C12029: MOVA: A Wearable Integrated System for Real-Time Anomaly Tracking
Novelty:
The current invention is a novel algorithmic framework for detection of anomalies in images from both real time stationary and non-stationary data sources.
Value Proposition:
The method for detection of Movement Anomalies, MovA, is an unsupervised and automatic method for anomaly object detection from stationary and/or non-stationary sensors (e.g., cameras, etc) that can be generalized to cover many scenarios. We have developed a real-time manifold distance learning (MDL) system that can track, distinguish, and count objects. These MDL methods can be integrated to embedded systems for remote devices such as night vision, drone and security apparatus. Advantages of the current invention include:
• Automatic anomaly detection and localization: Determination of contribution of each image to normal and anomaly patterns without need for training or human/expert supervision.
.
• Reduced computational load, making it suitable for real-time applications.
.
• Ability to detect the structure and changes in different images using the feature space and any associated objects.
.
Technical Details:
Johns Hopkins researchers have developed an embedded system with real-time manifold and incremental learning ability. MovA track moving objects within a frame and detect anomalies. The system can also detect localization based on tracking pattern of changes over time. The system does not need to store/access to previous data and video frames. It is comprised of two “sub-systems” that enable the prediction and graphing of the motion pattern at the inter-frame and intra-frame level in real-time, with the added ability to track moving objects in different frames. These tracking features allow for better differentiation of objects in frames over time, greater flexibility and easier deployment to assist the user in difficult arenas. For example, using MovA to define objects under night vision observation can yield a higher probability of success compared with current methods.
Looking for Partners:
To develop and commercialize the technology for object tracking.
Stage of Development:
Pre-Clinical
Data Availability:
Under CDA/NDA
Publications/Associated Cases:
Not available at this time.
