Award Data

Toyon Research Corporation proposes to develop algorithms that improve the precision and recall of neural networks for low-shot object classification and detection. Our approach is based upon developing a descriptive multi-domain feature representation of the low-shot target as well as the surrounding context. A multi-branch neural network merges the various domains of information to perform accur ...

Toyon proposes to research and develop algorithms for generalized salient change detection, and to incorporate these algorithms into software tools implemented on the cloud. Our approach leverages the two most promising methods from Phase I, both based on supervised learning. The first method is the entropy-based feature vector and corresponding neural network, which we will apply at a coarse sear ...

Implement methods to improve our Phase I architecture for complex SAR imagery collected from different systems and with different collection parameters. Implement techniques to boost performance across all ground features of interest. Reduce the amount of human intervention needed for training our SAR image segmentation architecture.  SAR segmentation can be applied to a number of areas: Intellig ...

This SBIR Phase II project will develop biologically inspired computational models and algorithms to enable low-shot and one-shot detectionof objects-of-interest in remote sensing imagery. The Phase II effort will build upon our Phase I work including multi-scale representationlearning framework and deep-learning based feature extraction and matching techniques for low-shot target detection. The P ...

With respect to digital imaging sensors and systems, obtaining as many pixels on target is a necessity to classify or identify a target for real-time operations and forensic analysis within the intelligence community. Rather than relying on improved sensors, the approach originally solicited and further refined here utilizes super-resolution image processing techniques to provide more detail in th ...

This SBIR Phase II project proposes a probabilistic approach to determine a vehicle’s location using onboard video and Lidar sensors and foundation map data in GPS denied environments. The proposed system does not rely on only one type of information source, instead it combines proposals from a variety of location estimators to find a vehicles location in GPS-denied environments. The system take ...

With the ever-growing number of imaging satellites in orbit the job of an analyst will change from eyes on pixels to analysis of informationfrom imagery thanks to automated image processing like object detection and change detection. Object detection algorithms have advancedto near human performance given that there is sufficient labeled data on which to train; however, obtaining this data is cost ...

Prescription eyeglass is a $100 billion industry that serves 3 billion people worldwide. It is also oneof the top 10 consumer markets for plastic consumption. Majority of all eyeglass lenses today aremade with Polycarbonate a Prop 65 regulated chemical. Significant health and environmentalconcerns prevail around the presence of Bis Phenol A or BPA due to its effects on reproductivehealth. This SBI ...

Vaccination has been the most effective means of controlling influenza infections of pigs. Howeverprotective antibody responses are largely strain-specific and are not effective against different influenzaviruses. Therefore the development of a vaccine candidate that elicits broad-spectrum protection againstdiverse influenza A viruses (IAV) i.e. a universal swine influenza vaccine is critically ne ...

The ability to freeze boar semen provides several advantages including disease reduction, more rapid genetic transfer, genetic banking, and more precise correlation with optimum female fertility. Although freezing is routine technology in many livestock species, it is not so in teh swine industry. Unfortunately, in boars, sperm quality inconsistencies and poor fertility make this technology commer ...