There is a need to understand the variability of biological conclusions due to the choice of a similarity metric, and due to the software quality and parameters of similarity computations. The goal of our effort is to advance high throughput and high confidence image comparisons accessed from any stationary or mobile computer device. We built a web accessible and computationally scalable system composed of image similarity metrics. The similarity metrics are validated regularly by per-configured tests. We have also added a workflow editor that allows access from a variety of platforms.
Our approach is based on organizing and evaluating image similarity metrics first according to several existing surveys of image similarities. The similarity metrics are represented by a triplet consisting of image loaders and color space representations, image descriptors, and proximity measures. The proximity measures are grouped into those that can operate on histogram descriptors, contiguous image segments, clusters of image pixels or raw pixel values. This classification of individual computations and their sub-categories allows us to build a simple tree taxonomy encapsulating image loading/representation, image characterization and comparison, and to map the taxonomy into intuitive web interfaces.
Data flow of image similarity comparisons
We have designed and integrated 40 similarity metrics into the Versus framework. These similarity metric implementations were thoroughly validated by using inputs and their corresponding numerically predicted outputs, and by processing input synthetic images and expected image outputs. The validation suite of tests can be executed regularly to verify correctness of similarity computations. Next, we deployed the web accessible image similarity computations on several virtual machines that form a computational cloud. The browser interface enables access to the computational resources via mobile devices that have the sensing and data exchange capabilities. Our current prototype web interface has been optimized to accommodate a tree-based taxonomy of similarity metrics, and to propagate any relevant information from a server to the web browser interface (e.g., definition of metrics or error messages).
Web Access to Image Similarity Measurements
Access to Measurements from Mobile Devices