基于仿生智能的人脸检测方法研究-信息与通信工程专业论文.docxVIP

Study Study on the Method for Face Detection Based on Bionic Intelligence Wang Wenj nan(Information and Communication Engineering) Directed by Professor Wang Yanj iang Abstract Detecting human faces robustly and fast is the most important issue during the study of face detection．Combined with bionics and artificial intelligence methods，this thesis studied on how to improve the robustness and the speed of face detection，some new methods were proposed in the thesis and listed as follows， 1．The speed of face detection is very important during real—time detection for input images or input video sequences．A novel face window searching algorithm based on evolutionary agent is proposed which is used to detect faces in gray·scale images．It Can quickly find the candidate face windows through the evolutionary computation of distributed agents，which represent different kind of windows．Experimental results show that the evolutionary agent—based searching algorithm Can increase the detection speed by 5-7 times compared with the traditional exhaustive searching method used in some general algorithm． 2．A novel face detection method based on bionic pattern recognition is proposed．Face detection can be regarded as the classification of“face’’and“non—face”．According to the basic idea of the continuity of the same kind of samples in feature space，the face detection algorithm can be constructed by training face samples which distribute manifold． Experimental results show that the face detection based on bionic pattern recognition has the advantages of low rej ection rate and high detection rate． Key words：Face Detection；Evolutionary Computation；Agent；Bionic Pattern Recognition；Searching Algorithm 关于学位论文的独创性声明本人郑重声明：所呈交的论文是本人在指导教师指导下独立进行研究工作所取得的 关于学位论文的独创性声明 本人郑重声明：所呈交的论文是本人在指导教师指导下独立进行研究工作所取得的 成果，论文中有关资料和数据是实事求是的。尽我所知，除文中已经加以标注和致谢外， 本论文不包含其他人已经发表或撰写的研究成果，也不包含本人或他人为获得中国石油 大学(华东)或其它教育机构的学位或学历证书而使用过的材料。与我一同工作的同志 对研究所做的任何贡献均已在论文中作出了明确的说明。 若有不实之处，本人愿意承担相关法律责任。 学位论文作者签名： R期：鼬10年6月f口F1 学位论文使