Pub Date : 2023-10-16DOI: 10.1007/s41064-023-00263-x
Birger Winkel, David Nakath, Felix Woelk, Kevin Köser
Abstract To advance underwater computer vision and robotics from lab environments and clear water scenarios to the deep dark ocean or murky coastal waters, representative benchmarks and realistic datasets with ground truth information are required. In particular, determining the camera pose is essential for many underwater robotic or photogrammetric applications and known ground truth is mandatory to evaluate the performance of, e.g., simultaneous localization and mapping approaches in such extreme environments. This paper presents the conception, calibration, and implementation of an external reference system for determining the underwater camera pose in real time. The approach, based on an HTC Vive tracking system in air, calculates the underwater camera pose by fusing the poses of two controllers tracked above the water surface of a tank. It is shown that the mean deviation of this approach to an optical marker-based reference in air is less than 3 mm and 0.3 $$^{circ }$$ ∘ . Finally, the usability of the system for underwater applications is demonstrated.
{"title":"Design, Implementation, and Evaluation of an External Pose-Tracking System for Underwater Cameras","authors":"Birger Winkel, David Nakath, Felix Woelk, Kevin Köser","doi":"10.1007/s41064-023-00263-x","DOIUrl":"https://doi.org/10.1007/s41064-023-00263-x","url":null,"abstract":"Abstract To advance underwater computer vision and robotics from lab environments and clear water scenarios to the deep dark ocean or murky coastal waters, representative benchmarks and realistic datasets with ground truth information are required. In particular, determining the camera pose is essential for many underwater robotic or photogrammetric applications and known ground truth is mandatory to evaluate the performance of, e.g., simultaneous localization and mapping approaches in such extreme environments. This paper presents the conception, calibration, and implementation of an external reference system for determining the underwater camera pose in real time. The approach, based on an HTC Vive tracking system in air, calculates the underwater camera pose by fusing the poses of two controllers tracked above the water surface of a tank. It is shown that the mean deviation of this approach to an optical marker-based reference in air is less than 3 mm and 0.3 $$^{circ }$$ <mml:math xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"> <mml:msup> <mml:mrow /> <mml:mo>∘</mml:mo> </mml:msup> </mml:math> . Finally, the usability of the system for underwater applications is demonstrated.","PeriodicalId":56035,"journal":{"name":"PFG-Journal of Photogrammetry Remote Sensing and Geoinformation Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136116745","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-13DOI: 10.1007/s41064-023-00262-y
Markus Gerke, Michael Cramer
{"title":"Editorial for PFG Issue 5/2023","authors":"Markus Gerke, Michael Cramer","doi":"10.1007/s41064-023-00262-y","DOIUrl":"https://doi.org/10.1007/s41064-023-00262-y","url":null,"abstract":"","PeriodicalId":56035,"journal":{"name":"PFG-Journal of Photogrammetry Remote Sensing and Geoinformation Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135858445","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-06DOI: 10.1007/s41064-023-00260-0
Paweł Trybała, Jarosław Szrek, Fabio Remondino, Paulina Kujawa, Jacek Wodecki, Jan Blachowski, Radosław Zimroz
Abstract The research potential in the field of mobile mapping technologies is often hindered by several constraints. These include the need for costly hardware to collect data, limited access to target sites with specific environmental conditions or the collection of ground truth data for a quantitative evaluation of the developed solutions. To address these challenges, the research community has often prepared open datasets suitable for developments and testing. However, the availability of datasets that encompass truly demanding mixed indoor–outdoor and subterranean conditions, acquired with diverse but synchronized sensors, is currently limited. To alleviate this issue, we propose the MIN3D dataset (MultI-seNsor 3D mapping with an unmanned ground vehicle for mining applications) which includes data gathered using a wheeled mobile robot in two distinct locations: (i) textureless dark corridors and outside parts of a university campus and (ii) tunnels of an underground WW2 site in Walim (Poland). MIN3D comprises around 150 GB of raw data, including images captured by multiple co-calibrated monocular, stereo and thermal cameras, two LiDAR sensors and three inertial measurement units. Reliable ground truth (GT) point clouds were collected using a survey-grade terrestrial laser scanner. By openly sharing this dataset, we aim to support the efforts of the scientific community in developing robust methods for navigation and mapping in challenging underground conditions. In the paper, we describe the collected data and provide an initial accuracy assessment of some visual- and LiDAR-based simultaneous localization and mapping (SLAM) algorithms for selected sequences. Encountered problems, open research questions and areas that could benefit from utilizing our dataset are discussed. Data are available at https://3dom.fbk.eu/benchmarks .
{"title":"MIN3D Dataset: MultI-seNsor 3D Mapping with an Unmanned Ground Vehicle","authors":"Paweł Trybała, Jarosław Szrek, Fabio Remondino, Paulina Kujawa, Jacek Wodecki, Jan Blachowski, Radosław Zimroz","doi":"10.1007/s41064-023-00260-0","DOIUrl":"https://doi.org/10.1007/s41064-023-00260-0","url":null,"abstract":"Abstract The research potential in the field of mobile mapping technologies is often hindered by several constraints. These include the need for costly hardware to collect data, limited access to target sites with specific environmental conditions or the collection of ground truth data for a quantitative evaluation of the developed solutions. To address these challenges, the research community has often prepared open datasets suitable for developments and testing. However, the availability of datasets that encompass truly demanding mixed indoor–outdoor and subterranean conditions, acquired with diverse but synchronized sensors, is currently limited. To alleviate this issue, we propose the MIN3D dataset (MultI-seNsor 3D mapping with an unmanned ground vehicle for mining applications) which includes data gathered using a wheeled mobile robot in two distinct locations: (i) textureless dark corridors and outside parts of a university campus and (ii) tunnels of an underground WW2 site in Walim (Poland). MIN3D comprises around 150 GB of raw data, including images captured by multiple co-calibrated monocular, stereo and thermal cameras, two LiDAR sensors and three inertial measurement units. Reliable ground truth (GT) point clouds were collected using a survey-grade terrestrial laser scanner. By openly sharing this dataset, we aim to support the efforts of the scientific community in developing robust methods for navigation and mapping in challenging underground conditions. In the paper, we describe the collected data and provide an initial accuracy assessment of some visual- and LiDAR-based simultaneous localization and mapping (SLAM) algorithms for selected sequences. Encountered problems, open research questions and areas that could benefit from utilizing our dataset are discussed. Data are available at https://3dom.fbk.eu/benchmarks .","PeriodicalId":56035,"journal":{"name":"PFG-Journal of Photogrammetry Remote Sensing and Geoinformation Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135350313","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-05DOI: 10.1007/s41064-023-00258-8
Nando Metzger, Mehmet Özgür Türkoglu, Rodrigo Caye Daudt, Jan Dirk Wegner, Konrad Schindler
Abstract Forecasting where and when new buildings will emerge is a rather unexplored topic, but one that is very useful in many disciplines such as urban planning, agriculture, resource management, and even autonomous flying. In the present work, we present a method that accomplishes this task with a deep neural network and a custom pretraining procedure. In Stage 1 , a U-Net backbone is pretrained within a Siamese network architecture that aims to solve a (building) change detection task. In Stage 2 , the backbone is repurposed to forecast the emergence of new buildings based solely on one image acquired before its construction. Furthermore, we also present a model that forecasts the time range within which the change will occur. We validate our approach using the SpaceNet7 dataset, which covers an area of 960 km $$^2$$ 2 at 24 points in time across 2 years. In our experiments, we found that our proposed pretraining method consistently outperforms the traditional pretraining using the ImageNet dataset. We also show that it is to some degree possible to predict in advance when building changes will occur.
{"title":"Urban Change Forecasting from Satellite Images","authors":"Nando Metzger, Mehmet Özgür Türkoglu, Rodrigo Caye Daudt, Jan Dirk Wegner, Konrad Schindler","doi":"10.1007/s41064-023-00258-8","DOIUrl":"https://doi.org/10.1007/s41064-023-00258-8","url":null,"abstract":"Abstract Forecasting where and when new buildings will emerge is a rather unexplored topic, but one that is very useful in many disciplines such as urban planning, agriculture, resource management, and even autonomous flying. In the present work, we present a method that accomplishes this task with a deep neural network and a custom pretraining procedure. In Stage 1 , a U-Net backbone is pretrained within a Siamese network architecture that aims to solve a (building) change detection task. In Stage 2 , the backbone is repurposed to forecast the emergence of new buildings based solely on one image acquired before its construction. Furthermore, we also present a model that forecasts the time range within which the change will occur. We validate our approach using the SpaceNet7 dataset, which covers an area of 960 km $$^2$$ <mml:math xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"> <mml:msup> <mml:mrow /> <mml:mn>2</mml:mn> </mml:msup> </mml:math> at 24 points in time across 2 years. In our experiments, we found that our proposed pretraining method consistently outperforms the traditional pretraining using the ImageNet dataset. We also show that it is to some degree possible to predict in advance when building changes will occur.","PeriodicalId":56035,"journal":{"name":"PFG-Journal of Photogrammetry Remote Sensing and Geoinformation Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135481560","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-04DOI: 10.1007/s41064-023-00257-9
Rim Katlane, David Doxaran, Boubaker ElKilani, Chaïma Trabelsi
{"title":"Remote Sensing of Turbidity in Optically Shallow Waters Using Sentinel-2 MSI and PRISMA Satellite Data","authors":"Rim Katlane, David Doxaran, Boubaker ElKilani, Chaïma Trabelsi","doi":"10.1007/s41064-023-00257-9","DOIUrl":"https://doi.org/10.1007/s41064-023-00257-9","url":null,"abstract":"","PeriodicalId":56035,"journal":{"name":"PFG-Journal of Photogrammetry Remote Sensing and Geoinformation Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135592115","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-26DOI: 10.1007/s41064-023-00256-w
Maryam Teimouri, Mehdi Mokhtarzade, Nicolas Baghdadi, Christian Heipke
Abstract Convolutional neural networks (CNNs) have shown results superior to most traditional image understanding approaches in many fields, incl. crop classification from satellite time series images. However, CNNs require a large number of training samples to properly train the network. The process of collecting and labeling such samples using traditional methods can be both, time-consuming and costly. To address this issue and improve classification accuracy, generating virtual training labels (VTL) from existing ones is a promising solution. To this end, this study proposes a novel method for generating VTL based on sub-dividing the training samples of each crop using self-organizing maps (SOM), and then assigning labels to a set of unlabeled pixels based on the distance to these sub-classes. We apply the new method to crop classification from Sentinel images. A three-dimensional (3D) CNN is utilized for extracting features from the fusion of optical and radar time series. The results of the evaluation show that the proposed method is effective in generating VTL, as demonstrated by the achieved overall accuracy (OA) of 95.3% and kappa coefficient (KC) of 94.5%, compared to 91.3% and 89.9% for a solution without VTL. The results suggest that the proposed method has the potential to enhance the classification accuracy of crops using VTL.
{"title":"Generating Virtual Training Labels for Crop Classification from Fused Sentinel-1 and Sentinel-2 Time Series","authors":"Maryam Teimouri, Mehdi Mokhtarzade, Nicolas Baghdadi, Christian Heipke","doi":"10.1007/s41064-023-00256-w","DOIUrl":"https://doi.org/10.1007/s41064-023-00256-w","url":null,"abstract":"Abstract Convolutional neural networks (CNNs) have shown results superior to most traditional image understanding approaches in many fields, incl. crop classification from satellite time series images. However, CNNs require a large number of training samples to properly train the network. The process of collecting and labeling such samples using traditional methods can be both, time-consuming and costly. To address this issue and improve classification accuracy, generating virtual training labels (VTL) from existing ones is a promising solution. To this end, this study proposes a novel method for generating VTL based on sub-dividing the training samples of each crop using self-organizing maps (SOM), and then assigning labels to a set of unlabeled pixels based on the distance to these sub-classes. We apply the new method to crop classification from Sentinel images. A three-dimensional (3D) CNN is utilized for extracting features from the fusion of optical and radar time series. The results of the evaluation show that the proposed method is effective in generating VTL, as demonstrated by the achieved overall accuracy (OA) of 95.3% and kappa coefficient (KC) of 94.5%, compared to 91.3% and 89.9% for a solution without VTL. The results suggest that the proposed method has the potential to enhance the classification accuracy of crops using VTL.","PeriodicalId":56035,"journal":{"name":"PFG-Journal of Photogrammetry Remote Sensing and Geoinformation Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134960018","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-19DOI: 10.1007/s41064-023-00254-y
Umut Gunes Sefercik, Ismail Colkesen, Taskin Kavzoglu, Nizamettin Ozdogan, Muhammed Yusuf Ozturk
{"title":"Assessing the Physical and Chemical Characteristics of Marine Mucilage Utilizing In-Situ and Remote Sensing Data (Sentinel-1, -2, -3)","authors":"Umut Gunes Sefercik, Ismail Colkesen, Taskin Kavzoglu, Nizamettin Ozdogan, Muhammed Yusuf Ozturk","doi":"10.1007/s41064-023-00254-y","DOIUrl":"https://doi.org/10.1007/s41064-023-00254-y","url":null,"abstract":"","PeriodicalId":56035,"journal":{"name":"PFG-Journal of Photogrammetry Remote Sensing and Geoinformation Science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135060842","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-06DOI: 10.1007/s41064-023-00255-x
A. Sedighi, S. Hamzeh, M. K. Firozjaei, Hamid Valipoori Goodarzi, A. Naseri
{"title":"Comparative Analysis of Multispectral and Hyperspectral Imagery for Mapping Sugarcane Varieties","authors":"A. Sedighi, S. Hamzeh, M. K. Firozjaei, Hamid Valipoori Goodarzi, A. Naseri","doi":"10.1007/s41064-023-00255-x","DOIUrl":"https://doi.org/10.1007/s41064-023-00255-x","url":null,"abstract":"","PeriodicalId":56035,"journal":{"name":"PFG-Journal of Photogrammetry Remote Sensing and Geoinformation Science","volume":null,"pages":null},"PeriodicalIF":4.1,"publicationDate":"2023-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82060067","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-07-28DOI: 10.1007/s41064-023-00252-0
Waseem Iqbal, J. Paffenholz, M. Mehltretter
{"title":"Guiding Deep Learning with Expert Knowledge for Dense Stereo Matching","authors":"Waseem Iqbal, J. Paffenholz, M. Mehltretter","doi":"10.1007/s41064-023-00252-0","DOIUrl":"https://doi.org/10.1007/s41064-023-00252-0","url":null,"abstract":"","PeriodicalId":56035,"journal":{"name":"PFG-Journal of Photogrammetry Remote Sensing and Geoinformation Science","volume":null,"pages":null},"PeriodicalIF":4.1,"publicationDate":"2023-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87870794","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-07-24DOI: 10.1007/s41064-023-00253-z
P. U. Hatipoglu, C. Iyigun, Sinan Kalkan
{"title":"Crowd-aware Thresholded Loss for Object Detection in Wide Area Motion Imagery","authors":"P. U. Hatipoglu, C. Iyigun, Sinan Kalkan","doi":"10.1007/s41064-023-00253-z","DOIUrl":"https://doi.org/10.1007/s41064-023-00253-z","url":null,"abstract":"","PeriodicalId":56035,"journal":{"name":"PFG-Journal of Photogrammetry Remote Sensing and Geoinformation Science","volume":null,"pages":null},"PeriodicalIF":4.1,"publicationDate":"2023-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80218625","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：