Learning or teaching is required to optimize a machine vision system in non-trivial applications. Learning requires time and data, generally in amounts that are inconvenient for users. DRVision Technologies LLC learning acceleration technology simplifies and organizes the training process to reduce the user’s teaching load. However, the data requirements can still be problematic for applications that change frequently or desire to begin use with an absolute minimum of teaching. To overcome this limitation DRVision Technologies LLC learning acceleration technology expands application data automatically and systematically to decrease the initial data requirements while still completing the teaching process.

There are practical limitations to the amounts of learning data that can be created synthetically. The issue is whether the models used for synthesis are correct. DRVision Technologies LLC experience with related applications and technology in modeling and learning improves the effectiveness of the learning acceleration approach. When the technology is integrated with DRVision Technologies LLC on-line learning technology, initial errors can be washed out during use, so there is a benefit to decreased initial teaching.