Combining Human and Machine Capabilities for Improved Accuracy and Speed in Visual Recognition Tasks

Abstract. This study investigated methods of enhancing human computer interaction in applications of pattern recognition where higher accuracy is required than is currently achievable by automated systems, but where there is enough time for a limited amount of human interaction. On a flower identification task, methods were explored to improve the accuracy of human-only and machine-only recognition by employing human-computer interaction. Human involvement in the feature extraction process, especially in color selection, was found to be most beneficial