CAJ | 학술논문

针对红外序列图像中运动弱小点目标的检测问题,重点研究序列关联检测算法.在多帧关联检测的基础上,引入决策融合概念,建立了基于模糊决策融合的序列关联检测方法.基于贝叶斯最小风险准则建立分布式决策模糊融合方法,有效融合单帧检测的结果,提高序列检测算法的性能.实测数据的实验结果验证了基于贝叶斯最小风险准则的分布式决策模糊融合方法的可行性和有效性.
침대홍외서렬도상중운동약소점목표적검측문제,중점연구서렬관련검측산법.재다정관련검측적기출상,인입결책융합개념,건립료기우모호결책융합적서렬관련검측방법.기우패협사최소풍험준칙건립분포식결책모호융합방법,유효융합단정검측적결과,제고서렬검측산법적성능.실측수거적실험결과험증료기우패협사최소풍험준칙적분포식결책모호융합방법적가행성화유효성.
Toward detection of moving point target in infrared sequential images, the sequential association detection algorithm is research in this paper. Based on the single detection, the decision fusion is introduced and the sequential detection based on fuzzy decision fusion is build. Because the fuzzy decision fusion is built based on minimum Bayesian risk criterion, it improves the detection performance. Experimental results to real data proof the feasibility and validity of distributed fuzzy decision fusion method based on minimum Bayesian risk criterion.