COMPUTER-IMPLEMENTED METHOD FOR AUTOMATED OBJECT RECOGNITION AND CLASSIFICATION IN SCENES USING SEGMENT-BASED OBJECT EXTRACTION

摘要

One embodiment relates to a computer-implemented method (100) for automated object recognition and classification in scenes using segment-based object extraction. The method includes automated procedures for receiving (202) video images, creating (204) segmentation maps from said images, grouping (206) segments so as to form extracted objects, extracting (400) features from said extracted objects, and classifying (600) said extracted objects using said features. Other features, aspects and embodiments are also disclosed.