Leanne Miller
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Leanne Miller has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
- Pedro Javier Navarro
División de Sistemas e Ingeniería Electrónica (DSIE), Campus Muralla del Mar, s/n, Universidad Politécnica de Cartagena, 30202 Cartagena, Spain
9 shared publications - Francisca Rosique
División de Sistemas e Ingeniería Electrónica (DSIE), Campus Muralla del Mar, s/n, Universidad Politécnica de Cartagena, 30202 Cartagena, Spain
5 shared publications - Alberto Gila-Navarro
Genética Molecular, Instituto de Biotecnología Vegetal, Edificio I+D+I, Plaza del Hospital s/n, Universidad Politécnica de Cartagena, 30202 Cartagena, Spain
3 shared publications - Eduardo Salas
División de Sistemas e Ingeniería Electrónica (DSIE), Campus Muralla del Mar, s/n, Universidad Politécnica de Cartagena, 30202 Cartagena, Spain
2 shared publications - Marcos Egea-Cortines
Instituto de Biotecnología Vegetal, Universidad Politécnica de Cartagena, 30202 Cartagena, Spain
2 shared publications - María Victoria Díaz-Galián
Genética Molecular, Instituto de Biotecnología Vegetal, Edificio I+D+I, Plaza del Hospital s/n, Universidad Politécnica de Cartagena, 30202 Cartagena, Spain
2 shared publications
Latest Publications
- Open Access
EdgeNet: An End-to-End Deep Neural Network Pretrained with Synthetic Data for a Real-World Autonomous Driving Application
This paper presents a novel end-to-end architecture based on edge detection for autonomous driving. The architecture has been designed to bridge the domain gap between synthetic and real-world images for end-to-end autonomous driving applications and includes custom edge detection layers before the Efficient Net convolutional module. To train the architecture, RGB and depth images were used together with inertial data as inputs to predict the driving speed and steering wheel angle. To pretrain the architecture, a synthetic multimodal dataset for autonomous driving applications was created. The dataset includes driving data from 100 diverse weather and traffic scenarios, gathered from multiple sensors including cameras and an IMU as well as from vehicle control variables. The results show that including edge detection layers in the architecture improves performance for transfer learning when using synthetic and real-world data. In addition, pretraining with synthetic data reduces training time and enhances model performance when using real-world data.
Open AccessEdgeNet: An End-to-End Deep Neural Network Pretrained with Synthetic Data for a Real-World Autonomous Driving Application
This paper presents a novel end-to-end architecture based on edge detection for autonomous driving. The architecture has been designed to bridge the domain gap between synthetic and real-world images for end-to-end autonomous driving applications and includes custom edge detection layers before the Efficient Net convolutional module. To train the architecture, RGB and depth images were used together with inertial data as inputs to predict the driving speed and steering wheel angle. To pretrain the architecture, a synthetic multimodal dataset for autonomous driving applications was created. The dataset includes driving data from 100 diverse weather and traffic scenarios, gathered from multiple sensors including cameras and an IMU as well as from vehicle control variables. The results show that including edge detection layers in the architecture improves performance for transfer learning when using synthetic and real-world data. In addition, pretraining with synthetic data reduces training time and enhances model performance when using real-world data.
- Open Access
- 14 November 2024
- Published by MDPI AG in Preprints.org
EdgeNet: An End-to-End Deep Neural Network pretrained with Synthetic Data for a Real-World Autonomous Driving Application
This study presents a novel end-to-end architecture based on edge detection for autonomous driving. The architecture has been designed to bridge the domain gap between synthetic and real-world images for end-to-end autonomous driving applications and includes custom edge detection layers before the Efficient Net convolutional module. To train the architecture, RGB and depth images were used together with inertial data as inputs to predict the driving speed and steering wheel angle. To pretrain the architecture, a synthetic multimodal dataset for autonomous driving applications was created. The dataset includes driving data from 100 diverse weather and traffic scenarios, gathered from multiple sensors including cameras and an IMU as well as vehicle control variables. The results show that including edge detection layers in the architecture improves performance for transfer learning when using synthetic and real-world data. In addition, pretraining with synthetic data reduces training time and enhances model performance when using real-world data.
- 14 November 2024
- Published by MDPI AG in Preprints.org
Open AccessEdgeNet: An End-to-End Deep Neural Network pretrained with Synthetic Data for a Real-World Autonomous Driving Application
This study presents a novel end-to-end architecture based on edge detection for autonomous driving. The architecture has been designed to bridge the domain gap between synthetic and real-world images for end-to-end autonomous driving applications and includes custom edge detection layers before the Efficient Net convolutional module. To train the architecture, RGB and depth images were used together with inertial data as inputs to predict the driving speed and steering wheel angle. To pretrain the architecture, a synthetic multimodal dataset for autonomous driving applications was created. The dataset includes driving data from 100 diverse weather and traffic scenarios, gathered from multiple sensors including cameras and an IMU as well as vehicle control variables. The results show that including edge detection layers in the architecture improves performance for transfer learning when using synthetic and real-world data. In addition, pretraining with synthetic data reduces training time and enhances model performance when using real-world data.
- Open Access
Autonomous Vehicle Dataset with Real Multi-Driver Scenes and Biometric Data
The development of autonomous vehicles is becoming increasingly popular and gathering real-world data is considered a valuable task. Many datasets have been published recently in the autonomous vehicle sector, with synthetic datasets gaining particular interest due to availability and cost. For a real implementation and correct evaluation of vehicles at higher levels of autonomy, it is also necessary to consider human interaction, which is precisely something that lacks in existing datasets. In this article the UPCT dataset is presented, a public dataset containing high quality, multimodal data obtained using state-of-the-art sensors and equipment installed onboard the UPCT’s CICar autonomous vehicle. The dataset includes data from a variety of perception sensors including 3D LiDAR, cameras, IMU, GPS, encoders, as well as driver biometric data and driver behaviour questionnaires. In addition to the dataset, the software developed for data synchronisation and processing has been made available. The quality of the dataset was validated using an end-to-end neural network model with multiple inputs to obtain the speed and steering wheel angle and it obtained very promising results.
Open AccessAutonomous Vehicle Dataset with Real Multi-Driver Scenes and Biometric Data
The development of autonomous vehicles is becoming increasingly popular and gathering real-world data is considered a valuable task. Many datasets have been published recently in the autonomous vehicle sector, with synthetic datasets gaining particular interest due to availability and cost. For a real implementation and correct evaluation of vehicles at higher levels of autonomy, it is also necessary to consider human interaction, which is precisely something that lacks in existing datasets. In this article the UPCT dataset is presented, a public dataset containing high quality, multimodal data obtained using state-of-the-art sensors and equipment installed onboard the UPCT’s CICar autonomous vehicle. The dataset includes data from a variety of perception sensors including 3D LiDAR, cameras, IMU, GPS, encoders, as well as driver biometric data and driver behaviour questionnaires. In addition to the dataset, the software developed for data synchronisation and processing has been made available. The quality of the dataset was validated using an end-to-end neural network model with multiple inputs to obtain the speed and steering wheel angle and it obtained very promising results.
- Open Access
- 28 January 2023
- Published by MDPI AG in Preprints.org
Autonomous Vehicle Dataset with Real Multi-Driver Scenes and Biometric Data
The development of autonomous vehicles is becoming increasingly popular and gathering real world data is considered a valuable task. Many datasets have been published recently in the autonomous vehicle sector, with synthetic datasets gaining particular interest due to availability and cost. For a real implementation and correct evaluation of vehicles at higher levels of autonomy it is also necessary to consider human interaction, which is precisely something that lacks in existing datasets. In this article the UPCT dataset is presented, a public dataset containing high quality, multimodal data obtained using state of the art sensors and equipment installed onboard the UPCT’s CICar autonomous vehicle. The dataset includes data from a variety of perception sensors including 3D LiDAR, cameras, IMU, GPS, encoders, as well as driver biometric data and driver behaviour questionnaires. In addition to the dataset, the software developed for data synchronisation and processing has been made available. The quality of the dataset was validated using an end-to-end neural network model with multiple inputs to obtain speed and steering wheel angle and obtained very promising results.
- 28 January 2023
- Published by MDPI AG in Preprints.org
Open AccessAutonomous Vehicle Dataset with Real Multi-Driver Scenes and Biometric Data
The development of autonomous vehicles is becoming increasingly popular and gathering real world data is considered a valuable task. Many datasets have been published recently in the autonomous vehicle sector, with synthetic datasets gaining particular interest due to availability and cost. For a real implementation and correct evaluation of vehicles at higher levels of autonomy it is also necessary to consider human interaction, which is precisely something that lacks in existing datasets. In this article the UPCT dataset is presented, a public dataset containing high quality, multimodal data obtained using state of the art sensors and equipment installed onboard the UPCT’s CICar autonomous vehicle. The dataset includes data from a variety of perception sensors including 3D LiDAR, cameras, IMU, GPS, encoders, as well as driver biometric data and driver behaviour questionnaires. In addition to the dataset, the software developed for data synchronisation and processing has been made available. The quality of the dataset was validated using an end-to-end neural network model with multiple inputs to obtain speed and steering wheel angle and obtained very promising results.
- Open Access
- 14 June 2022
- Published by Oxford University Press (OUP) in GigaScience
- Vol. 11
A novel ground truth multispectral image dataset with weight, anthocyanins, and Brix index measures of grape berries tested for its utility in machine learning pipelines
Background: The combination of computer vision devices such as multispectral cameras coupled with artificial intelligence has provided a major leap forward in image-based analysis of biological processes. Supervised artificial intelligence algorithms require large ground truth image datasets for model training, which allows to validate or refute research hypotheses and to carry out comparisons between models. However, public datasets of images are scarce and ground truth images are surprisingly few considering the numbers required for training algorithms. Results: We created a dataset of 1,283 multidimensional arrays, using berries from five different grape varieties. Each array has 37 images of wavelengths between 488.38 and 952.76 nm obtained from single berries. Coupled to each multispectral image, we added a dataset with measurements including, weight, anthocyanin content, and Brix index for each independent grape. Thus, the images have paired measures, creating a ground truth dataset. We tested the dataset with 2 neural network algorithms: multilayer perceptron (MLP) and 3-dimensional convolutional neural network (3D-CNN). A perfect (100% accuracy) classification model was fit with either the MLP or 3D-CNN algorithms. Conclusions: This is the first public dataset of grape ground truth multispectral images. Associated with each multispectral image, there are measures of the weight, anthocyanins, and Brix index. The dataset should be useful to develop deep learning algorithms for classification, dimensionality reduction, regression, and prediction analysis.
- 14 June 2022
- Published by Oxford University Press (OUP) in GigaScience
- Vol. 11
Open AccessA novel ground truth multispectral image dataset with weight, anthocyanins, and Brix index measures of grape berries tested for its utility in machine learning pipelines
Background: The combination of computer vision devices such as multispectral cameras coupled with artificial intelligence has provided a major leap forward in image-based analysis of biological processes. Supervised artificial intelligence algorithms require large ground truth image datasets for model training, which allows to validate or refute research hypotheses and to carry out comparisons between models. However, public datasets of images are scarce and ground truth images are surprisingly few considering the numbers required for training algorithms. Results: We created a dataset of 1,283 multidimensional arrays, using berries from five different grape varieties. Each array has 37 images of wavelengths between 488.38 and 952.76 nm obtained from single berries. Coupled to each multispectral image, we added a dataset with measurements including, weight, anthocyanin content, and Brix index for each independent grape. Thus, the images have paired measures, creating a ground truth dataset. We tested the dataset with 2 neural network algorithms: multilayer perceptron (MLP) and 3-dimensional convolutional neural network (3D-CNN). A perfect (100% accuracy) classification model was fit with either the MLP or 3D-CNN algorithms. Conclusions: This is the first public dataset of grape ground truth multispectral images. Associated with each multispectral image, there are measures of the weight, anthocyanins, and Brix index. The dataset should be useful to develop deep learning algorithms for classification, dimensionality reduction, regression, and prediction analysis.
- Open Access
- 25 May 2021
- Published by MDPI AG in Electronics
- Vol. 10 (11) , 1266
End-to-End Deep Neural Network Architectures for Speed and Steering Wheel Angle Prediction in Autonomous Driving
The complex decision-making systems used for autonomous vehicles or advanced driver-assistance systems (ADAS) are being replaced by end-to-end (e2e) architectures based on deep-neural-networks (DNN). DNNs can learn complex driving actions from datasets containing thousands of images and data obtained from the vehicle perception system. This work presents the classification, design and implementation of six e2e architectures capable of generating the driving actions of speed and steering wheel angle directly on the vehicle control elements. The work details the design stages and optimization process of the convolutional networks to develop six e2e architectures. In the metric analysis the architectures have been tested with different data sources from the vehicle, such as images, XYZ accelerations and XYZ angular speeds. The best results were obtained with a mixed data e2e architecture that used front images from the vehicle and angular speeds to predict the speed and steering wheel angle with a mean error of 1.06%. An exhaustive optimization process of the convolutional blocks has demonstrated that it is possible to design lightweight e2e architectures with high performance more suitable for the final implementation in autonomous driving.
- 25 May 2021
- Published by MDPI AG in Electronics
- Vol. 10 (11) , 1266
Open AccessEnd-to-End Deep Neural Network Architectures for Speed and Steering Wheel Angle Prediction in Autonomous Driving
The complex decision-making systems used for autonomous vehicles or advanced driver-assistance systems (ADAS) are being replaced by end-to-end (e2e) architectures based on deep-neural-networks (DNN). DNNs can learn complex driving actions from datasets containing thousands of images and data obtained from the vehicle perception system. This work presents the classification, design and implementation of six e2e architectures capable of generating the driving actions of speed and steering wheel angle directly on the vehicle control elements. The work details the design stages and optimization process of the convolutional networks to develop six e2e architectures. In the metric analysis the architectures have been tested with different data sources from the vehicle, such as images, XYZ accelerations and XYZ angular speeds. The best results were obtained with a mixed data e2e architecture that used front images from the vehicle and angular speeds to predict the speed and steering wheel angle with a mean error of 1.06%. An exhaustive optimization process of the convolutional blocks has demonstrated that it is possible to design lightweight e2e architectures with high performance more suitable for the final implementation in autonomous driving.
- Open Access
- 17 February 2021
- Published by MDPI AG in Remote Sensing
- Vol. 13 (4) , 729
3DeepM: An Ad Hoc Architecture Based on Deep Learning Methods for Multispectral Image Classification
Current predefined architectures for deep learning are computationally very heavy and use tens of millions of parameters. Thus, computational costs may be prohibitive for many experimental or technological setups. We developed an ad hoc architecture for the classification of multispectral images using deep learning techniques. The architecture, called 3DeepM, is composed of 3D filter banks especially designed for the extraction of spatial-spectral features in multichannel images. The new architecture has been tested on a sample of 12210 multispectral images of seedless table grape varieties: Autumn Royal, Crimson Seedless, Itum4, Itum5 and Itum9. 3DeepM was able to classify 100% of the images and obtained the best overall results in terms of accuracy, number of classes, number of parameters and training time compared to similar work. In addition, this paper presents a flexible and reconfigurable computer vision system designed for the acquisition of multispectral images in the range of 400 nm to 1000 nm. The vision system enabled the creation of the first dataset consisting of 12210 37-channel multispectral images (12 VIS + 25 IR) of five seedless table grape varieties that have been used to validate the 3DeepM architecture. Compared to predefined classification architectures such as AlexNet, ResNet or ad hoc architectures with a very high number of parameters, 3DeepM shows the best classification performance despite using 130-fold fewer parameters than the architecture to which it was compared. 3DeepM can be used in a multitude of applications that use multispectral images, such as remote sensing or medical diagnosis. In addition, the small number of parameters of 3DeepM make it ideal for application in online classification systems aboard autonomous robots or unmanned vehicles.
- 17 February 2021
- Published by MDPI AG in Remote Sensing
- Vol. 13 (4) , 729
Open Access3DeepM: An Ad Hoc Architecture Based on Deep Learning Methods for Multispectral Image Classification
Current predefined architectures for deep learning are computationally very heavy and use tens of millions of parameters. Thus, computational costs may be prohibitive for many experimental or technological setups. We developed an ad hoc architecture for the classification of multispectral images using deep learning techniques. The architecture, called 3DeepM, is composed of 3D filter banks especially designed for the extraction of spatial-spectral features in multichannel images. The new architecture has been tested on a sample of 12210 multispectral images of seedless table grape varieties: Autumn Royal, Crimson Seedless, Itum4, Itum5 and Itum9. 3DeepM was able to classify 100% of the images and obtained the best overall results in terms of accuracy, number of classes, number of parameters and training time compared to similar work. In addition, this paper presents a flexible and reconfigurable computer vision system designed for the acquisition of multispectral images in the range of 400 nm to 1000 nm. The vision system enabled the creation of the first dataset consisting of 12210 37-channel multispectral images (12 VIS + 25 IR) of five seedless table grape varieties that have been used to validate the 3DeepM architecture. Compared to predefined classification architectures such as AlexNet, ResNet or ad hoc architectures with a very high number of parameters, 3DeepM shows the best classification performance despite using 130-fold fewer parameters than the architecture to which it was compared. 3DeepM can be used in a multitude of applications that use multispectral images, such as remote sensing or medical diagnosis. In addition, the small number of parameters of 3DeepM make it ideal for application in online classification systems aboard autonomous robots or unmanned vehicles.
- 5 September 2019
- Published by Springer Nature
- p. 277-286
Time of Flight Camera Calibration and Obstacle Detection Application for an Autonomous Vehicle
- 5 September 2019
- Published by Springer Nature
- p. 277-286
Time of Flight Camera Calibration and Obstacle Detection Application for an Autonomous Vehicle
- 1 September 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Obstacle Detection using a Time of Flight Range Camera
The objective of this work consisted in the design, implementation and evaluation of a robust object detection system, using only information acquired by a Time of Flight (ToF) camera. For object detection and discrimination against other elements that may appear in the scene, different descriptors and classifiers have been studied. To evaluate the algorithm, images obtained with the Sentis3D-M420 Time of Flight camera of the DSIE research group of the Polytechnic University of Cartagena have been used. The system has been designed to cover the blind spot area of a vehicle or integrated as part of an autonomous driving system.
- 1 September 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Obstacle Detection using a Time of Flight Range Camera
The objective of this work consisted in the design, implementation and evaluation of a robust object detection system, using only information acquired by a Time of Flight (ToF) camera. For object detection and discrimination against other elements that may appear in the scene, different descriptors and classifiers have been studied. To evaluate the algorithm, images obtained with the Sentis3D-M420 Time of Flight camera of the DSIE research group of the Polytechnic University of Cartagena have been used. The system has been designed to cover the blind spot area of a vehicle or integrated as part of an autonomous driving system.