533 results about "Ground truth" patented technology

A method for identifying a botnet in a network, including analyzing historical network data using a pre-determined heuristic to determine values of a feature in the historical network data, obtaining a ground truth data set having labels assigned to data units in the historical network data identifying known malicious nodes in the network, analyzing the historical network data and the ground truth data set using a machine learning algorithm to generate a model representing the labels as a function of the values of the feature, analyzing real-time network data using the pre-determined heuristic to determine a value of the feature for a data unit in the real-time network data, assigning a label to the data unit by applying the model to the value of the feature, and categorizing the data unit as associated with the botnet based on the label.

A method for identifying a botnet in a network, including analyzing historical network data using a pre-determined heuristic to determine values of a connectivity graph based feature in the historical network data, obtaining a ground truth data set having labels assigned to data units in the historical network data identifying known malicious nodes in the network, analyzing the historical network data and the ground truth data set using a machine learning algorithm to generate a model representing the labels as a function of the values of the connectivity graph based feature, analyzing real-time network data using the pre-determined heuristic to determine a value of the connectivity graph based feature for a data unit in the real-time network data, assigning a label to the data unit by applying the model to the value of the connectivity graph based feature, and categorizing the data unit as associated with the botnet based on the label.

In various examples, detected object data representative of locations of detected objects in a field of view may be determined. One or more clusters of the detected objects may be generated based at least in part on the locations and features of the cluster may be determined for use as inputs to a machine learning model(s). A confidence score, computed by the machine learning model(s) based at least in part on the inputs, may be received, where the confidence score may be representative of a probability that the cluster corresponds to an object depicted at least partially in the field of view. Further examples provide approaches for determining ground truth data for training object detectors, such as for determining coverage values for ground truth objects using associated shapes, and for determining soft coverage values for ground truth objects.

Characterization of carotid atherosclerosis and classification of plaque into symptomatic or asymptomatic along with the risk score estimation are key steps necessary for allowing the vascular surgeons to decide if the patient has to definitely undergo risky treatment procedures that are needed to unblock the stenosis. This application describes a statistical (a) Computer Aided Diagnostic (CAD) technique for symptomatic versus asymptomatic plaque automated classification of carotid ultrasound images and (b) presents a cardiovascular stroke risk score computation. We demonstrate this for longitudinal Ultrasound, CT, MR modalities and extendable to 3D carotid Ultrasound. The on-line system consists of Atherosclerotic Wall Region estimation using AtheroEdge™ for longitudinal Ultrasound or Athero-CTView™ for CT or Athero-MRView from MR. This greyscale Wall Region is then fed to a feature extraction processor which computes: (a) Higher Order Spectra; (b) Discrete Wavelet Transform (DWT); (c) Texture and (d) Wall Variability. The output of the Feature Processor is fed to the Classifier which is trained off-line from the Database of similar Atherosclerotic Wall Region images. The off-line Classifier is trained from the significant features from (a) Higher Order Spectra; (b) Discrete Wavelet Transform (DWT); (c) Texture and (d) Wall Variability, selected using t-test. Symptomatic ground truth information about the training patients is drawn from cross modality imaging such as CT or MR or 3D ultrasound in the form of 0 or 1. Support Vector Machine (SVM) supervised classifier of varying kernel functions is used off-line for training. The Atheromatic™ system is also demonstrated for Radial Basis Probabilistic Neural Network (RBPNN), or Nearest Neighbor (KNN) classifier or Decision Trees (DT) Classifier for symptomatic versus asymptomatic plaque automated classification. The obtained training parameters are then used to evaluate the test set. The system also yields the cardiovascular stroke risk score value on the basis of the four set of wall features.

A method and apparatus for automated liver segmentation in a 3D medical image of a patient is disclosed. A 3D medical image, such as a 3D computed tomography (CT) volume, of a patient is received. The 3D medical image of the patient is input to a trained deep image-to-image network. The trained deep image-to-image network is trained in an adversarial network together with a discriminative network that distinguishes between predicted liver segmentation masks generated by the deep image-to-image network from input training volumes and ground truth liver segmentation masks. A liver segmentation mask defining a segmented liver region in the 3D medical image of the patient is generated using the trained deep image-to-image network.

A system and method for detecting global positioning system (GPS) spoofing attacks includes collecting GPS readings along with inertial navigational system (INS) readings as a ground truth, and sequentially testing the GPS readings and INS readings through the use of a sequential probability ratio testing (SPRT) process.

Weather forecasting systems and methods for deriving and producing very short-term weather forecasts. Weather data and forecast information maybe input to derive very short-term weather forecasts in the 0-6 hour time frame. The system and methods include an expert system employing a hybrid or blending of forecasting approaches, with dynamic or intelligent arbitration of the different numerical, statistical, and human-driven techniques to produce more accurate weather forecasts for very short-term time periods. A calibration or internal verification against ground truth may be run independently with a time lag to produce more accurate future predictions of atmospheric state. The systems and methods may run in modes that produce either high resolution weather forecasts in the short-term period of 0-6 hours or a combination of synthetic, real-time weather observations / conditions and high resolution, short-term weather forecasts.

A system and method for ground truth estimation of an autonomous navigation system is described. The method includes calibrating a ground truth estimation system for the navigator, by determining a calibration pose of the navigator as disposed in relation to each of a plurality of landmarks during a calibration period. The method also includes directing the navigator to travel to a sequence of waypoints, the waypoints including a selected sequence of the landmarks. The method further includes determining the ground truth estimation based on an accuracy pose of the navigator as disposed in relation to the sequence of landmarks, and the calibration poses for the sequence of landmarks.

A prediction-based system and method for trajectory planning of autonomous vehicles are disclosed. A particular embodiment is configured to: receive training data and ground truth data from a training data collection system, the training data including perception data and context data corresponding to human driving behaviors; perform a training phase for training a trajectory prediction module using the training data; receive perception data associated with a host vehicle; and perform an operational phase for extracting host vehicle feature data and proximate vehicle context data from the perception data, using the trained trajectory prediction module to generate predicted trajectories for each of one or more proximate vehicles near the host vehicle, generating a proposed trajectory for the host vehicle, determining if the proposed trajectory for the host vehicle will conflict with any of the predicted trajectories of the proximate vehicles, and modifying the proposed trajectory for the host vehicle until conflicts are eliminated.

A prediction-based system and method for trajectory planning of autonomous vehicles are disclosed. A particular embodiment is configured to: receive training data and ground truth data from a training data collection system, the training data including perception data and context data corresponding to human driving behaviors; perform a training phase for training a trajectory prediction module using the training data; receive perception data associated with a host vehicle; and perform an operational phase for extracting host vehicle feature data and proximate vehicle context data from the perception data, generating a proposed trajectory for the host vehicle, using the trained trajectory prediction module to generate predicted trajectories for each of one or more proximate vehicles near the host vehicle based on the proposed host vehicle trajectory, determining if the proposed trajectory for the host vehicle will conflict with any of the predicted trajectories of the proximate vehicles, and modifying the proposed trajectory for the host vehicle until conflicts are eliminated.

In an automatic fingerprint authentication or identification system, the fingerprint image acquisition is severely effected by the limitations of the acquisition process. The two modes of input, viz. scanning inked fingerprints from paper records or directly from a finger using live-scan fingerprint scanners suffer from the following noise sources in the input in addition to standard noise in the camera. Non-uniform ink application, uneven pressure while rolling on the paper or pressing on the scanner surface and external dirt like oil and climatic variations in the moisture content of skin are some of the main causes for the ridges and valleys not to be imaged clearly. This invention deals with a method of learning a set of partitioned least-sqaures filters that can be derived from a given set of images and ground truth pairs as an offline process. The learned filters are convolved with input fingerprint images to obtain the enhanced image.

Various modes and embodiment of a method, apparatus, user interface, article of manufacture including a computer readable medium, computer data signals embodied on a carrier wave, and computer system for a GUI-based ground truth tool and insertion of user algorithms written in multiple programming languages. One embodiment comprises user interface for inserting a custom algorithm in a data-mining application. Another embodiment comprises a ground truth tool in a data-mining-application. A third embodiment comprises seamless insertion of custom algorithms in a data-mining application using tap points.

In general, a “Stereoscopic Video Converter” (SVC) provides various techniques for automatically converting arbitrary 2D video sequences into perceptually plausible stereoscopic or “3D” versions while optionally generating dense depth maps for every frame of the video sequence. In particular, the automated 2D-to-3D conversion process first automatically estimates scene depth for each frame of an input video sequence via a label transfer process that matches features extracted from those frames with features from a database of images and videos having known ground truth depths. The estimated depth distributions for all image frames of the input video sequence are then used by the SVC for automatically generating a “right view” of a corresponding stereoscopic image for each frame (assuming that each original input frame represents the “left view” of the stereoscopic image).

A method and tools for virtually aligning object detection sensors on a vehicle without having to physically adjust the sensors. A sensor misalignment condition is detected during normal driving of a host vehicle by comparing different sensor readings to each other. At a vehicle service facility, the host vehicle is placed in an alignment target fixture, and alignment of all object detection sensors is compared to ground truth to determine alignment calibration parameters. Alignment calibration can be further refined by driving the host vehicle in a controlled environment following a leading vehicle. Final alignment calibration parameters are authorized and stored in system memory, and applications which use object detection data henceforth adjust the sensor readings according to the calibration parameters.

A method, system, and computer program product is provided, for example, for predicting a location of a road sign. In an example embodiment, the method may include receiving a set of pre-processed road observations and extracting a plurality of features from the set of pre-processed road observations, wherein the plurality of features comprise at least a plurality of sensor based features and a plurality of map based features. Further, the method may include associating a set of sensor based features from the plurality of sensor based features and a set of map based features from the plurality of map based features with at least one of a plurality of ground truth points in ground truth data. Additionally, the method may include training a machine learning model based on the association of the ground truth data with the set of sensor based features and the set of map based features with at least one of a plurality of ground truth points in the ground truth data and further predicting the location of the road sign based on the trained machine learning model.

A system and a method are disclosed for generating video. Object information is received. A path of motion of the object relative to a reference point is generated. A series of images and ground for a reference frame are generated from the ground truth and the generated path. A system and a method are disclosed for generating an image. Object information is received. Image data and ground truth may be generated using position, the image description, the camera characteristics, and image distortion parameters. A positional relationship between the document and a reference point is determined. An image of the document and ground truth are generated from the object information and the positional relationship and in response to user specified environment of the document.

The technology disclosed relates to generating ground truth training data to train a neural network-based template generator for cluster metadata determination task. In particular, it relates to accessing sequencing images, obtaining, from a base caller, a base call classifying each subpixel in the sequencing images as one of four bases (A, C, T, and G), generating a cluster map that identifies clusters as disjointed regions of contiguous subpixels which share a substantially matching base call sequence, determining cluster metadata based on the disjointed regions in the cluster map, and using the cluster metadata to generate the ground truth training data for training the neural network-based template generator for the cluster metadata determination task.

The technology disclosed assigns quality scores to bases called by a neural network-based base caller by (i) quantizing classification scores of predicted base calls produced by the neural network-based base caller in response to processing training data during training, (ii) selecting a set of quantized classification scores, (iii) for each quantized classification score in the set, determining a base calling error rate by comparing its predicted base calls to corresponding ground truth base calls, (iv) determining a fit between the quantized classification scores and their base calling error rates, and (v) correlating the quality scores to the quantized classification scores based on the fit.

There is provided a discriminative framework for image alignment. Image alignment is generally the process of moving and deforming a template to minimize the distance between the template and an image. There are essentially three elements to image alignment, namely template representation, distance metric, and optimization method. For template representation, given a face dataset with ground truth landmarks, a boosting-based classifier is trained that is able to learn the decision boundary between two classes-the warped images from ground truth landmarks (e.g., positive class) and those from perturbed landmarks (e.g., negative class). A set of trained weak classifiers based on Haar-like rectangular features determines a boosted appearance model. A distance metric is a score from the strong classifier, and image alignment is the process of optimizing (e.g., maximizing) the classification score. On the generic face alignment problem, the proposed framework greatly improves the robustness, accuracy, and efficiency of alignment.

OCR errors are identified and corrected through learning. An error probability estimator is trained using ground truths to learn error probability estimation. Multiple OCR engines process a text image, and convert it into texts. The error probability estimator compares the outcomes of the multiple OCR engines for mismatches, and determines an error probability for each of the mismatches. If the error probability of a mismatch exceeds an error probability threshold, a suspect is generated and grouped together with similar suspects in a cluster. A question for the cluster is generated and rendered to a human operator for answering. The answer from the human operator is then applied to all suspects in the cluster to correct OCR errors in the resulting text. The answer is also used to further train the error probability estimator.

A method, system and a computer program product are provided for classifying elements in a ground truth training set by iteratively assigning machine-annotated training set elements to clusters which are analyzed to identify a prioritized cluster containing one or more elements which are frequently misclassified and display machine-annotated training set elements associated with the first prioritized cluster along with a warning that the first prioritized cluster contains one or more elements which are frequently misclassified to solicit verification or correction feedback from a human subject matter expert (SME) for inclusion in an accepted training set.

A personality model is created for a population and used as an input to a text generation system. Alternative texts are created based upon the emotional effect of the generated text. Certain words or phrases are “pinned” in the output, reducing the variability of the generated text so as to preserve required information content, and a number of tests provide input to a discriminator network so that proposed outputs both match an outside objective regarding the information content, emotional affect, and grammatical acceptability. A feedback loop provides new “ground truth” data points for refining the personality model and associated generated text.

A method for predicting reactiveness of MMI users comprises:reacting to a message with a mobile user device which is a receiver of the message,collecting ground-truth data (11) for a machine-learning classifier,extracting from the collected ground-truth data (11) a list of features (12) which determines a current or past context of the user, and each feature having a feature's prediction strength calculated as fraction of classes misclassified when removing the feature;selecting the list of features (12) based on each feature's prediction strength;defining a plurality of reactiveness classes (101); both the extracted list of features (12) and the reactiveness classes (101) being input to the machine-learning classifier;classifying (102) the user according to the defined reactiveness classes (101);predicting the user's reactiveness for the given current or past context of the user by determining the most likely reactiveness class via the machine-learning classifier.

The present disclosure relates to training and utilizing an image exposure transformation network to generate a long-exposure image from a single short-exposure image (e.g., still image). In various embodiments, the image exposure transformation network is trained using adversarial learning, long-exposure ground truth images, and a multi-term loss function. In some embodiments, the image exposure transformation network includes an optical flow prediction network and / or an appearance guided attention network. Trained embodiments of the image exposure transformation network generate realistic long-exposure images from single short-exposure images without additional information.

The prognostic tool disclosed here decomposes the problem of estimating the remaining useful life (RUL) of a component or sub-system into two separate regression problems: the feature-to-damage mapping and the operational conditions-to-damage-rate mapping. These maps are initially generated in off-line mode. One or more regression algorithms are used to generate each of these maps from measurements (and features derived from these), operational conditions, and ground truth information. This decomposition technique allows for the explicit quantification and management of different sources of uncertainty present in the process. Next, the maps are used in an on-line mode where run-time data (sensor measurements and operational conditions) are used in conjunction with the maps generated in off-line mode to estimate both current damage state as well as future damage accumulation. Remaining life is computed by subtracting the instance when the extrapolated damage reaches the failure threshold from the instance when the prediction is made.

The invention relates to a method of quickly segmenting a moving target in a non-restrictive scene based on a full convolution network, which belongs to the technical field of video object segmentation. The method comprises steps: firstly, framing is carried out on the video, and a result after framing is used for making a Ground Truth set S for a sample image; a full convolution neural network trained through a PASCAL VOC standard library is adopted to predict a target in each frame of the video, a deep feature estimator for an image foreground target is acquired, target maximum intra-class likelihood mapping information in all frames is obtained hereby, and initial prediction on the foreground and the background in the video frames is realized; and then, through a Markov random field, deep feature estimators for the foreground and the background are refined, and thus, segmentation on the video foreground moving target in the non-restrictive scene video can be realized. The information of the moving target can be effectively acquired, high-efficiency and accurate segmentation on the moving target can be realized, and the analysis precision of the video foreground-background information is improved.