JW
J. Wu
info
Please Note
<p>This page displays the records of the person named above and is not linked to a unique person identifier. This record may need to be merged to a profile.</p>
2 records found
1
Building material identification plays an important role in applications such as sustainable building assessment, facade inspection, energy efficiency analysis, and urban environmental studies. Traditional material identification approaches mainly rely on optical imagery or geometric information, which are often limited in capturing the thermophysical characteristics of facade materials. Thermal imagery provides additional information related to heat transfer behavior, emissivity, and thermal inertia, offering an alternative perspective for building material analysis.
This thesis investigates the use of spatio-temporal thermal imagery combined with a Convolutional Long Short-Term Memory (Convolutional Long Short-Term Memory (ConvLSTM))- based deep learning framework for pixel-level building material classification. Groundbased thermal image sequences of building facades were collected under different environmental and temporal conditions using a thermal camera. Building materials including glass, concrete, and brick were manually annotated through Red Green Blue (RGB) imagery and Segment Anything Model (SAM)-assisted segmentation. Image registration techniques were subsequently applied to align the RGB and thermal image sequences, enabling the transfer of material labels into the thermal domain.
A ConvLSTM-based semantic segmentation model was developed to capture both spatial thermal distributions and temporal thermal dynamics from sequential thermal imagery. Different dataset splitting strategies and temporal datasets were evaluated to investigate the influence of spatial generalization and temporal variation on classification performance. The experimental results demonstrate that spatio-temporal thermal sequences provide meaningful discriminative information for building material classification. Under the random sequence-level split strategy, the proposed framework achieved a mean F1-score of approximately 0.686 and a mean Intersection over Union (IoU) of approximately 0.524 across the evaluated material classes.
The results indicate that temporal thermal behavior significantly contributes to distinguishing facade materials. Brick surfaces exhibited relatively stable and spatially uniform thermal characteristics, while glass surfaces demonstrated stronger temporal variation and environmental sensitivity. Concrete exhibited intermediate thermal behavior, making it comparatively more difficult to classify. The findings suggest that incorporating temporal thermal information improves segmentation performance compared with single-frame thermal analysis.
...
This thesis investigates the use of spatio-temporal thermal imagery combined with a Convolutional Long Short-Term Memory (Convolutional Long Short-Term Memory (ConvLSTM))- based deep learning framework for pixel-level building material classification. Groundbased thermal image sequences of building facades were collected under different environmental and temporal conditions using a thermal camera. Building materials including glass, concrete, and brick were manually annotated through Red Green Blue (RGB) imagery and Segment Anything Model (SAM)-assisted segmentation. Image registration techniques were subsequently applied to align the RGB and thermal image sequences, enabling the transfer of material labels into the thermal domain.
A ConvLSTM-based semantic segmentation model was developed to capture both spatial thermal distributions and temporal thermal dynamics from sequential thermal imagery. Different dataset splitting strategies and temporal datasets were evaluated to investigate the influence of spatial generalization and temporal variation on classification performance. The experimental results demonstrate that spatio-temporal thermal sequences provide meaningful discriminative information for building material classification. Under the random sequence-level split strategy, the proposed framework achieved a mean F1-score of approximately 0.686 and a mean Intersection over Union (IoU) of approximately 0.524 across the evaluated material classes.
The results indicate that temporal thermal behavior significantly contributes to distinguishing facade materials. Brick surfaces exhibited relatively stable and spatially uniform thermal characteristics, while glass surfaces demonstrated stronger temporal variation and environmental sensitivity. Concrete exhibited intermediate thermal behavior, making it comparatively more difficult to classify. The findings suggest that incorporating temporal thermal information improves segmentation performance compared with single-frame thermal analysis.
...
Building material identification plays an important role in applications such as sustainable building assessment, facade inspection, energy efficiency analysis, and urban environmental studies. Traditional material identification approaches mainly rely on optical imagery or geometric information, which are often limited in capturing the thermophysical characteristics of facade materials. Thermal imagery provides additional information related to heat transfer behavior, emissivity, and thermal inertia, offering an alternative perspective for building material analysis.
This thesis investigates the use of spatio-temporal thermal imagery combined with a Convolutional Long Short-Term Memory (Convolutional Long Short-Term Memory (ConvLSTM))- based deep learning framework for pixel-level building material classification. Groundbased thermal image sequences of building facades were collected under different environmental and temporal conditions using a thermal camera. Building materials including glass, concrete, and brick were manually annotated through Red Green Blue (RGB) imagery and Segment Anything Model (SAM)-assisted segmentation. Image registration techniques were subsequently applied to align the RGB and thermal image sequences, enabling the transfer of material labels into the thermal domain.
A ConvLSTM-based semantic segmentation model was developed to capture both spatial thermal distributions and temporal thermal dynamics from sequential thermal imagery. Different dataset splitting strategies and temporal datasets were evaluated to investigate the influence of spatial generalization and temporal variation on classification performance. The experimental results demonstrate that spatio-temporal thermal sequences provide meaningful discriminative information for building material classification. Under the random sequence-level split strategy, the proposed framework achieved a mean F1-score of approximately 0.686 and a mean Intersection over Union (IoU) of approximately 0.524 across the evaluated material classes.
The results indicate that temporal thermal behavior significantly contributes to distinguishing facade materials. Brick surfaces exhibited relatively stable and spatially uniform thermal characteristics, while glass surfaces demonstrated stronger temporal variation and environmental sensitivity. Concrete exhibited intermediate thermal behavior, making it comparatively more difficult to classify. The findings suggest that incorporating temporal thermal information improves segmentation performance compared with single-frame thermal analysis.
This thesis investigates the use of spatio-temporal thermal imagery combined with a Convolutional Long Short-Term Memory (Convolutional Long Short-Term Memory (ConvLSTM))- based deep learning framework for pixel-level building material classification. Groundbased thermal image sequences of building facades were collected under different environmental and temporal conditions using a thermal camera. Building materials including glass, concrete, and brick were manually annotated through Red Green Blue (RGB) imagery and Segment Anything Model (SAM)-assisted segmentation. Image registration techniques were subsequently applied to align the RGB and thermal image sequences, enabling the transfer of material labels into the thermal domain.
A ConvLSTM-based semantic segmentation model was developed to capture both spatial thermal distributions and temporal thermal dynamics from sequential thermal imagery. Different dataset splitting strategies and temporal datasets were evaluated to investigate the influence of spatial generalization and temporal variation on classification performance. The experimental results demonstrate that spatio-temporal thermal sequences provide meaningful discriminative information for building material classification. Under the random sequence-level split strategy, the proposed framework achieved a mean F1-score of approximately 0.686 and a mean Intersection over Union (IoU) of approximately 0.524 across the evaluated material classes.
The results indicate that temporal thermal behavior significantly contributes to distinguishing facade materials. Brick surfaces exhibited relatively stable and spatially uniform thermal characteristics, while glass surfaces demonstrated stronger temporal variation and environmental sensitivity. Concrete exhibited intermediate thermal behavior, making it comparatively more difficult to classify. The findings suggest that incorporating temporal thermal information improves segmentation performance compared with single-frame thermal analysis.
Student report
(2024)
-
N.P. Alting, H. Baba, Derian Der Derian Auliyaa Bainus, H.Y. Cheng, J. Wu, E. Verbree, Niels van der Vaart, A.N. Yunisya
This project presents an indoor navigation system based on image matching, aiming to address the challenges of localization and navigation in indoor environments. The system utilizes Simultaneous Localization and Mapping (SLAM) technology to capture high-resolution images and point cloud data, combined with the VGG16 model from Convolutional Neural Networks (CNN) for image processing, feature extraction, and matching.
In our research, we conducted experiments at the Faculty of Architecture and the Built Environment of Delft University of Technology, using a SLAM scanner to obtain 360-degree panoramic images and point cloud data of the indoor environment. Through cube mapping projection, we converted the panoramic images into six planar views, selecting the front, right, and left views as positioning references. Additionally, we reconstructed the indoor environment structure and designed node networks for positioning and navigation.
The technical architecture of this system comprises three main components: VGG16-based image feature extraction, cosine similarity-based image matching, and DBSCAN algorithm for location clustering. Through this method, the system can achieve real-time localization results after image capture and provide users with optimal paths using the A* navigation algorithm.
Experimental results show that when using single image matching, the system's room localization accuracy reaches 74.65\%. When employing multiple image matching and DBSCAN clustering methods, the accuracy significantly improves. In our final evaluation involving 116 positions, the system successfully matched 111 of these positions to their correct rooms, achieving a localization accuracy of 95.69\%.
This research not only provides an innovative solution for indoor positioning and navigation but also points the way for future research, including support for multi-floor navigation, enhancing CNN model performance, and automating building processing. This technology has the potential for widespread application in complex indoor environments such as large buildings, conference centers, and university campuses, offering users accurate, real-time positioning and navigation services. ...
In our research, we conducted experiments at the Faculty of Architecture and the Built Environment of Delft University of Technology, using a SLAM scanner to obtain 360-degree panoramic images and point cloud data of the indoor environment. Through cube mapping projection, we converted the panoramic images into six planar views, selecting the front, right, and left views as positioning references. Additionally, we reconstructed the indoor environment structure and designed node networks for positioning and navigation.
The technical architecture of this system comprises three main components: VGG16-based image feature extraction, cosine similarity-based image matching, and DBSCAN algorithm for location clustering. Through this method, the system can achieve real-time localization results after image capture and provide users with optimal paths using the A* navigation algorithm.
Experimental results show that when using single image matching, the system's room localization accuracy reaches 74.65\%. When employing multiple image matching and DBSCAN clustering methods, the accuracy significantly improves. In our final evaluation involving 116 positions, the system successfully matched 111 of these positions to their correct rooms, achieving a localization accuracy of 95.69\%.
This research not only provides an innovative solution for indoor positioning and navigation but also points the way for future research, including support for multi-floor navigation, enhancing CNN model performance, and automating building processing. This technology has the potential for widespread application in complex indoor environments such as large buildings, conference centers, and university campuses, offering users accurate, real-time positioning and navigation services. ...
This project presents an indoor navigation system based on image matching, aiming to address the challenges of localization and navigation in indoor environments. The system utilizes Simultaneous Localization and Mapping (SLAM) technology to capture high-resolution images and point cloud data, combined with the VGG16 model from Convolutional Neural Networks (CNN) for image processing, feature extraction, and matching.
In our research, we conducted experiments at the Faculty of Architecture and the Built Environment of Delft University of Technology, using a SLAM scanner to obtain 360-degree panoramic images and point cloud data of the indoor environment. Through cube mapping projection, we converted the panoramic images into six planar views, selecting the front, right, and left views as positioning references. Additionally, we reconstructed the indoor environment structure and designed node networks for positioning and navigation.
The technical architecture of this system comprises three main components: VGG16-based image feature extraction, cosine similarity-based image matching, and DBSCAN algorithm for location clustering. Through this method, the system can achieve real-time localization results after image capture and provide users with optimal paths using the A* navigation algorithm.
Experimental results show that when using single image matching, the system's room localization accuracy reaches 74.65\%. When employing multiple image matching and DBSCAN clustering methods, the accuracy significantly improves. In our final evaluation involving 116 positions, the system successfully matched 111 of these positions to their correct rooms, achieving a localization accuracy of 95.69\%.
This research not only provides an innovative solution for indoor positioning and navigation but also points the way for future research, including support for multi-floor navigation, enhancing CNN model performance, and automating building processing. This technology has the potential for widespread application in complex indoor environments such as large buildings, conference centers, and university campuses, offering users accurate, real-time positioning and navigation services.
In our research, we conducted experiments at the Faculty of Architecture and the Built Environment of Delft University of Technology, using a SLAM scanner to obtain 360-degree panoramic images and point cloud data of the indoor environment. Through cube mapping projection, we converted the panoramic images into six planar views, selecting the front, right, and left views as positioning references. Additionally, we reconstructed the indoor environment structure and designed node networks for positioning and navigation.
The technical architecture of this system comprises three main components: VGG16-based image feature extraction, cosine similarity-based image matching, and DBSCAN algorithm for location clustering. Through this method, the system can achieve real-time localization results after image capture and provide users with optimal paths using the A* navigation algorithm.
Experimental results show that when using single image matching, the system's room localization accuracy reaches 74.65\%. When employing multiple image matching and DBSCAN clustering methods, the accuracy significantly improves. In our final evaluation involving 116 positions, the system successfully matched 111 of these positions to their correct rooms, achieving a localization accuracy of 95.69\%.
This research not only provides an innovative solution for indoor positioning and navigation but also points the way for future research, including support for multi-floor navigation, enhancing CNN model performance, and automating building processing. This technology has the potential for widespread application in complex indoor environments such as large buildings, conference centers, and university campuses, offering users accurate, real-time positioning and navigation services.