基于学习的人脸面部图像快速检测算法研究-计算机应用技术专业论文.docxVIP

北京工业大学工学硕士学位论文Abstract 北京工业大学工学硕士学位论文 Abstract Human face detection is a process of finding the human face image extents and locations information of human face in a given image taking no account of 3D—pose， illumination and other conditions．Face detection is an arduous work mainlv because extracting face fcaturcs in f如e pattem is very difficuIt．Face features vary largely in d铂hrem cases such as dif艳rent human face Doses、 dif传rent illuminations、 unkllown 1ighting、occlusion a11d many other conditions． This thesis discussed the rapid face detection algori恤lbased on le锄ing Which is one ofpmmising trends of face detection tecllnica． 111ere are three factors must research in the raDid face detection method based on leaming：leaming algoritllrIl，出e dennition of features a11d the method of building claSsifler based on these features，the m籼er of combining ef先ctivelv these f宅atures． This method 6rstly defines a tvpe of feature fomlat and then bases on it to build several weak classifiers using the method of building classi6er whose perfornlance is less good．Leaming algorithm chooses and combines the weak classifier，generated the strong classiner combined by、veak classiflers，and then it can be used to detect face．Most ef南ctively combing the weak classifiers is imponant work for increasing the speed of detecting task． The thesis discussed respectively these three fhctors in the rapid detecting algorit№based on leaming，studied enlphatically the method of buil‘ling classier based on features and the combin窟of the weak classiners which is needed to build rapid face detection system．About building classiner based on Harr— like fbatures，the thesis put forward a novel method based on the symmetry of face． In the final，the thesis build a rapid face detection system using a cascaded classifiers based on Adaboost leaming algorithrn，Harr—like feature and the method of building classmer．The flnal system detects image whose size is 320+240 in less thaJl 0．1 second with detection mte ofabov