The Experts below are selected from a list of 318 Experts worldwide ranked by ideXlab platform
Alina Zare - One of the best experts on this subject based on the ideXlab platform.
-
Multiple Instance Hybrid Estimator for Learning Target Signatures
arXiv: Computer Vision and Pattern Recognition, 2017Co-Authors: Changzhe Jiao, Alina ZareAbstract:Signature-based detectors for hyperspectral Target detection rely on knowing the specific Target signature in advance. However, Target signature are often difficult or impossible to obtain. Furthermore, common methods for obtaining Target signatures, such as from laboratory measurements or manual selection from an image scene, usually do not capture the discriminative features of Target class. In this paper, an approach for estimating a discriminative Target signature from imprecise labels is presented. The proposed approach maximizes the response of the hybrid sub-pixel detector within a multiple instance Learning framework and estimates a set of discriminative Target signatures. After Learning Target signatures, any signature based detector can then be applied on test data. Both simulated and real hyperspectral Target detection experiments are shown to illustrate the effectiveness of the method.
-
IGARSS - Multiple instance hybrid estimator for Learning Target signatures
2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), 2017Co-Authors: Changzhe Jiao, Alina ZareAbstract:Signature-based detectors for hyperspectral Target detection rely on knowing the specific Target signature in advance. However, Target signatures are often difficult or impossible to obtain. Furthermore, common methods for obtaining Target signatures, such as from laboratory measurements or manual selection from an image scene, usually do not capture the discriminative features of Target class. In this paper, an approach for estimating a discriminative Target signature from imprecise labels is presented. The proposed approach maximizes the response of the hybrid sub-pixel detector within a multiple instance Learning framework and estimates a set of discriminative Target signatures. After Learning Target signatures, any signature based detector can then be applied on test data. Both simulated and real hyperspectral Target detection experiments are shown to illustrate the effectiveness of the method.
-
Functions of Multiple Instances for Learning Target Signatures
IEEE Transactions on Geoscience and Remote Sensing, 2015Co-Authors: Changzhe Jiao, Alina ZareAbstract:The functions of multiple instances (FUMI) approach for Learning Target and nonTarget signatures is introduced. FUMI is a generalization of the multiple-instance Learning (MIL) approach for supervised Learning. FUMI differs significantly from standard MIL and supervised Learning approaches because only data points which are functions of class concepts/signatures are available. In particular, this paper addresses the problem in which data points are convex combinations of Target and nonTarget signatures. Two algorithms, convex FUMI ( $c$ FUMI) and extended $c$ FUMI ( $e$ FUMI) , are presented and applied to the problem of hyperspectral unmixing and Target detection. $c$ FUMI learns Target and nonTarget signatures (i.e., Target and nonTarget endmembers) , the number of nonTarget signatures, and the proportion of each signature for every data point. The $e$ FUMI algorithm extends the $c$ FUMI to allow for additional “bag level” uncertainty in training labels. For these methods, training data need only binary labels indicating whether a data point (or some spatial area in the case of $e$ FUMI) contains or does not contain some proportion of the Target; the specific Target proportions for the training data are not needed. After Learning the Target signature using the binary-labeled training data, Target detection can be performed on test data. Results for subpixel Target detection on simulated and real airborne hyperspectral data are shown.
Changzhe Jiao - One of the best experts on this subject based on the ideXlab platform.
-
Multiple Instance Hybrid Estimator for Learning Target Signatures
arXiv: Computer Vision and Pattern Recognition, 2017Co-Authors: Changzhe Jiao, Alina ZareAbstract:Signature-based detectors for hyperspectral Target detection rely on knowing the specific Target signature in advance. However, Target signature are often difficult or impossible to obtain. Furthermore, common methods for obtaining Target signatures, such as from laboratory measurements or manual selection from an image scene, usually do not capture the discriminative features of Target class. In this paper, an approach for estimating a discriminative Target signature from imprecise labels is presented. The proposed approach maximizes the response of the hybrid sub-pixel detector within a multiple instance Learning framework and estimates a set of discriminative Target signatures. After Learning Target signatures, any signature based detector can then be applied on test data. Both simulated and real hyperspectral Target detection experiments are shown to illustrate the effectiveness of the method.
-
IGARSS - Multiple instance hybrid estimator for Learning Target signatures
2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), 2017Co-Authors: Changzhe Jiao, Alina ZareAbstract:Signature-based detectors for hyperspectral Target detection rely on knowing the specific Target signature in advance. However, Target signatures are often difficult or impossible to obtain. Furthermore, common methods for obtaining Target signatures, such as from laboratory measurements or manual selection from an image scene, usually do not capture the discriminative features of Target class. In this paper, an approach for estimating a discriminative Target signature from imprecise labels is presented. The proposed approach maximizes the response of the hybrid sub-pixel detector within a multiple instance Learning framework and estimates a set of discriminative Target signatures. After Learning Target signatures, any signature based detector can then be applied on test data. Both simulated and real hyperspectral Target detection experiments are shown to illustrate the effectiveness of the method.
-
Functions of Multiple Instances for Learning Target Signatures
IEEE Transactions on Geoscience and Remote Sensing, 2015Co-Authors: Changzhe Jiao, Alina ZareAbstract:The functions of multiple instances (FUMI) approach for Learning Target and nonTarget signatures is introduced. FUMI is a generalization of the multiple-instance Learning (MIL) approach for supervised Learning. FUMI differs significantly from standard MIL and supervised Learning approaches because only data points which are functions of class concepts/signatures are available. In particular, this paper addresses the problem in which data points are convex combinations of Target and nonTarget signatures. Two algorithms, convex FUMI ( $c$ FUMI) and extended $c$ FUMI ( $e$ FUMI) , are presented and applied to the problem of hyperspectral unmixing and Target detection. $c$ FUMI learns Target and nonTarget signatures (i.e., Target and nonTarget endmembers) , the number of nonTarget signatures, and the proportion of each signature for every data point. The $e$ FUMI algorithm extends the $c$ FUMI to allow for additional “bag level” uncertainty in training labels. For these methods, training data need only binary labels indicating whether a data point (or some spatial area in the case of $e$ FUMI) contains or does not contain some proportion of the Target; the specific Target proportions for the training data are not needed. After Learning the Target signature using the binary-labeled training data, Target detection can be performed on test data. Results for subpixel Target detection on simulated and real airborne hyperspectral data are shown.
Peter Corke - One of the best experts on this subject based on the ideXlab platform.
-
Towards vision-based deep reinforcement Learning for robotic motion control
Australasian Conference on Robotics and Automation ACRA, 2015Co-Authors: Fangyi Zhang, Michael J. Milford, Jurgen Leitner, Ben Upcroft, Peter CorkeAbstract:This paper introduces a machine Learning based system for controlling a robotic manipulator with visual perception only. The capability to autonomously learn robot controllers solely from raw-pixel images and without any prior knowledge of configuration is shown for the first time. We build upon the success of recent deep reinforcement Learning and develop a system for Learning Target reaching with a three-joint robot manipulator using external visual observation. A Deep Q Network (DQN) was demonstrated to perform Target reaching after training in simulation. Transferring the network to real hardware and real observation in a naive approach failed, but experiments show that the network works when replacing camera images with synthetic images.
Mohamed Fares - One of the best experts on this subject based on the ideXlab platform.
-
Multiple virtual screening approaches for finding new Hepatitis c virus RNA-dependent RNA polymerase inhibitors: Structure-based screens and molecular dynamics for the pursue of new poly pharmacological inhibitors
BMC Bioinformatics, 2012Co-Authors: Mahmoud Elhefnawi, Mohammad Elgamacy, Mohamed FaresAbstract:The RNA polymerase NS5B of Hepatitis C virus (HCV) is a well-characterised drug Target with an active site and four allosteric binding sites. This work presents a workflow for virtual screening and its application to Drug Bank screening Targeting the Hepatitis C Virus (HCV) RNA polymerase non-nucleoside binding sites. Potential polypharmacological drugs are sought with predicted active inhibition on viral replication, and with proven positive pharmaco-clinical profiles. The approach adopted was receptor-based. Docking screens, guided with contact pharmacophores and neural-network activity prediction models on all allosteric binding sites and MD simulations, constituted our analysis workflow for identification of potential hits. Steps included: 1) using a two-phase docking screen with Surflex and Glide Xp. 2) Ranking based on scores, and important H interactions. 3) a machine-Learning Target-trained artificial neural network PIC prediction model used for ranking. This provided a better correlation of IC50 values of the training sets for each site with different docking scores and sub-scores. 4) interaction pharmacophores-through retrospective analysis of protein-inhibitor complex X-ray structures for the interaction pharmacophore (common interaction modes) of inhibitors for the five non-nucleoside binding sites were constructed. These were used for filtering the hits according to the critical binding feature of formerly reported inhibitors. This filtration process resulted in identification of potential new inhibitors as well as formerly reported ones for the thumb II and Palm I sites (HCV-81) NS5B binding sites. Eventually molecular dynamics simulations were carried out, confirming the binding hypothesis and resulting in 4 hits.
Fangyi Zhang - One of the best experts on this subject based on the ideXlab platform.
-
Towards vision-based deep reinforcement Learning for robotic motion control
Australasian Conference on Robotics and Automation ACRA, 2015Co-Authors: Fangyi Zhang, Michael J. Milford, Jurgen Leitner, Ben Upcroft, Peter CorkeAbstract:This paper introduces a machine Learning based system for controlling a robotic manipulator with visual perception only. The capability to autonomously learn robot controllers solely from raw-pixel images and without any prior knowledge of configuration is shown for the first time. We build upon the success of recent deep reinforcement Learning and develop a system for Learning Target reaching with a three-joint robot manipulator using external visual observation. A Deep Q Network (DQN) was demonstrated to perform Target reaching after training in simulation. Transferring the network to real hardware and real observation in a naive approach failed, but experiments show that the network works when replacing camera images with synthetic images.