摘要: In this paper, we learn to discover composite operators and features that are evolved from combinations of primitive image processing operations extract regions-of-interest (ROIs) in images. Our approach is based on genetic programming (GP). The motivation for using GP there a great many ways combining these the human expert, limited by experience, knowledge time, can only try very small number conventional combination. Genetic programming, other hand, attempts unconventional combination may never be imagined experts. some cases, yield exceptionally good results. experimental results show find operators, consist designed effectively regions interest images learned applied ROIs similar
参考文章(6)
Stephen A. Stanhope, Jason M. Daida, Genetic programming for automatic target classification and recognition in synthetic aperture radar imagery Lecture Notes in Computer Science. pp. 735- 744 ,(1998) , 10.1007/BFB0040824
Riccardo Poli, Genetic Programming for Feature Detection and Image Segmentation artificial intelligence and the simulation of behaviour. pp. 110- 125 ,(1996) , 10.1007/BFB0032777
John R. Koza, Genetic Programming II: Automatic Discovery of Reusable Programs ,(1994)
Jason M. Daida, Stephen A. Stanhope, Genetic Programming for Automatic Target Classification and Recognition Evolutionary Programming. pp. 735- 744 ,(1998)
Daniel Howard, SimonC. Roberts, Richard Brankin, Target detection in SAR imagery by genetic programming Advances in Engineering Software. ,vol. 30, pp. 303- 311 ,(1999) , 10.1016/S0965-9978(98)00093-3
Simon C. Roberts, Daniel Howard, Evolution of Vehicle Detectors for Infrared Line Scan Imagery Lecture Notes in Computer Science. pp. 111- 125 ,(1999) , 10.1007/10704703_9