Visual perception is regularly used by humans and robots for navigation. By either implicitly or ... more Visual perception is regularly used by humans and robots for navigation. By either implicitly or explicitly mapping the environment, ego-motion can be determined and a path of actions can be planned. The process of mapping and navigation are delicately intertwined; therefore, improving one can often lead to an improvement of the other. Both processes are sensitive to the interior orientation parameters of the camera system and mathematically modelling these systematic errors can often improve the precision and accuracy of the overall solution. This paper presents an automatic camera calibration method suitable for any lens, without having prior knowledge about the sensor. Statistical inference is performed to map the environment and localize the camera simultaneously. K-nearest neighbour regression is used to model the geometric distortions of the images. A normal-angle lens Nikon camera and wide-angle lens GoPro camera were calibrated using the proposed method, as well as the conventional bundle adjustment with self-calibration method (for comparison). Results showed that the mapping error was reduced from an average of 14.9 mm to 1.2 mm (i.e. a 92% improvement) and 66.6 mm to 1.5 mm (i.e. a 98% improvement) using the proposed method for the Nikon and GoPro cameras, respectively. In contrast, the conventional approach achieved an average 3D error of 0.9 mm (i.e. 94% improvement) and 3.3 mm (i.e. 95% improvement) for the Nikon and GoPro cameras, respectively. Thus, the proposed method performs well irrespective of the lens/sensor used: it yields results that are comparable to the conventional approach for normal-angle lens cameras, and it has the additional benefit of improving calibration results for wide-angle lens cameras.
X-ray imaging is a fundamental tool of routine clinical diagnosis. Fluoroscopic imaging can furth... more X-ray imaging is a fundamental tool of routine clinical diagnosis. Fluoroscopic imaging can further acquire X-ray images at video frame rates, thus enabling non-invasive in-vivo motion studies of joints, gastrointestinal tract, etc. For both the qualitative and quantitative analysis of static and dynamic X-ray images, the data should be free of systematic biases. Besides precise fabrication of hardware, software-based calibration solutions are commonly used for modelling the distortions. In this primary research study, a robust photogrammetric bundle adjustment was used to model the projective geometry of two fluoroscopic X-ray imaging systems. However, instead of relying on an expert photogrammetrist's knowledge and judgement to decide on a parametric model for describing the systematic errors, a self-tuning data-driven approach is used to model the complex non-linear distortion profile of the sensors. Quality control from the experiment showed that 0.06 mm to 0.09 mm 3D reconstruction accuracy was achievable post-calibration using merely 15 X-ray images. As part of the bundle adjustment, the location of the virtual fluoroscopic system relative to the target field can also be spatially resected with an RMSE between 3.10 mm and 3.31 mm.
Inertial measurement units (IMUs) are fundamental for attitude control of drones. With the advanc... more Inertial measurement units (IMUs) are fundamental for attitude control of drones. With the advancements in micro-electro-mechanical systems (MEMS) fabrication processes, size, power consumption, and price of these sensors have reduced significantly and attracted many new applications. However, this came at the expense of sensors requiring frequent recalibration, as they are highly contaminated with systematic errors. This paper presents a novel method to jointly calibrate the accelerometer, gyroscope, and magnetometer triad in a MEMS IMU without additional equipment. Opportunistic zero change in velocity and position updates, and inclination updates were used in conjunction with relative orientation updates from magnetometers in a robust batch least-squares adjustment. Solutions from the proposed self-calibration were compared to existing calibration methods. Empirical results suggest that the new method is robust against magnetic distortions and can achieve performance similar to a specialized calibration that uses a more accurate (and expensive) IMU as reference. The jointly estimated accelerometer and gyroscope calibration parameters can deliver a more accurate dead-reckoning solution than the popular multi-position calibration method (i.e., 54% improvement in orientation accuracy) by recovering the gyroscope scale error and other systematic errors. In addition, it can improve parameter observability as well as reduce calibration time by incorporating dynamic data with static orientations. The proposed calibration was also applied on-site pre-mission by simply waving the sensor by hand and was able to improve the orientation tracking accuracy by 73%.
In the absence of external reference position information (e.g. surveyed targets or Global Naviga... more In the absence of external reference position information (e.g. surveyed targets or Global Navigation Satellite Systems) Simultaneous Localization and Mapping (SLAM) has proven to be an effective method for indoor navigation. The positioning drift can be reduced with regular loop-closures and global relaxation as the backend, thus achieving a good balance between exploration and exploitation. Although vision-based systems like laser scanners are typically deployed for SLAM, these sensors are heavy, energy inefficient, and expensive, making them unattractive for wearables or smartphone applications. However, the concept of SLAM can be extended to non-optical systems such as magnetometers. Instead of matching features such as walls and furniture using some variation of the Iterative Closest Point algorithm, the local magnetic field can be matched to provide loop-closure and global trajectory updates in a Gaussian Process (GP) SLAM framework. With a MEMS-based inertial measurement unit providing a continuous trajectory, and the matching of locally distinct magnetic field maps, experimental results in this paper show that a drift-free navigation solution in an indoor environment with millimetre-level accuracy can be achieved. The GP-SLAM approach presented can be formulated as a maximum a posteriori estimation problem and it can naturally perform loop-detection, feature-to-feature distance minimization, global trajectory optimization, and magnetic field map estimation simultaneously. Spatially continuous features (i.e. smooth magnetic field signatures) are used instead of discrete feature correspondences (e.g. point-to-point) as in conventional vision-based SLAM. These position updates from the ambient magnetic field also provide enough information for calibrating the accelerometer bias and gyroscope bias in-use. The only restriction for this method is the need for magnetic disturbances (which is typically not an issue for indoor environments); however, no assumptions are required for the general motion of the sensor (e.g. static periods).
Sensor fusion of a MEMS IMU with a magnetometer is a popular system design, because such 9-DoF (d... more Sensor fusion of a MEMS IMU with a magnetometer is a popular system design, because such 9-DoF (degrees of freedom) systems are capable of achieving drift-free 3D orientation tracking. However, these systems are often vulnerable to ambient magnetic distortions and lack useful position information; in the absence of external position aiding (e.g. satellite/ultra-wideband positioning systems) the dead-reckoned position accuracy from a 9-DoF MEMS IMU deteriorates rapidly due to unmodelled errors. Positioning information is valuable in many satellite-denied geomatics applications (e.g. indoor navigation, location-based services, etc.). This paper proposes an improved 9-DoF IMU indoor pose tracking method using batch optimization. By adopting a robust in-situ user self-calibration approach to model the systematic errors of the accelerometer, gyroscope, and magnetometer simultaneously in a tightly-coupled post-processed least-squares framework, the accuracy of the estimated trajectory from a 9-DoF MEMS IMU can be improved. Through a combination of relative magnetic measurement updates and a robust weight function, the method is able to tolerate a high level of magnetic distortions. The proposed auto-calibration method was tested in-use under various heterogeneous magnetic field conditions to mimic a person walking with the sensor in their pocket, a person checking their phone, and a person walking with a smartwatch. In these experiments, the presented algorithm improved the in-situ dead-reckoning orientation accuracy by 79.8 – 89.5% and the dead-reckoned positioning accuracy by 72.9 – 92.8%, thus reducing the relative positioning error from metre-level to decimetre-level after ten seconds of integration, without making assumptions about the user's dynamics.
Data Revues 18770568 Unassign S1877056814001935, Jan 10, 2014
Measuring dynamic in vivo shoulder kinematics is crucial to better understanding numerous patholo... more Measuring dynamic in vivo shoulder kinematics is crucial to better understanding numerous pathologies. Motion capture systems using skin-mounted markers offer good solutions for non-invasive assessment of shoulder kinematics during dynamic movement. However, none of the current motion capture techniques have been used to study translation values at the joint, which is crucial to assess shoulder instability. The aim of the present study was to develop a dedicated patient-specific measurement technique based on motion capture and magnetic resonance imaging (MRI) to determine shoulder kinematics accurately. Estimation of both rotations and translations at the shoulder joint using motion capture is feasible thanks to a patient-specific kinematic chain of the shoulder complex reconstructed from MRI data. We implemented a patient-specific kinematic chain model of the shoulder complex with loose constraints on joint translation. To assess the effectiveness of the technique, six subjects underwent data acquisition simultaneously with fluoroscopy and motion capture during flexion and empty-can abduction. The reference 3D shoulder kinematics was reconstructed from fluoroscopy and compared to that obtained from the new technique using skin markers. Root mean square errors (RMSE) for shoulder orientation were within 4° (mean range: 2.0°-3.4°) for each anatomical axis and each motion. For glenohumeral translations, maximum RMSE for flexion was 3.7mm and 3.5mm for empty-can abduction (mean range: 1.9-3.3mm). Although the translation errors were significant, the computed patterns of humeral translation showed good agreement with published data. To our knowledge, this study is the first attempt to calculate both rotations and translations at the shoulder joint based on skin-mounted markers. Results were encouraging and can serve as reference for future developments. The proposed technique could provide valuable kinematic data for the study of shoulder pathologies. Basic Science Study.
Orthopaedics & traumatology, surgery & research : OTSR, 2014
Measuring dynamic in vivo shoulder kinematics is crucial to better understanding numerous patholo... more Measuring dynamic in vivo shoulder kinematics is crucial to better understanding numerous pathologies. Motion capture systems using skin-mounted markers offer good solutions for non-invasive assessment of shoulder kinematics during dynamic movement. However, none of the current motion capture techniques have been used to study translation values at the joint, which is crucial to assess shoulder instability. The aim of the present study was to develop a dedicated patient-specific measurement technique based on motion capture and magnetic resonance imaging (MRI) to determine shoulder kinematics accurately. Estimation of both rotations and translations at the shoulder joint using motion capture is feasible thanks to a patient-specific kinematic chain of the shoulder complex reconstructed from MRI data. We implemented a patient-specific kinematic chain model of the shoulder complex with loose constraints on joint translation. To assess the effectiveness of the technique, six subjects un...
This paper focuses on the use of control linear features extraction from a terrestrial laser scan... more This paper focuses on the use of control linear features extraction from a terrestrial laser scanning (TLS) surface to produce a good datum definition for a close range photogrammetric model. The difficulty of identifying conjugate points between both datasets is overcome by the derivation of automatic features extraction. Hence, the extracted features from the terrestrial measurements represent a good candidate for co-registration. Field work is performed to validate the proposed methods. It combines a TLS acquisition on different city buildings, while a complete dataset of close range images of the same area is measured. Some additional geodetic survey control is added to the overall data to perform global registration. Those experiments are based on planar patches automatic recognition derived from segmentation techniques. Planar segmentation is the extraction of planes from a point cloud. Assuming that linear features are at the intersection of those planes, we are able to gener...
Over the last few years, LiDAR has become a popular technology for the direct acquisition of topo... more Over the last few years, LiDAR has become a popular technology for the direct acquisition of topographic information. In spite of the increasing utilization of this technology in several applications, its accuracy potential has not been fully explored. Most of current LiDAR calibration techniques are based on empirical and proprietary procedures that demand the system’s raw measurements, which may not be always available to the end-user. As a result, we can still observe systematic discrepancies between conjugate surface elements in overlapping LiDAR strips. In this paper, two alternative calibration procedures that overcome the existing limitations are introduced. The first procedure, denoted as “Simplified method”, makes use of the LiDAR point cloud from parallel LiDAR strips acquired by a steady platform (e.g., fixed wing aircraft) over an area with moderately varying elevation. The second procedure, denoted as “Quasi-rigorous method”, can deal with non-parallel strips, but requi...
Current advances in digital and electronic products have led to the availability of inexpensive a... more Current advances in digital and electronic products have led to the availability of inexpensive and reliable Medium Format Digital Cameras (MFDCs) that can be used in many photogrammetric applications. In this research, the impact of camera and system calibration on object space reconstruction is investigated under different georeferencing scenarios (i.e., indirect georeferencing and integrated sensor orientation). First, camera calibration is conducted using a MFDC (i.e., the Rollei-P65). Based on different camera calibration datasets – such as indoor, in-situ, and camera calibration certificates, the equivalency of the calibration techniques as well as the adequacy of the distortion models are evaluated while considering relative and absolute quantitative measures. Previously developed camera stability analysis technique will be used for testing the adequacy of the utilized distortion model as well as the equivalency of different calibration techniques. Afterwards, system calibrat...
ABSTRACT The geometric calibration of time-of-flight range cameras is a necessary quality assuran... more ABSTRACT The geometric calibration of time-of-flight range cameras is a necessary quality assurance measure performed to estimate the interior orientation parameters. Self-calibration from a network of range imagery of an array of signalized targets arranged in one or two planes can be used for this purpose. The latter configuration requires the addition of a parametric model for internal light scattering biases in the range observations to the background plane due to the presence of the foreground plane. In a previous study of MESA Imaging SwissRanger range cameras, such a model was developed and shown to be effective. A new parametric model is proposed here because the scattering error behavior is camera model dependent. The new model was tested on two pmdtechnologies range cameras, the CamCube 3.0 and CamBoard nano, and its effectiveness was demonstrated both graphically and statistically. The improvement gained in the root-mean square of the self-calibration range residuals of 22 and 32%, respectively, indicates the model’s ability to compensate for the scattering error. A reduction in correlation between the camera position and rangefinder offset of up to 10% was achieved, which is consistent with previous findings. In addition, a systematic approach for designing the optimal separation between the foreground and background planes is presented.
Visual perception is regularly used by humans and robots for navigation. By either implicitly or ... more Visual perception is regularly used by humans and robots for navigation. By either implicitly or explicitly mapping the environment, ego-motion can be determined and a path of actions can be planned. The process of mapping and navigation are delicately intertwined; therefore, improving one can often lead to an improvement of the other. Both processes are sensitive to the interior orientation parameters of the camera system and mathematically modelling these systematic errors can often improve the precision and accuracy of the overall solution. This paper presents an automatic camera calibration method suitable for any lens, without having prior knowledge about the sensor. Statistical inference is performed to map the environment and localize the camera simultaneously. K-nearest neighbour regression is used to model the geometric distortions of the images. A normal-angle lens Nikon camera and wide-angle lens GoPro camera were calibrated using the proposed method, as well as the conventional bundle adjustment with self-calibration method (for comparison). Results showed that the mapping error was reduced from an average of 14.9 mm to 1.2 mm (i.e. a 92% improvement) and 66.6 mm to 1.5 mm (i.e. a 98% improvement) using the proposed method for the Nikon and GoPro cameras, respectively. In contrast, the conventional approach achieved an average 3D error of 0.9 mm (i.e. 94% improvement) and 3.3 mm (i.e. 95% improvement) for the Nikon and GoPro cameras, respectively. Thus, the proposed method performs well irrespective of the lens/sensor used: it yields results that are comparable to the conventional approach for normal-angle lens cameras, and it has the additional benefit of improving calibration results for wide-angle lens cameras.
X-ray imaging is a fundamental tool of routine clinical diagnosis. Fluoroscopic imaging can furth... more X-ray imaging is a fundamental tool of routine clinical diagnosis. Fluoroscopic imaging can further acquire X-ray images at video frame rates, thus enabling non-invasive in-vivo motion studies of joints, gastrointestinal tract, etc. For both the qualitative and quantitative analysis of static and dynamic X-ray images, the data should be free of systematic biases. Besides precise fabrication of hardware, software-based calibration solutions are commonly used for modelling the distortions. In this primary research study, a robust photogrammetric bundle adjustment was used to model the projective geometry of two fluoroscopic X-ray imaging systems. However, instead of relying on an expert photogrammetrist's knowledge and judgement to decide on a parametric model for describing the systematic errors, a self-tuning data-driven approach is used to model the complex non-linear distortion profile of the sensors. Quality control from the experiment showed that 0.06 mm to 0.09 mm 3D reconstruction accuracy was achievable post-calibration using merely 15 X-ray images. As part of the bundle adjustment, the location of the virtual fluoroscopic system relative to the target field can also be spatially resected with an RMSE between 3.10 mm and 3.31 mm.
Inertial measurement units (IMUs) are fundamental for attitude control of drones. With the advanc... more Inertial measurement units (IMUs) are fundamental for attitude control of drones. With the advancements in micro-electro-mechanical systems (MEMS) fabrication processes, size, power consumption, and price of these sensors have reduced significantly and attracted many new applications. However, this came at the expense of sensors requiring frequent recalibration, as they are highly contaminated with systematic errors. This paper presents a novel method to jointly calibrate the accelerometer, gyroscope, and magnetometer triad in a MEMS IMU without additional equipment. Opportunistic zero change in velocity and position updates, and inclination updates were used in conjunction with relative orientation updates from magnetometers in a robust batch least-squares adjustment. Solutions from the proposed self-calibration were compared to existing calibration methods. Empirical results suggest that the new method is robust against magnetic distortions and can achieve performance similar to a specialized calibration that uses a more accurate (and expensive) IMU as reference. The jointly estimated accelerometer and gyroscope calibration parameters can deliver a more accurate dead-reckoning solution than the popular multi-position calibration method (i.e., 54% improvement in orientation accuracy) by recovering the gyroscope scale error and other systematic errors. In addition, it can improve parameter observability as well as reduce calibration time by incorporating dynamic data with static orientations. The proposed calibration was also applied on-site pre-mission by simply waving the sensor by hand and was able to improve the orientation tracking accuracy by 73%.
In the absence of external reference position information (e.g. surveyed targets or Global Naviga... more In the absence of external reference position information (e.g. surveyed targets or Global Navigation Satellite Systems) Simultaneous Localization and Mapping (SLAM) has proven to be an effective method for indoor navigation. The positioning drift can be reduced with regular loop-closures and global relaxation as the backend, thus achieving a good balance between exploration and exploitation. Although vision-based systems like laser scanners are typically deployed for SLAM, these sensors are heavy, energy inefficient, and expensive, making them unattractive for wearables or smartphone applications. However, the concept of SLAM can be extended to non-optical systems such as magnetometers. Instead of matching features such as walls and furniture using some variation of the Iterative Closest Point algorithm, the local magnetic field can be matched to provide loop-closure and global trajectory updates in a Gaussian Process (GP) SLAM framework. With a MEMS-based inertial measurement unit providing a continuous trajectory, and the matching of locally distinct magnetic field maps, experimental results in this paper show that a drift-free navigation solution in an indoor environment with millimetre-level accuracy can be achieved. The GP-SLAM approach presented can be formulated as a maximum a posteriori estimation problem and it can naturally perform loop-detection, feature-to-feature distance minimization, global trajectory optimization, and magnetic field map estimation simultaneously. Spatially continuous features (i.e. smooth magnetic field signatures) are used instead of discrete feature correspondences (e.g. point-to-point) as in conventional vision-based SLAM. These position updates from the ambient magnetic field also provide enough information for calibrating the accelerometer bias and gyroscope bias in-use. The only restriction for this method is the need for magnetic disturbances (which is typically not an issue for indoor environments); however, no assumptions are required for the general motion of the sensor (e.g. static periods).
Sensor fusion of a MEMS IMU with a magnetometer is a popular system design, because such 9-DoF (d... more Sensor fusion of a MEMS IMU with a magnetometer is a popular system design, because such 9-DoF (degrees of freedom) systems are capable of achieving drift-free 3D orientation tracking. However, these systems are often vulnerable to ambient magnetic distortions and lack useful position information; in the absence of external position aiding (e.g. satellite/ultra-wideband positioning systems) the dead-reckoned position accuracy from a 9-DoF MEMS IMU deteriorates rapidly due to unmodelled errors. Positioning information is valuable in many satellite-denied geomatics applications (e.g. indoor navigation, location-based services, etc.). This paper proposes an improved 9-DoF IMU indoor pose tracking method using batch optimization. By adopting a robust in-situ user self-calibration approach to model the systematic errors of the accelerometer, gyroscope, and magnetometer simultaneously in a tightly-coupled post-processed least-squares framework, the accuracy of the estimated trajectory from a 9-DoF MEMS IMU can be improved. Through a combination of relative magnetic measurement updates and a robust weight function, the method is able to tolerate a high level of magnetic distortions. The proposed auto-calibration method was tested in-use under various heterogeneous magnetic field conditions to mimic a person walking with the sensor in their pocket, a person checking their phone, and a person walking with a smartwatch. In these experiments, the presented algorithm improved the in-situ dead-reckoning orientation accuracy by 79.8 – 89.5% and the dead-reckoned positioning accuracy by 72.9 – 92.8%, thus reducing the relative positioning error from metre-level to decimetre-level after ten seconds of integration, without making assumptions about the user's dynamics.
Data Revues 18770568 Unassign S1877056814001935, Jan 10, 2014
Measuring dynamic in vivo shoulder kinematics is crucial to better understanding numerous patholo... more Measuring dynamic in vivo shoulder kinematics is crucial to better understanding numerous pathologies. Motion capture systems using skin-mounted markers offer good solutions for non-invasive assessment of shoulder kinematics during dynamic movement. However, none of the current motion capture techniques have been used to study translation values at the joint, which is crucial to assess shoulder instability. The aim of the present study was to develop a dedicated patient-specific measurement technique based on motion capture and magnetic resonance imaging (MRI) to determine shoulder kinematics accurately. Estimation of both rotations and translations at the shoulder joint using motion capture is feasible thanks to a patient-specific kinematic chain of the shoulder complex reconstructed from MRI data. We implemented a patient-specific kinematic chain model of the shoulder complex with loose constraints on joint translation. To assess the effectiveness of the technique, six subjects underwent data acquisition simultaneously with fluoroscopy and motion capture during flexion and empty-can abduction. The reference 3D shoulder kinematics was reconstructed from fluoroscopy and compared to that obtained from the new technique using skin markers. Root mean square errors (RMSE) for shoulder orientation were within 4° (mean range: 2.0°-3.4°) for each anatomical axis and each motion. For glenohumeral translations, maximum RMSE for flexion was 3.7mm and 3.5mm for empty-can abduction (mean range: 1.9-3.3mm). Although the translation errors were significant, the computed patterns of humeral translation showed good agreement with published data. To our knowledge, this study is the first attempt to calculate both rotations and translations at the shoulder joint based on skin-mounted markers. Results were encouraging and can serve as reference for future developments. The proposed technique could provide valuable kinematic data for the study of shoulder pathologies. Basic Science Study.
Orthopaedics & traumatology, surgery & research : OTSR, 2014
Measuring dynamic in vivo shoulder kinematics is crucial to better understanding numerous patholo... more Measuring dynamic in vivo shoulder kinematics is crucial to better understanding numerous pathologies. Motion capture systems using skin-mounted markers offer good solutions for non-invasive assessment of shoulder kinematics during dynamic movement. However, none of the current motion capture techniques have been used to study translation values at the joint, which is crucial to assess shoulder instability. The aim of the present study was to develop a dedicated patient-specific measurement technique based on motion capture and magnetic resonance imaging (MRI) to determine shoulder kinematics accurately. Estimation of both rotations and translations at the shoulder joint using motion capture is feasible thanks to a patient-specific kinematic chain of the shoulder complex reconstructed from MRI data. We implemented a patient-specific kinematic chain model of the shoulder complex with loose constraints on joint translation. To assess the effectiveness of the technique, six subjects un...
This paper focuses on the use of control linear features extraction from a terrestrial laser scan... more This paper focuses on the use of control linear features extraction from a terrestrial laser scanning (TLS) surface to produce a good datum definition for a close range photogrammetric model. The difficulty of identifying conjugate points between both datasets is overcome by the derivation of automatic features extraction. Hence, the extracted features from the terrestrial measurements represent a good candidate for co-registration. Field work is performed to validate the proposed methods. It combines a TLS acquisition on different city buildings, while a complete dataset of close range images of the same area is measured. Some additional geodetic survey control is added to the overall data to perform global registration. Those experiments are based on planar patches automatic recognition derived from segmentation techniques. Planar segmentation is the extraction of planes from a point cloud. Assuming that linear features are at the intersection of those planes, we are able to gener...
Over the last few years, LiDAR has become a popular technology for the direct acquisition of topo... more Over the last few years, LiDAR has become a popular technology for the direct acquisition of topographic information. In spite of the increasing utilization of this technology in several applications, its accuracy potential has not been fully explored. Most of current LiDAR calibration techniques are based on empirical and proprietary procedures that demand the system’s raw measurements, which may not be always available to the end-user. As a result, we can still observe systematic discrepancies between conjugate surface elements in overlapping LiDAR strips. In this paper, two alternative calibration procedures that overcome the existing limitations are introduced. The first procedure, denoted as “Simplified method”, makes use of the LiDAR point cloud from parallel LiDAR strips acquired by a steady platform (e.g., fixed wing aircraft) over an area with moderately varying elevation. The second procedure, denoted as “Quasi-rigorous method”, can deal with non-parallel strips, but requi...
Current advances in digital and electronic products have led to the availability of inexpensive a... more Current advances in digital and electronic products have led to the availability of inexpensive and reliable Medium Format Digital Cameras (MFDCs) that can be used in many photogrammetric applications. In this research, the impact of camera and system calibration on object space reconstruction is investigated under different georeferencing scenarios (i.e., indirect georeferencing and integrated sensor orientation). First, camera calibration is conducted using a MFDC (i.e., the Rollei-P65). Based on different camera calibration datasets – such as indoor, in-situ, and camera calibration certificates, the equivalency of the calibration techniques as well as the adequacy of the distortion models are evaluated while considering relative and absolute quantitative measures. Previously developed camera stability analysis technique will be used for testing the adequacy of the utilized distortion model as well as the equivalency of different calibration techniques. Afterwards, system calibrat...
ABSTRACT The geometric calibration of time-of-flight range cameras is a necessary quality assuran... more ABSTRACT The geometric calibration of time-of-flight range cameras is a necessary quality assurance measure performed to estimate the interior orientation parameters. Self-calibration from a network of range imagery of an array of signalized targets arranged in one or two planes can be used for this purpose. The latter configuration requires the addition of a parametric model for internal light scattering biases in the range observations to the background plane due to the presence of the foreground plane. In a previous study of MESA Imaging SwissRanger range cameras, such a model was developed and shown to be effective. A new parametric model is proposed here because the scattering error behavior is camera model dependent. The new model was tested on two pmdtechnologies range cameras, the CamCube 3.0 and CamBoard nano, and its effectiveness was demonstrated both graphically and statistically. The improvement gained in the root-mean square of the self-calibration range residuals of 22 and 32%, respectively, indicates the model’s ability to compensate for the scattering error. A reduction in correlation between the camera position and rangefinder offset of up to 10% was achieved, which is consistent with previous findings. In addition, a systematic approach for designing the optimal separation between the foreground and background planes is presented.
Corporate insolvency can have a devastating effect on the economy. With an increasing number of c... more Corporate insolvency can have a devastating effect on the economy. With an increasing number of companies making expansion overseas to capitalize on foreign resources, a multinational corporate bankruptcy can disrupt the world’s financial ecosystem. Corporations do not fail instantaneously; objective measures and rigorous analysis of qualitative (e.g. brand) and quantitative (e.g. econometric factors) data can help identify a company’s financial risk. Gathering and storage of data about a corporation has become less difficult with recent advancements in communication and information technologies. The remaining challenge lies in mining relevant information about a company’s health hidden under the vast amounts of data, and using it to forecast insolvency so that managers and stakeholders have time to react. In recent years, machine learning has become a popular field in big data analytics because of its success in learning complicated models. Methods such as support vector machines, adaptive boosting, artificial neural networks, and Gaussian processes can be used for recognizing patterns in the data (with a high degree of accuracy) that may not be apparent to human analysts. This thesis studied corporate bankruptcy of manufacturing companies in Korea and Poland using experts’ opinions and financial measures, respectively. Using publicly available datasets, several machine learning methods were applied to learn the relationship between the company’s current state and its fate in the near future. Results showed that predictions with accuracy greater than 95% were achievable using any machine learning technique when informative features like experts’ assessment were used. However, when using purely financial factors to predict whether or not a company will go bankrupt, the correlation is not as strong. More features are required to better describe the data, but this results in a higher dimensional problem where the thousands of published companies’ data are insufficient to populate this space with high enough density. Due to this “curse of dimensionality”, flexible nonlinear models tend to over-fit to the training samples and thus fail to generalize to unseen data. For the high-dimensional Polish bankruptcy dataset, simpler models such as logistic regression could forecast a company’s bankruptcy one year into the future with 66.4% accuracy.
Uploads