboratory, Faculty of Engineering and Natural Sciences, Tampere University, Tampere, Finland
2 shared publications - Samuel Olaiya Afolaranmi
Faculty of Engineering and Natural Sciences, Tampere University, 33720 Tampere, Finland
2 shared publications
Latest Publications
- Open Access
- 27 October 2025
- Published by Springer Nature in Research Square
Multi-Objective Optimization of Container Loading: Balancing Space Utilization, Unloading Obstacles, and Cargo Stability
The Container Loading Problem (CLP) involves arranging rectangular boxes within a container while satisfying practical constraints such as dimensional fit, weight capacity, stackability, delivery order, customer priorities, and cargo stability. In real-world multi-drop delivery scenarios, additional considerations include minimizing unloading obstacles (ULOs) and maintaining cargo balance to ensure transport safety. This study formally defines a tri-objective CLP through a mathematical model that (i) maximizes space utilization, (ii) minimizes ULOs, and (iii) constrains the center of gravity (CG) of the loaded cargo within specified limits along each axis. The model integrates diverse practical constraints into a unified formulation. However, due to its complexity, it is used here for problem definition rather than being solved directly to optimality. A constraint-aware variant of the Non-dominated Sorting Genetic Algorithm II (NSGA-II) is developed to generate high-quality feasible solutions. Computational results on benchmark instances from Bischoff and Ratcliff, adapted to include CG balance requirements, demonstrate that the algorithm can efficiently capture trade-offs between volume utilization, unloading efficiency, and cargo balance. The approach addresses current limitations of CLP studies by jointly considering multiple operational objectives and stability constraints, providing realistic insight into how different priorities affect container space use.
- 27 October 2025
- Published by Springer Nature in Research Square
Open AccessMulti-Objective Optimization of Container Loading: Balancing Space Utilization, Unloading Obstacles, and Cargo Stability
The Container Loading Problem (CLP) involves arranging rectangular boxes within a container while satisfying practical constraints such as dimensional fit, weight capacity, stackability, delivery order, customer priorities, and cargo stability. In real-world multi-drop delivery scenarios, additional considerations include minimizing unloading obstacles (ULOs) and maintaining cargo balance to ensure transport safety. This study formally defines a tri-objective CLP through a mathematical model that (i) maximizes space utilization, (ii) minimizes ULOs, and (iii) constrains the center of gravity (CG) of the loaded cargo within specified limits along each axis. The model integrates diverse practical constraints into a unified formulation. However, due to its complexity, it is used here for problem definition rather than being solved directly to optimality. A constraint-aware variant of the Non-dominated Sorting Genetic Algorithm II (NSGA-II) is developed to generate high-quality feasible solutions. Computational results on benchmark instances from Bischoff and Ratcliff, adapted to include CG balance requirements, demonstrate that the algorithm can efficiently capture trade-offs between volume utilization, unloading efficiency, and cargo balance. The approach addresses current limitations of CLP studies by jointly considering multiple operational objectives and stability constraints, providing realistic insight into how different priorities affect container space use.
- Open Access
- 20 June 2024
- Published by Springer Nature in Machine Vision and Applications
- Vol. 35 (4) , 1-14
Generation of realistic synthetic cable images to train deep learning segmentation models
Semantic segmentation is one of the most important and studied problems in machine vision, which has been solved with high accuracy by many deep learning models. However, all these models present a significant drawback, they require large and diverse datasets to be trained. Gathering and annotating all these images manually would be extremely time-consuming, hence, numerous researchers have proposed approaches to facilitate or automate the process. Nevertheless, when the objects to be segmented are deformable, such as cables, the automation of this process becomes more challenging, as the dataset needs to represent their high diversity of shapes while keeping a high level of realism, and none of the existing solutions have been able to address it effectively. Therefore, this paper proposes a novel methodology to automatically generate highly realistic synthetic datasets of cables for training deep learning models in image segmentation tasks. This methodology utilizes Blender to create photo-realistic cable scenes and a Python pipeline to introduce random variations and natural deformations. To prove its performance, a dataset composed of 25000 synthetic cable images and their corresponding masks was generated and used to train six popular deep learning segmentation models. These models were then utilized to segment real cable images achieving outstanding results (over 70% IoU and 80% Dice coefficient for all the models). Both the methodology and the generated dataset are publicly available in the project’s repository.
- 20 June 2024
- Published by Springer Nature in Machine Vision and Applications
- Vol. 35 (4) , 1-14
Open AccessGeneration of realistic synthetic cable images to train deep learning segmentation models
Semantic segmentation is one of the most important and studied problems in machine vision, which has been solved with high accuracy by many deep learning models. However, all these models present a significant drawback, they require large and diverse datasets to be trained. Gathering and annotating all these images manually would be extremely time-consuming, hence, numerous researchers have proposed approaches to facilitate or automate the process. Nevertheless, when the objects to be segmented are deformable, such as cables, the automation of this process becomes more challenging, as the dataset needs to represent their high diversity of shapes while keeping a high level of realism, and none of the existing solutions have been able to address it effectively. Therefore, this paper proposes a novel methodology to automatically generate highly realistic synthetic datasets of cables for training deep learning models in image segmentation tasks. This methodology utilizes Blender to create photo-realistic cable scenes and a Python pipeline to introduce random variations and natural deformations. To prove its performance, a dataset composed of 25000 synthetic cable images and their corresponding masks was generated and used to train six popular deep learning segmentation models. These models were then utilized to segment real cable images achieving outstanding results (over 70% IoU and 80% Dice coefficient for all the models). Both the methodology and the generated dataset are publicly available in the project’s repository.
- Open Access
- 23 February 2024
- Published by Springer Nature in Research Square
Generation of realistic synthetic cable images to train Deep Learning segmentation models
One of the main contemporary challenges in robotics is the automation of tasks that involve handling Deformable Linear Objects (DLOs), such as cables or ropes. Due to their changing shape, constant perception is required, which makes computer vision, and in particular, image segmentation, fundamental. Even though image segmentation is a very studied problem, which has been addressed with high accuracy by many different Deep Learning models, they need to be properly trained to segment the objects of interest (DLOs in this case). To do this, it is necessary to have a large and diverse dataset. However, such a dataset hasn’t been created for cable images, and doing it manually would be complicated and extremely time-consuming. This paper addresses this issue with a novel methodology, which automatically generates synthetic cable datasets to train image segmentation models. This methodology utilizes Blender to create photo-realistic scenes and a Python pipeline to interact with them. To ensure the diversity of the dataset, before the generation of each image, the pipeline performs random variations on the elements of the scene. After this, the scene is animated, dropping the cables from a certain height and letting them deform naturally after landing, which results in realistic arrangements of the cables. The effectiveness of the methodology was demonstrated by training six popular segmentation models with synthetic datasets and using them to segment real cable images with great results (IoU over 70% and Dice coefficient over 80% for all the models).
- 23 February 2024
- Published by Springer Nature in Research Square
Open AccessGeneration of realistic synthetic cable images to train Deep Learning segmentation models
One of the main contemporary challenges in robotics is the automation of tasks that involve handling Deformable Linear Objects (DLOs), such as cables or ropes. Due to their changing shape, constant perception is required, which makes computer vision, and in particular, image segmentation, fundamental. Even though image segmentation is a very studied problem, which has been addressed with high accuracy by many different Deep Learning models, they need to be properly trained to segment the objects of interest (DLOs in this case). To do this, it is necessary to have a large and diverse dataset. However, such a dataset hasn’t been created for cable images, and doing it manually would be complicated and extremely time-consuming. This paper addresses this issue with a novel methodology, which automatically generates synthetic cable datasets to train image segmentation models. This methodology utilizes Blender to create photo-realistic scenes and a Python pipeline to interact with them. To ensure the diversity of the dataset, before the generation of each image, the pipeline performs random variations on the elements of the scene. After this, the scene is animated, dropping the cables from a certain height and letting them deform naturally after landing, which results in realistic arrangements of the cables. The effectiveness of the methodology was demonstrated by training six popular segmentation models with synthetic datasets and using them to segment real cable images with great results (IoU over 70% and Dice coefficient over 80% for all the models).
- Top Cited PapersOpen Access
- 7 December 2023
- Published by Springer Nature in Discover Artificial Intelligence
- Vol. 3 (1) , 1-78
A comprehensive literature review of the applications of AI techniques through the lifecycle of industrial equipment
Driven by the ongoing migration towards Industry 4.0, the increasing adoption of artificial intelligence (AI) has empowered smart manufacturing and digital transformation. AI enhances the migration towards industry 4.0 through AI-based decision-making by analyzing real-time data to optimize different processes such as production planning, predictive maintenance, quality control etc., thus guaranteeing reduced costs, high precision, efficiency and accuracy. This paper explores AI-driven smart manufacturing, revolutionizing traditional approaches and unlocking new possibilities throughout the major phases of the industrial equipment lifecycle. Through a comprehensive review, we delve into a wide range of AI techniques employed to tackle challenges such as optimizing process control, machining parameters, facilitating decision-making, and elevating maintenance strategies within the major phases of an industrial equipment lifecycle. These phases encompass design, manufacturing, maintenance, and recycling/retrofitting. As reported in the 2022 McKinsey Global Survey (https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai-in-2022-and-a-half-decade-in-review), the adoption of AI has witnessed more than a two-fold increase since 2017. This has contributed to an increase in AI research within the last six years. Therefore, from a meticulous search of relevant electronic databases, we carefully selected and synthesized 42 articles spanning from 01 January 2017 to 20 May 2023 to highlight and review the most recent research, adhering to specific inclusion and exclusion criteria, and shedding light on the latest trends and popular AI techniques adopted by researchers. This includes AI techniques such as Convolutional Neural Networks (CNN), Generative Adversarial Networks (GAN), Bayesian Networks, Support Vector Machines (SVM) etc., which are extensively discussed in this paper. Additionally, we provide insights into the advantages (e.g., enhanced decision making) and challenges (e.g., AI integration with legacy systems due to technical complexities and compatibilities) of integrating AI across the major stages of industrial equipment operations. Strategically implementing AI techniques in each phase enables industries to achieve enhanced productivity, improved product quality, cost-effectiveness, and sustainability. This exploration of the potential of AI in smart manufacturing fosters agile and resilient processes, keeping industries at the forefront of technological advancements and harnessing the full potential of AI-driven solutions to improve manufacturing processes and products.
- 7 December 2023
- Published by Springer Nature in Discover Artificial Intelligence
- Vol. 3 (1) , 1-78
Top Cited PapersOpen AccessA comprehensive literature review of the applications of AI techniques through the lifecycle of industrial equipment
Driven by the ongoing migration towards Industry 4.0, the increasing adoption of artificial intelligence (AI) has empowered smart manufacturing and digital transformation. AI enhances the migration towards industry 4.0 through AI-based decision-making by analyzing real-time data to optimize different processes such as production planning, predictive maintenance, quality control etc., thus guaranteeing reduced costs, high precision, efficiency and accuracy. This paper explores AI-driven smart manufacturing, revolutionizing traditional approaches and unlocking new possibilities throughout the major phases of the industrial equipment lifecycle. Through a comprehensive review, we delve into a wide range of AI techniques employed to tackle challenges such as optimizing process control, machining parameters, facilitating decision-making, and elevating maintenance strategies within the major phases of an industrial equipment lifecycle. These phases encompass design, manufacturing, maintenance, and recycling/retrofitting. As reported in the 2022 McKinsey Global Survey (https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai-in-2022-and-a-half-decade-in-review), the adoption of AI has witnessed more than a two-fold increase since 2017. This has contributed to an increase in AI research within the last six years. Therefore, from a meticulous search of relevant electronic databases, we carefully selected and synthesized 42 articles spanning from 01 January 2017 to 20 May 2023 to highlight and review the most recent research, adhering to specific inclusion and exclusion criteria, and shedding light on the latest trends and popular AI techniques adopted by researchers. This includes AI techniques such as Convolutional Neural Networks (CNN), Generative Adversarial Networks (GAN), Bayesian Networks, Support Vector Machines (SVM) etc., which are extensively discussed in this paper. Additionally, we provide insights into the advantages (e.g., enhanced decision making) and challenges (e.g., AI integration with legacy systems due to technical complexities and compatibilities) of integrating AI across the major stages of industrial equipment operations. Strategically implementing AI techniques in each phase enables industries to achieve enhanced productivity, improved product quality, cost-effectiveness, and sustainability. This exploration of the potential of AI in smart manufacturing fosters agile and resilient processes, keeping industries at the forefront of technological advancements and harnessing the full potential of AI-driven solutions to improve manufacturing processes and products.