Abstract: A system and method for assessing patient movement for Parkinson's disease includes capturing a video of a subject performing a finger tapping sequence comprising a predetermined number of open and close periods. According to an exemplary embodiment, a system and method includes extracting a region of interest for each frame of the video and generating a projection of the region of interest for each frame of the video using perpendicular vector projections in a direction or plurality of directions.

Abstract: What is disclosed is a system and method for extracting photoplethysmographic (PPG) signal (i.e., a cardiac signal) on a continuous basis from a time-series signals obtained from video images captured of a subject being monitored for cardiac function in a non-contact remote sensing environment involves the following. First, a time-series signal obtained from video images captured of a region of exposed skin where a photoplethysmographic (PPG) signal of a subject of interest can be registered. A sliding window is then used to define consecutive sequential segments of the time-series signal for processing. Each of the consecutive time-series signal segments is detrended such that low frequency variations and non-stationary components are removed. The detrended signals are processed to obtain, for each segment, a PPG signal. The PPG signal segments are then stitched together using a stitching method, as disclosed herein, to obtain a continuous PPG signal for the subject.

Abstract: A method and system for domain adaptation based on multi-layer fusion in a convolutional neural network architecture for feature extraction and a two-step training and fine-tuning scheme. The architecture concatenates features extracted at different depths of the network to form a fully connected layer before the classification step. First, the network is trained with a large set of images from a source domain as a feature extractor. Second, for each new domain (including the source domain), the classification step is fine-tuned with images collected from the corresponding site. The features from different depths are concatenated with and fine-tuned with weights adjusted for a specific task. The architecture is used for classifying high occupancy vehicle images.

Abstract: A computer-based apparatus including a computer including a processor arranged to select a first video regarding a medical condition; create a second video including segments from the first video; transmit the second video for viewing by qualified medical personnel; receive input from the personnel; based on the input confirm accuracy of a first segment or modify a second segment or delete a third segment; create, from the second video, by at least including the first or second segment or deleting the third segment; transmit the third video for viewing by viewers; receive a respective response from each viewer identifying a respective fourth segment of the third video deemed relevant to the medical condition or enjoyable; create a fourth video including at least a portion of the respective fourth segments; and store the fourth video for inclusion in a video regarding the medical condition.

Abstract: A system and method that includes training a classifier using uniquely defined landmark points along the windshield region based on an elastic deformation model. The deformation model uses mixtures of trees with a shared pool of parts and can be globally optimized with dynamic programming and still capture much relevant global elastic structure. Once a candidate area is identified in the scene, a learned threshold is applied to the classification score of the candidate area to determine if the candidate area is a windshield. The identified area is then cropped out for further downstream process.

Abstract: A system and method of providing annotated trajectories by receiving image frames from a video camera and determining a location based on the image frames from the video camera. The system and method can further include the steps of determining that the location is associated with a preexisting annotation and displaying the preexisting annotation. Additionally or alternatively, the system and method can further include the steps of generating a new annotation automatically or based on a user input and associating the new annotation with the current location.

Abstract: What is disclosed is a system and method for processing image data acquired using a multi-band infrared camera system with a spectral mosaic filter arranged in a geometric pattern without having to perform a demosaicing that is typical with processing data from an array of sensors. In one embodiment, image data that has been captured using a camera system that has a spectral filter mosaic comprising a plurality of spectral filters arrayed on a grid. A material index is determined, using intensity values collected by sensor elements associated with a cell's respective spectral filters. All of the material indices collectively generate a material index image. Thereafter, material identification is performed on the material index image using, for example, pixel classification. Because the demosaicing step can be effectively avoided, image processing time is reduced. The teachings hereof find their uses in a wide array of applications including automated HOV/HOT violation detection.

Abstract: What is disclosed is a video system and method that accounts for differences in imaging characteristics of differing video systems used to acquire video of respective regions of interest of a subject being monitored for a desired physiological function. In one embodiment, video is captured using N video imaging devices, where N?2, of respective regions of interest of a subject being monitored for a desired physiological function (i.e., a respiratory or cardiac function). Each video imaging device is different but has complimentary imaging characteristics. A reliability factor f is determined for each of the devices in a manner more fully disclosed herein. A time-series signal is generated from each of the videos. Each time-series signal is weighted by each respective reliability factor and combined to obtain a composite signal. A physiological signal can be then extracted from the composite signal. The processed physiological signal corresponds to the desired physiological function.

Abstract: What is disclosed is a system and method for detecting febrile seizure using a thermal video camera. In one embodiment, a video is received comprising time-sequential thermal images of a subject. The video is acquired of the subject in real-time using a thermal video system. Each thermal image comprises a plurality of pixels with an intensity value of each pixel corresponding to a temperature. The thermal images are processed to determine an occurrence of a febrile seizure. The processing involves identifying a region of interest in the thermal image and determining a temperature for the region of interest based on values of the pixels isolated in that region of interest. Thereafter, a rate of change of temperatures is obtained for the subject in real-time on a per-frame basis. If the rate of change is determined to have exceeded a pre-defined threshold level, then the subject is having a febrile seizure.

Abstract: What is disclosed is a system and method for compensating for motion during processing of a video of a subject being monitored for physiological function assessment. In one embodiment, image frames are received. Successive batches of N video frames are processed to isolate pixels associated with a body region of the subject where a physiological signal is registered by the camera. The pixels are processed to obtain a time-series signal for each batch. A determination is made whether movement during video acquisition of this batch of image frames exceeds a threshold level. If so then a size N of the next batch of image frames is changed to: N=N+M1, where N+M1?Nmax. Otherwise, a size N of a next batch is changed to: N=N?M2, where N?M2?Nmin. Thereafter, processing repeats in a real-time continuous manner as the next batch of the N image frames is received. Various embodiments are disclosed.

Abstract: What is disclosed is a system for compensating for motion induced artifacts in a physiological signal obtained from multiple videos of a first and second region of interest a subject being monitored for a desired physiological function. At least one of the videos being of the first region and at least one of the videos being of the second region. The first region being at least one area of exposed skin where a desired signal corresponding to the physiological function can be registered by a video imaging device. The second region being an area where a movement by the subject is likely to induce motion artifacts into the signal. The videos are processed to isolate pixels associated with the first and second regions. Processed pixels of the isolated first regions to obtain a composite time-series signal. From the composite signal, a physiological signal corresponding to the physiological function is extracted.

Abstract: What is disclosed is a system and method for compensating for motion induce artifacts in a physiological signal obtained from a video. In one embodiment, a video of a first and second region of interest of a subject being monitored for a desired physiological function is captured by a video device. The first region is an area of exposed skin wherein a desired signal corresponding to the physiological function can be registered. The second region is an area where movement is likely to induce motion artifacts into that signal. The video is processed to isolate pixels in the image frames associated with these regions. Pixels of the first region are processed to obtain a time-series signal. A physiological signal is extracted from the time-series signal. Pixels of the second region are analyzed to identify motion. The physiological signal is processed to compensate for the identified motion.

Abstract: A system and method for enhancing images including an image capture device operably connected to a data processing device that captures an image of a target vehicle, and a processor-usable medium embodying computer code, said processor-usable medium being coupled to said data processing device, said computer program code comprising instructions executable by said processor. The instructions configured for identifying a region within the image including a window of the target vehicle, applying a first image enhancement effect to the identified region, applying a second image enhancement effect to a remainder of the image not including the identified region, the second image enhancement effect different than the first image enhancement effect.

Abstract: A computer-vision based method for validating an activity workflow of a human performer includes identifying a target activity. The method includes determining an expected sequence of actions associated with the target activity. The method includes receiving a video stream from an image capture device monitoring an activity performed by an associated human performer. The method includes determining an external cue in the video stream. The method includes associating a frame capturing the external cue as a first frame in a key frame sequence. The method includes determining an action being performed by the associated human performer in the key frame sequence. In response to determining the action in the key frame sequence matching an expected action in the target activity, the method includes verifying the action as being performed in the monitored activity.

Abstract: A method for reconstructing an image of a scene captured using a compressed sensing device. A mask is received which identifies at least one region of interest in an image of a scene. Measurements are then obtained of the scene using a compressed sensing device comprising, at least in part, a spatial light modulator configuring a plurality of spatial patterns according to a set of basis functions each having a different spatial resolution. A spatial resolution is adaptively modified according to the mask. Each pattern focuses incoming light of the scene onto a detector which samples sequential measurements of light. These measurements comprise a sequence of projection coefficients corresponding to the scene. Thereafter, an appearance of the scene is reconstructed utilizing a compressed sensing framework which reconstructs the image from the sequence of projection coefficients.

Abstract: What is disclosed is system and method for contemporaneously reconstructing images of a scene illuminated with unstructured and structured illumination sources. In one embodiment, the system comprises capturing a first 2D image containing energy reflected from a scene being illuminated by a structured illumination source and a second 2D image containing energy reflected from the scene being illuminated by an unstructured illumination source. A controller effectuates a manipulation of the structured and unstructured illumination sources during capture of the video. A processor is configured to execute machine readable program instructions enabling the controller to manipulate the illumination sources, and for effectuating the contemporaneous reconstruction of a 2D intensity map of the scene using the second 2D image and of a 3D surface map of the scene using the first 2D image. The reconstruction is effectuated by manipulating the illumination sources.

Abstract: A method for detecting settle-down time in a space includes acquiring a sequence of frames capturing a select space from a first camera. The method includes determining an initial time for computing a duration it takes for an associated occupant to settle into a seat in the select space. The method includes determining one or more candidate frames from the sequence of frames where one or both of a sitting behavior and seat occupancy is observed at the seat. The method includes determining a final frame and a final time associated with the final frame from the one or more candidate frames. The method includes computing the settle-down time using the initial and the final times.

Abstract: A method for detecting sitting behavior includes acquiring a sequence of frames capturing a scene-of-interest at an overhead view. The method includes detecting at least one empty seat within the scene-of-interest and associating the seat as being unoccupied and the frame as a reference frame. The method includes extracting reference features describing a region of the unoccupied seat in the reference frame and quantifying the reference features to form a reference feature vector. The method includes extracting features describing the region in a given frame and quantifying the features to form a current feature vector. The method includes measuring a change in a feature vector over time using the reference feature vector and the current feature vector. The method includes and determining a status of the seat in the given frame as being one of occupied and unoccupied based on the change in the feature vector.

Abstract: A system and method of video-based chew counting by receiving image frames from a video camera, determining feature points within the image frames from the video camera, generating a motion signal based on movement of the feature points across the image frames from the video camera, and determining a chew count based on the motion signal.

Abstract: What is disclosed is a system and method for detecting an arrhythmic or non-arrhythmic event from a cardiac signal obtained from a subject. In one embodiment, a plurality of different cardiac signals are received and are transformed into frequency domain signals which, in turn, are changed such that a dominant frequency in each of the signals is substantially aligned to form a matrix of feature vectors. The feature matrix is used to train a classifier. A cardiac signal from the subject is received and transformed to a frequency domain signal. The frequency domain signal is changed such that a dominant is substantially aligned with a dominant frequency of signals used to train the classifier. The subject's frequency domain signal is provided as a new feature vector to the classifier. The classifier uses the new feature vector to classify the subject as having an arrhythmic or a non-arrhythmic event.