Computer vision-based technologies play a key role in precision livestock farming, and video-base... more Computer vision-based technologies play a key role in precision livestock farming, and video-based analysis approaches have been advocated as useful tools for automatic animal monitoring, behavior analysis, and efficient welfare measurement management. Accurately and efficiently segmenting animals’ contours from their backgrounds is a prerequisite for vision-based technologies. Deep learning-based segmentation methods have shown good performance through training models on a large amount of pixel-labeled images. However, it is challenging and time-consuming to label animal images due to their irregular contours and changing postures. In order to reduce the reliance on the number of labeled images, one-shot learning with a pseudo-labeling approach is proposed using only one labeled image frame to segment animals in videos. The proposed approach is mainly comprised of an Xception-based Fully Convolutional Neural Network (Xception-FCN) module and a pseudo-labeling (PL) module. Xception-...
Deep learning-based video segmentation methods can offer a good performance after being trained o... more Deep learning-based video segmentation methods can offer a good performance after being trained on the large-scale pixel labeled datasets. However, a pixel-wise manual labeling of animal images is challenging and time consuming due to irregular contours and motion blur. To achieve desirable tradeoffs between the accuracy and speed, a novel one-shot learning-based approach is proposed in this article to segment animal video with only one labeled frame. The proposed approach consists of the following three main modules: guidance frame selection utilizes “BubbleNet” to choose one frame for manual labeling, which can leverage the fine-tuning effects of the only labeled frame; Xception-based fully convolutional network localizes dense prediction using depthwise separable convolutions based on one single labeled frame; and postprocessing is used to remove outliers and sharpen object contours, which consists of two submodules—test time augmentation and conditional random field. Extensive experiments have been conducted on the DAVIS 2016 animal dataset. Our proposed video segmentation approach achieved mean intersection-over-union score of 89.5% on the DAVIS 2016 animal dataset with less run time, and outperformed the state-of-art methods (OSVOS and OSMN). The proposed one-shot learning-based approach achieves real-time and automatic segmentation of animals with only one labeled video frame. This can be potentially used further as a baseline for intelligent perception-based monitoring of animals and other domain-specific applications.11The source code, datasets, and pre-trained weights for this work are publicly [Online]. Available: https://github.com/tengfeixue-victor/One-Shot-Animal-Video-Segmentation.
Hyperspectral line-scan cameras are increasingly being deployed on mobile platforms operating in ... more Hyperspectral line-scan cameras are increasingly being deployed on mobile platforms operating in unstructured environments. To generate geometrically accurate hyperspectral composites, the intrinsic parameters of these cameras must be resolved. This article describes a method for determining the intrinsic parameters of a hyperspectral line-scan camera. The proposed method is based on a cross-ratio invariant calibration routine and is able to estimate the focal length, principal point, and radial distortion parameters in a hyperspectral line-scan camera. Compared to previous methods that use similar calibration targets, our approach extends the camera model to include radial distortion. It is able to utilize calibration data recorded from multiple camera view angles by optimizing the re-projection error of all calibration data jointly. The proposed method also includes an additional signal processing step that automatically detects calibration points in hyperspectral imagery of the calibration target. These contributions result in accurate estimates of the intrinsic parameters with minimal supervision. The proposed method is validated through comprehensive simulation and demonstrated on real hyperspectral line-scans.
2011 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2011
This paper presents and compares two different approaches to integrating sensor information from ... more This paper presents and compares two different approaches to integrating sensor information from an Iner- tial Measuring Unit (IMU), Global Positioning System (GPS) receiver and monocular vision camera mounted to a low- flying Unmanned Aerial Vehicle (UAV) for building large-scale 3D terrain reconstructions. Both approaches utilise a statisti- cally optimal bundle adjustment formulation that incorporates Vision, IMU and GPS observations
This paper presents a 3D online path planning algorithm for a 6DOF Rotary Unmanned Aerial Vehicle... more This paper presents a 3D online path planning algorithm for a 6DOF Rotary Unmanned Aerial Vehicle (RUAV) operating in a cluttered environment using a Gaussian Process (GP) occupancy map. Traditional grid-based occupancy maps suffer from the curse of dimensionality for platforms that operate in a high dimensional configuration space. In addition, the grid resolutions, shapes and orientations limit the flexibility of the path planner. A GP occupancy map however performs inference directly on the collected sensor data, removing the need of maintaining a full discrete map. It performs inference about the probability of collision for any query point in continuous 3D space that potentially maximizes the flexibility of the path planner. We embed the GP map approach into our previous work on smooth path planning and control for an aerial vehicle and present results in a simulated cluttered environment.
This paper presents a fusion method to combine aerial images from a low flying Unmanned Aerial Ve... more This paper presents a fusion method to combine aerial images from a low flying Unmanned Aerial Vehicle (UAV) with images of other spectral bands from sources such as satellites or commercial hyperspectral imagers. The proposed method propagates information from high-resolution images into other low-resolution modalities while allowing the images to have different spectral channels. This means the relationship between the high-resolution and low-resolution channels is expected to be non-deterministic, non-linear and non-stationary. A novel Gaussian Process (GP) framework was developed to define a stochastic prior over the estimated images. Its covariance function is computed to replicate the local structure of the high-resolution image, and allows the model to infer a high-resolution estimate from a low-resolution channel. Results are presented for natural images acquired by a UAV in a farmland mapping application.
2005 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2005
... intelligent guidance scheme for a UAV for improving navigation integrity when oper-ating over... more ... intelligent guidance scheme for a UAV for improving navigation integrity when oper-ating over unknown terrain ... MWMG Dissanayake, P. Newman, S. Clark, HF Durrant-Whyte, M. Csorba, ASolution to the Simultaneous Localization and Map Building (SLAM) Problem, IEEE ...
2003 IEEE International Conference on Robotics and Automation (Cat. No.03CH37422)
This paper presents results of the application of simultaneous localisation and map building (SLA... more This paper presents results of the application of simultaneous localisation and map building (SLAM) for an uninhabited aerial vehicle (UAV). Single vision camera and inertial measurement unit (IMU) are installed in a UAV platform. The data taken from a flight test is used to run the SLAM algorithm. Results show that both the map and the vehicle uncertainty are corrected even though the model of the system and observation are highly non-linear. The results, however, also indicate that further work of observability and the relationship between vehicle model drift and the number and the location of landmarks need to be further analysed given the highly dynamic nature of the system.
The ability of robots to meticulously cover large areas while gathering sensor data has widesprea... more The ability of robots to meticulously cover large areas while gathering sensor data has widespread applications in precision agriculture. For autonomous operations in orchards, a suitable information management system is required, within which we can gather and process data relating to the state and performance of the crop over time, such as distinct yield count, canopy volume, and crop health. An efficient way to structure an information system is to discretize it to the individual tree, for which tree segmentation/detection is a key component. This paper presents a tree trunk detection pipeline for identifying individual trees in a trellis structured apple orchard, using ground-based lidar and image data. A coarse observation of trunk candidates is initially made using a Hough transformation on point cloud lidar data. These candidates are projected into the camera images, where pixelwise classification is used to update their likelihood of being a tree trunk. Detection is achieved by using a hidden semi-Markov model to leverage from contextual information provided by the repetitive structure of an orchard. By repeating this over individual orchard rows, we are able to build a tree map over the farm, which can be either GPS localized or represented topologically by the row and tree number. The pipeline was evaluated at a commercial apple orchard near Melbourne, Australia. Data were collected at different times of year, covering an area of 1.6 ha containing different apple varieties planted on two types of trellis systems: a vertical I-trellis structure and a Güttingen V-trellis structure. The results show good trunk detection performance for both apple varieties and trellis structures during the preharvest season (87-96 % accuracy) and near perfect trunk detection performance (99% accuracy) during the flowering season.
Proceedings 2007 IEEE International Conference on Robotics and Automation, 2007
This paper presents an algorithm which can effectively constrain inertial navigation drift using ... more This paper presents an algorithm which can effectively constrain inertial navigation drift using monocular camera data. It is capable of operating in unknown and large scale environments and assumes no prior knowledge of the size, appearance or location of potential environmental features. Low cost inertial navigation units are found on most autonomous vehicles and a large number of smaller robots. Depending on the grade of the sensor, when used alone, inertial data for control and navigation will only be reliable for a matter of seconds or minutes. An algorithm is presented that simultaneously estimates relative feature location in sensor space and inertial position, velocity and attitude in world coordinates. Feature locations are maintained in sensor space to ensure measurement linearity. Image depth is represented by an inverse function which permits un-delayed feature initialization and improves linearity and convergence. It is shown that the resulting navigation solution is able to be constrained, providing results comparable to inertial-GPS systems. Results are presented for an autonomous aircraft operating in a large semi-structured environment.
2008 IEEE International Conference on Robotics and Automation, 2008
This paper identifies various scale factor biases commonly introduced into monocular SLAM impleme... more This paper identifies various scale factor biases commonly introduced into monocular SLAM implementations as a result of the true scale factor of the map not being observable. A way to make the scale factor observable and remove any scale biases via the use of an inertial measurement unit (IMU) is presented and implemented. Results show that with an IMU the true scale of the map becomes observable over time and the use of a square root information filter allows the effect of initial scale biases to be removed completely from the solution resulting in an unbiased solution no matter what the initial scale assumptions are.
approach to tree recognition and localization in orchard environments for tree-crop applications.... more approach to tree recognition and localization in orchard environments for tree-crop applications. The primary objective is to develop a pipeline for building detailed orchard maps and an algorithm to match subsequent lidar tree scans to the prior database, enabling correct data association for precision agricultural applications. Although global positioning systems (GPS) offer a simple solution, they are often unreliable in canopied environments due to satellite occlusion. The proposed method builds on the natural structure of the orchard. Lidar data are first segmented into individual trees using a hidden semi-Markov model. Then a descriptor for representing the characteristics or appearance of each tree is introduced, allowing a hidden Markov model based matching method to associate new observations with an existing map of the orchard. Short summary for practitioners Website Audiovisual material Links to other websites Additional comments Keywords Farming equipment and machinery Additional keywords Geographical location (NUTS) EU
In this paper we present an approach to tree recognition and localisation in orchard environments... more In this paper we present an approach to tree recognition and localisation in orchard environments for tree-crop applications. The method builds on the natural structure of the orchard by first segmenting the data into individual trees using a Hidden Semi-Markov Model. Second, a descriptor for representing the characteristics of the trees is introduced, allowing a Hidden Markov Model based matching method to associate new observations with an existing map of the orchard. The localisation method is evaluated on a dataset collected in an almond orchard, showing good performance and robustness both to segmentation errors and measurement noise.
2012 IEEE International Conference on Robotics and Automation, 2012
This paper introduces an unsupervised graph cut based object segmentation algorithm, ShadowCut, f... more This paper introduces an unsupervised graph cut based object segmentation algorithm, ShadowCut, for robotic aerial surveillance applications. By exploiting the spatial setting of the aerial imagery, ShadowCut algorithm differs from state-of-the-art object segmentation algorithms ([1] [2] [3] [4] [5]) by not requiring a large number of labelled training data set, nor constant user interaction ([6] [7] [8]). In this paper
Future unmanned aerial vehicle (UAV) applications will require high-accuracy localisation in envi... more Future unmanned aerial vehicle (UAV) applications will require high-accuracy localisation in environments in which navigation infrastructure such as the Global Positioning System (GPS) and prior terrain maps may be unavailable or unreliable. In these applications, long-term operation requires the vehicle to build up a spatial map of the environment while simultaneously localising itself within the map, a task known as
Abstract Text: Advances in technology could reduce time farmers spend on repetitive tasks. In pas... more Abstract Text: Advances in technology could reduce time farmers spend on repetitive tasks. In pasture-based systems, herding cows from grazing areas to the milking parlour is a repetitive task ideally suited to automation. We conducted a field study at Sydney University’s dairy farm to determine the behavioural response of dairy cows to a remotely operated unmanned ground vehicle (UGV) across time. Twenty milking cows were separated from the main herd at 0830 h and offered 0.5 ha of an ad-libitum kikuyu pasture allocation (50 kg DM/cow to ground level). A pre-defined figure eight route was determined for the UGV within this area. The UGV entered the pasture allocation at 0900 h and traversed this route at a target speed of 2.5 km/h, with the whole procedure repeated 6 times at 15 min intervals. The 0.5 ha was virtually split into four sectors. Four observers monitored cows exiting or entering each sector. Data were analysed by REML, where Cows out = Fixed (Robot (presence/absence) *...
In this paper we demonstrate a co-operative path-planning algorithm for multi-vehicle simultaneou... more In this paper we demonstrate a co-operative path-planning algorithm for multi-vehicle simultaneous localisation and mapping (SLAM) that uses information-based measures to maximize the accuracy of a feature map which is constructed from terrain observation made by each vehicle. The SLAM algorithm is distributed amongst the vehicles where each vehicle shares locally built map information via a central communications node. This information is used to assist in localisation which in turn increases the accuracy of the map information each vehicle provides. Each vehicle communicates to the central node potential trajectories it can take and the associated map information it will provide. The central communications node then co-ordinates the actions of each platform such as to maximise the accuracy of the globally constructed map. The vehicles each perform SLAM using a combination of on-board inertial sensors and an on-board terrain sensor. Results are presented using a 6-DoF simulation of several UAVs over an initially unexplored terrain.
2012 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2012
ABSTRACT This paper presents a novel approach to depth estimation from monocular images that is b... more ABSTRACT This paper presents a novel approach to depth estimation from monocular images that is based on the Gaussian Derivative Process (GDP) formulation. We use an inverse depth parametrisation and learn the mapping from image pixel coordinates to the inverse depth of the corresponding scene point given an estimate of the relative camera motion. We show that information about the geometry of the measurements can be integrated into Gaussian Process (GP) models and learnt jointly with the measurements. We provide a novel formulation of the inverse depth and its derivatives and learn their joint distribution. Experimental results are presented using synthesised examples and real monocular images captured from an Unmanned Aerial Vehicle (UAV). Results show improvement in depth estimation over standard Gaussian Process Regression (GPR). This improvement is presented by a reduction in the GP depth prediction errors and the predictive variance. Finally, we show mathematically that this improvement is due to the augmented derivative covariance terms and the correlations between the inverse depth and the derivatives.
The use of a multi-rotor unmanned air vehicle (UAV) in image acquisition tasks is promising for t... more The use of a multi-rotor unmanned air vehicle (UAV) in image acquisition tasks is promising for three-dimensional (3D) object modeling. Such an autonomous data acquisition system can be useful to handle the geometric complexity of objects such as trees and the inherent difficulties of image capture. In this paper, we address the problem of view planning for a camera-equipped multi-rotor UAV to acquire an adequate set of images that leads to more detailed and complete knowledge of the 3D tree model. The proposed algorithm based on shape-from-silhouette methods incorporates both expected new visual information and vehicle movement. Occupancy estimation for volumetric object model serves as a baseline measure of new information. The outlined approach determines next best views across the viewpoint space bounded by the sensor coverage and the capability of the UAV with minimal a priori knowledge of the object. Simulation studies conducted with virtual reality environments show the effectiveness of the algorithm.
Computer vision-based technologies play a key role in precision livestock farming, and video-base... more Computer vision-based technologies play a key role in precision livestock farming, and video-based analysis approaches have been advocated as useful tools for automatic animal monitoring, behavior analysis, and efficient welfare measurement management. Accurately and efficiently segmenting animals’ contours from their backgrounds is a prerequisite for vision-based technologies. Deep learning-based segmentation methods have shown good performance through training models on a large amount of pixel-labeled images. However, it is challenging and time-consuming to label animal images due to their irregular contours and changing postures. In order to reduce the reliance on the number of labeled images, one-shot learning with a pseudo-labeling approach is proposed using only one labeled image frame to segment animals in videos. The proposed approach is mainly comprised of an Xception-based Fully Convolutional Neural Network (Xception-FCN) module and a pseudo-labeling (PL) module. Xception-...
Deep learning-based video segmentation methods can offer a good performance after being trained o... more Deep learning-based video segmentation methods can offer a good performance after being trained on the large-scale pixel labeled datasets. However, a pixel-wise manual labeling of animal images is challenging and time consuming due to irregular contours and motion blur. To achieve desirable tradeoffs between the accuracy and speed, a novel one-shot learning-based approach is proposed in this article to segment animal video with only one labeled frame. The proposed approach consists of the following three main modules: guidance frame selection utilizes “BubbleNet” to choose one frame for manual labeling, which can leverage the fine-tuning effects of the only labeled frame; Xception-based fully convolutional network localizes dense prediction using depthwise separable convolutions based on one single labeled frame; and postprocessing is used to remove outliers and sharpen object contours, which consists of two submodules—test time augmentation and conditional random field. Extensive experiments have been conducted on the DAVIS 2016 animal dataset. Our proposed video segmentation approach achieved mean intersection-over-union score of 89.5% on the DAVIS 2016 animal dataset with less run time, and outperformed the state-of-art methods (OSVOS and OSMN). The proposed one-shot learning-based approach achieves real-time and automatic segmentation of animals with only one labeled video frame. This can be potentially used further as a baseline for intelligent perception-based monitoring of animals and other domain-specific applications.11The source code, datasets, and pre-trained weights for this work are publicly [Online]. Available: https://github.com/tengfeixue-victor/One-Shot-Animal-Video-Segmentation.
Hyperspectral line-scan cameras are increasingly being deployed on mobile platforms operating in ... more Hyperspectral line-scan cameras are increasingly being deployed on mobile platforms operating in unstructured environments. To generate geometrically accurate hyperspectral composites, the intrinsic parameters of these cameras must be resolved. This article describes a method for determining the intrinsic parameters of a hyperspectral line-scan camera. The proposed method is based on a cross-ratio invariant calibration routine and is able to estimate the focal length, principal point, and radial distortion parameters in a hyperspectral line-scan camera. Compared to previous methods that use similar calibration targets, our approach extends the camera model to include radial distortion. It is able to utilize calibration data recorded from multiple camera view angles by optimizing the re-projection error of all calibration data jointly. The proposed method also includes an additional signal processing step that automatically detects calibration points in hyperspectral imagery of the calibration target. These contributions result in accurate estimates of the intrinsic parameters with minimal supervision. The proposed method is validated through comprehensive simulation and demonstrated on real hyperspectral line-scans.
2011 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2011
This paper presents and compares two different approaches to integrating sensor information from ... more This paper presents and compares two different approaches to integrating sensor information from an Iner- tial Measuring Unit (IMU), Global Positioning System (GPS) receiver and monocular vision camera mounted to a low- flying Unmanned Aerial Vehicle (UAV) for building large-scale 3D terrain reconstructions. Both approaches utilise a statisti- cally optimal bundle adjustment formulation that incorporates Vision, IMU and GPS observations
This paper presents a 3D online path planning algorithm for a 6DOF Rotary Unmanned Aerial Vehicle... more This paper presents a 3D online path planning algorithm for a 6DOF Rotary Unmanned Aerial Vehicle (RUAV) operating in a cluttered environment using a Gaussian Process (GP) occupancy map. Traditional grid-based occupancy maps suffer from the curse of dimensionality for platforms that operate in a high dimensional configuration space. In addition, the grid resolutions, shapes and orientations limit the flexibility of the path planner. A GP occupancy map however performs inference directly on the collected sensor data, removing the need of maintaining a full discrete map. It performs inference about the probability of collision for any query point in continuous 3D space that potentially maximizes the flexibility of the path planner. We embed the GP map approach into our previous work on smooth path planning and control for an aerial vehicle and present results in a simulated cluttered environment.
This paper presents a fusion method to combine aerial images from a low flying Unmanned Aerial Ve... more This paper presents a fusion method to combine aerial images from a low flying Unmanned Aerial Vehicle (UAV) with images of other spectral bands from sources such as satellites or commercial hyperspectral imagers. The proposed method propagates information from high-resolution images into other low-resolution modalities while allowing the images to have different spectral channels. This means the relationship between the high-resolution and low-resolution channels is expected to be non-deterministic, non-linear and non-stationary. A novel Gaussian Process (GP) framework was developed to define a stochastic prior over the estimated images. Its covariance function is computed to replicate the local structure of the high-resolution image, and allows the model to infer a high-resolution estimate from a low-resolution channel. Results are presented for natural images acquired by a UAV in a farmland mapping application.
2005 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2005
... intelligent guidance scheme for a UAV for improving navigation integrity when oper-ating over... more ... intelligent guidance scheme for a UAV for improving navigation integrity when oper-ating over unknown terrain ... MWMG Dissanayake, P. Newman, S. Clark, HF Durrant-Whyte, M. Csorba, ASolution to the Simultaneous Localization and Map Building (SLAM) Problem, IEEE ...
2003 IEEE International Conference on Robotics and Automation (Cat. No.03CH37422)
This paper presents results of the application of simultaneous localisation and map building (SLA... more This paper presents results of the application of simultaneous localisation and map building (SLAM) for an uninhabited aerial vehicle (UAV). Single vision camera and inertial measurement unit (IMU) are installed in a UAV platform. The data taken from a flight test is used to run the SLAM algorithm. Results show that both the map and the vehicle uncertainty are corrected even though the model of the system and observation are highly non-linear. The results, however, also indicate that further work of observability and the relationship between vehicle model drift and the number and the location of landmarks need to be further analysed given the highly dynamic nature of the system.
The ability of robots to meticulously cover large areas while gathering sensor data has widesprea... more The ability of robots to meticulously cover large areas while gathering sensor data has widespread applications in precision agriculture. For autonomous operations in orchards, a suitable information management system is required, within which we can gather and process data relating to the state and performance of the crop over time, such as distinct yield count, canopy volume, and crop health. An efficient way to structure an information system is to discretize it to the individual tree, for which tree segmentation/detection is a key component. This paper presents a tree trunk detection pipeline for identifying individual trees in a trellis structured apple orchard, using ground-based lidar and image data. A coarse observation of trunk candidates is initially made using a Hough transformation on point cloud lidar data. These candidates are projected into the camera images, where pixelwise classification is used to update their likelihood of being a tree trunk. Detection is achieved by using a hidden semi-Markov model to leverage from contextual information provided by the repetitive structure of an orchard. By repeating this over individual orchard rows, we are able to build a tree map over the farm, which can be either GPS localized or represented topologically by the row and tree number. The pipeline was evaluated at a commercial apple orchard near Melbourne, Australia. Data were collected at different times of year, covering an area of 1.6 ha containing different apple varieties planted on two types of trellis systems: a vertical I-trellis structure and a Güttingen V-trellis structure. The results show good trunk detection performance for both apple varieties and trellis structures during the preharvest season (87-96 % accuracy) and near perfect trunk detection performance (99% accuracy) during the flowering season.
Proceedings 2007 IEEE International Conference on Robotics and Automation, 2007
This paper presents an algorithm which can effectively constrain inertial navigation drift using ... more This paper presents an algorithm which can effectively constrain inertial navigation drift using monocular camera data. It is capable of operating in unknown and large scale environments and assumes no prior knowledge of the size, appearance or location of potential environmental features. Low cost inertial navigation units are found on most autonomous vehicles and a large number of smaller robots. Depending on the grade of the sensor, when used alone, inertial data for control and navigation will only be reliable for a matter of seconds or minutes. An algorithm is presented that simultaneously estimates relative feature location in sensor space and inertial position, velocity and attitude in world coordinates. Feature locations are maintained in sensor space to ensure measurement linearity. Image depth is represented by an inverse function which permits un-delayed feature initialization and improves linearity and convergence. It is shown that the resulting navigation solution is able to be constrained, providing results comparable to inertial-GPS systems. Results are presented for an autonomous aircraft operating in a large semi-structured environment.
2008 IEEE International Conference on Robotics and Automation, 2008
This paper identifies various scale factor biases commonly introduced into monocular SLAM impleme... more This paper identifies various scale factor biases commonly introduced into monocular SLAM implementations as a result of the true scale factor of the map not being observable. A way to make the scale factor observable and remove any scale biases via the use of an inertial measurement unit (IMU) is presented and implemented. Results show that with an IMU the true scale of the map becomes observable over time and the use of a square root information filter allows the effect of initial scale biases to be removed completely from the solution resulting in an unbiased solution no matter what the initial scale assumptions are.
approach to tree recognition and localization in orchard environments for tree-crop applications.... more approach to tree recognition and localization in orchard environments for tree-crop applications. The primary objective is to develop a pipeline for building detailed orchard maps and an algorithm to match subsequent lidar tree scans to the prior database, enabling correct data association for precision agricultural applications. Although global positioning systems (GPS) offer a simple solution, they are often unreliable in canopied environments due to satellite occlusion. The proposed method builds on the natural structure of the orchard. Lidar data are first segmented into individual trees using a hidden semi-Markov model. Then a descriptor for representing the characteristics or appearance of each tree is introduced, allowing a hidden Markov model based matching method to associate new observations with an existing map of the orchard. Short summary for practitioners Website Audiovisual material Links to other websites Additional comments Keywords Farming equipment and machinery Additional keywords Geographical location (NUTS) EU
In this paper we present an approach to tree recognition and localisation in orchard environments... more In this paper we present an approach to tree recognition and localisation in orchard environments for tree-crop applications. The method builds on the natural structure of the orchard by first segmenting the data into individual trees using a Hidden Semi-Markov Model. Second, a descriptor for representing the characteristics of the trees is introduced, allowing a Hidden Markov Model based matching method to associate new observations with an existing map of the orchard. The localisation method is evaluated on a dataset collected in an almond orchard, showing good performance and robustness both to segmentation errors and measurement noise.
2012 IEEE International Conference on Robotics and Automation, 2012
This paper introduces an unsupervised graph cut based object segmentation algorithm, ShadowCut, f... more This paper introduces an unsupervised graph cut based object segmentation algorithm, ShadowCut, for robotic aerial surveillance applications. By exploiting the spatial setting of the aerial imagery, ShadowCut algorithm differs from state-of-the-art object segmentation algorithms ([1] [2] [3] [4] [5]) by not requiring a large number of labelled training data set, nor constant user interaction ([6] [7] [8]). In this paper
Future unmanned aerial vehicle (UAV) applications will require high-accuracy localisation in envi... more Future unmanned aerial vehicle (UAV) applications will require high-accuracy localisation in environments in which navigation infrastructure such as the Global Positioning System (GPS) and prior terrain maps may be unavailable or unreliable. In these applications, long-term operation requires the vehicle to build up a spatial map of the environment while simultaneously localising itself within the map, a task known as
Abstract Text: Advances in technology could reduce time farmers spend on repetitive tasks. In pas... more Abstract Text: Advances in technology could reduce time farmers spend on repetitive tasks. In pasture-based systems, herding cows from grazing areas to the milking parlour is a repetitive task ideally suited to automation. We conducted a field study at Sydney University’s dairy farm to determine the behavioural response of dairy cows to a remotely operated unmanned ground vehicle (UGV) across time. Twenty milking cows were separated from the main herd at 0830 h and offered 0.5 ha of an ad-libitum kikuyu pasture allocation (50 kg DM/cow to ground level). A pre-defined figure eight route was determined for the UGV within this area. The UGV entered the pasture allocation at 0900 h and traversed this route at a target speed of 2.5 km/h, with the whole procedure repeated 6 times at 15 min intervals. The 0.5 ha was virtually split into four sectors. Four observers monitored cows exiting or entering each sector. Data were analysed by REML, where Cows out = Fixed (Robot (presence/absence) *...
In this paper we demonstrate a co-operative path-planning algorithm for multi-vehicle simultaneou... more In this paper we demonstrate a co-operative path-planning algorithm for multi-vehicle simultaneous localisation and mapping (SLAM) that uses information-based measures to maximize the accuracy of a feature map which is constructed from terrain observation made by each vehicle. The SLAM algorithm is distributed amongst the vehicles where each vehicle shares locally built map information via a central communications node. This information is used to assist in localisation which in turn increases the accuracy of the map information each vehicle provides. Each vehicle communicates to the central node potential trajectories it can take and the associated map information it will provide. The central communications node then co-ordinates the actions of each platform such as to maximise the accuracy of the globally constructed map. The vehicles each perform SLAM using a combination of on-board inertial sensors and an on-board terrain sensor. Results are presented using a 6-DoF simulation of several UAVs over an initially unexplored terrain.
2012 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2012
ABSTRACT This paper presents a novel approach to depth estimation from monocular images that is b... more ABSTRACT This paper presents a novel approach to depth estimation from monocular images that is based on the Gaussian Derivative Process (GDP) formulation. We use an inverse depth parametrisation and learn the mapping from image pixel coordinates to the inverse depth of the corresponding scene point given an estimate of the relative camera motion. We show that information about the geometry of the measurements can be integrated into Gaussian Process (GP) models and learnt jointly with the measurements. We provide a novel formulation of the inverse depth and its derivatives and learn their joint distribution. Experimental results are presented using synthesised examples and real monocular images captured from an Unmanned Aerial Vehicle (UAV). Results show improvement in depth estimation over standard Gaussian Process Regression (GPR). This improvement is presented by a reduction in the GP depth prediction errors and the predictive variance. Finally, we show mathematically that this improvement is due to the augmented derivative covariance terms and the correlations between the inverse depth and the derivatives.
The use of a multi-rotor unmanned air vehicle (UAV) in image acquisition tasks is promising for t... more The use of a multi-rotor unmanned air vehicle (UAV) in image acquisition tasks is promising for three-dimensional (3D) object modeling. Such an autonomous data acquisition system can be useful to handle the geometric complexity of objects such as trees and the inherent difficulties of image capture. In this paper, we address the problem of view planning for a camera-equipped multi-rotor UAV to acquire an adequate set of images that leads to more detailed and complete knowledge of the 3D tree model. The proposed algorithm based on shape-from-silhouette methods incorporates both expected new visual information and vehicle movement. Occupancy estimation for volumetric object model serves as a baseline measure of new information. The outlined approach determines next best views across the viewpoint space bounded by the sensor coverage and the capability of the UAV with minimal a priori knowledge of the object. Simulation studies conducted with virtual reality environments show the effectiveness of the algorithm.
Uploads
Papers by Salah Sukkarieh