Model-based detection, segmentation, and classification of compact objects

Mark J. Carlotto

doi:10.1117/12.2179810

Translator Disclaimer

21 May 2015 Model-based detection, segmentation, and classification of compact objects

Mark J. Carlotto

Author Affiliations +

Proceedings Volume 9474, Signal Processing, Sensor/Information Fusion, and Target Recognition XXIV; 94740Z (2015) https://doi.org/10.1117/12.2179810
Event: SPIE Defense + Security, 2015, Baltimore, Maryland, United States

Abstract

A unified model-based approach to ATR that uses 3D models to control detection, segmentation, and classification is described. Objects are modeled by rectangular boxes whose dimensions are Gaussian random variables. A fast predictor estimates the size and shape of expected objects in the image, which controls detection and segmentation algorithms. Segmentation fits oriented rectangles (length x width @ pose) to object-like regions detected using a multi‐level thresholding/region tracking approach. Detections are classified by comparing measured to predicted region length and width in the pose direction. The method is fast and requires only a coarse characterization of objects/object classes.

Citation Download Citation

Mark J. Carlotto "Model-based detection, segmentation, and classification of compact objects", Proc. SPIE 9474, Signal Processing, Sensor/Information Fusion, and Target Recognition XXIV, 94740Z (21 May 2015); https://doi.org/10.1117/12.2179810

ACCESS THE FULL ARTICLE