ÄãºÃ£¬»¶Ó­À´µ½¾­¹ÜÖ®¼Ò [µÇ¼] [×¢²á]

ÉèΪÊ×Ò³ | ¾­¹ÜÖ®¼ÒÊ×Ò³ | Êղر¾Õ¾

»ùÓÚÖ§³ÖÏòÁ¿»úµÄÈËÁ³Ê¶±ð¼¼ÊõÑо¿_ͨÐŹ¤³ÌרҵÂÛÎÄ·¶ÎÄ

·¢²¼Ê±¼ä£º2015-01-24 À´Ô´£ºÈË´ó¾­¼ÃÂÛ̳
ͨÐŹ¤³ÌרҵÂÛÎÄ Ä¿ ¼ Õª ÒªI AbstractII ͼĿ¼III ±íĿ¼IV Ä¿ ¼V 1. Ð÷ ÂÛ1 1.1 ¿ÎÌâÑо¿±³¾°1 1.2Ïà¹Øѧ¿ÆÑо¿½øÕ¹3 1.3×Ô¶¯ÈËÁ³Ê¶±ðϵͳ4 1.4±¾ÎÄÑо¿ÄÚÈݼ°½á¹¹5 2. ÈËÁ³Ê¶±ð×ÛÊö7 2.1 »ùÓÚÈ˹¤¶¨ÒåÌØÕ÷µÄʶ±ð·½·¨7 2.2 »ùÓÚ×Ô¶¯»ñÈ¡ÌØÕ÷µÄʶ±ð·½·¨7 2.2.1 »ùÓÚÖ§³ÖÏòÁ¿»úµÄʶ±ð·½·¨7 2.2.2 »ùÓÚÉñ¾­ÍøÂçµÄʶ±ð·½·¨9 2.2.3 »ùÓÚͳ¼ÆÌØÕ÷µÄʶ±ð·½·¨10 2.2.4 »ùÓÚС²¨ÌØÕ÷µÄµ¯ÐÔÆ¥Åä·½·¨12 3. Ö§³ÖÏòÁ¿»úµÄ»ù±¾ÀíÂÛºÍËã·¨16 3.1 Ö§³ÖÏòÁ¿»úµÄÌصãÓëÓ¦ÓÃ16 3.2 Ö§³ÖÏòÁ¿»ú¸ÅÊö21 3.2.1 Ö§³ÖÏòÁ¿»ú£¨SVM£©µÄÈËÁ³Ê¶±ð½á¹¹21 3.2.2 Ö§³ÖÏòÁ¿»ú£¨SVM£©µÄÈËÁ³Ê¶±ðËã·¨22 4. »ùÓÚDCTºÍÖ§³ÖÏòÁ¿»úµÄÈËÁ³Ê¶±ðϵͳ24 4.1 ÈËÁ³ÌØÕ÷µÄÌáÈ¡24 4.1.1 ÀëÉ¢ÓàÏұ任24 4.1.2 ϵÊýÑ¡È¡26 4.1.3 ±¾ÊµÑéÖеÄDCT´¦Àí¹ý³Ì31 4.2 Ö§³ÖÏòÁ¿»úµÄ½á¹¹Éè¼Æ31 5. ÐÔÄÜÆÀ¼Û33 5.1 ÈËÁ³Êý¾Ý¿â33 5.2 ʵÑé½á¹û34 5.2.1 ²»Í¬ÌØÕ÷άÊýÐÔÄܱȽÏ34 5.2.2 ²»Í¬ºËº¯ÊýµÄÐÔÄܱȽÏ35 6. ×ܽá37 л´Ç38 ²Î¿¼ÎÄÏ×39 ¸½Â¼²¿·Ö41 ¸½Â¼A41 Õª Òª ÈËÁ³Ê¶±ð¼¼ÊõÊǹúÄÚÍ⹲ͬ¹Ø×¢µÄÒ»¸öÇ°ÑØ¿ÎÌ⣬ÔÚÏÖ´ú¾­¼ÃºÍÉç»áµÄ·¢Õ¹ÖÐÓÐ×ÅÊ®·Ö¹ã·ºµÄÓ¦ÓÃÁìÓòºÍÓ¦ÓÃÇ°¾°£¬È簲ȫϵͳ¡¢×ﷸʶ±ð¡¢µçÊÓ»áÒéµÈ£¬Òò¶øÒѾ­³ÉΪµ±Ç°Ä£Ê½Ê¶±ðºÍÈ˹¤ÖÇÄܵÄÒ»¸öÑо¿Èȵ㡣 ±¾ÎÄ×ܽáÁËÈËÁ³Ê¶±ð¼¼ÊõµÄÑо¿ÏÖ×´£¬ÌÖÂÛ½«Ö§³ÖÏòÁ¿»úÓÃÓÚģʽʶ±ðµÄÀíÂÛ£¬Ñо¿ÁËÆäÖеĹؼü¼¼ÊõºÍÄѵ㣬²¢½øÐÐÁË·ÖÎöºÍ±È½Ï¡£±¾ÎÄÌá³öÁËÒ»ÖÖ»ùÓÚͼƬ·Ö¸îµÄÈËÁ³ÌØÕ÷ÌáÈ¡·½·¨£¬¸Ã·½·¨ÀûÓöþάÀëÉ¢ÓàÏұ任¶Ôÿ¸ö×ÓͼƬ½øÐзֽ⣬²¢ÀûÓÃÖ§³ÖÏòÁ¿»ú×÷Ϊ·ÖÀàÆ÷À´Ê¶±ð²»Í¬µÄÈËÁ³¡£»ùÓÚͼÏóµÄ»®·Ö£¬Ò»¸öеÄͼÏóÌáÈ¡µÄ·½·¨£¬ËüʹÓÃ2άµÄÀëÉ¢ÓàÏұ任À´·Ö½âͼÏó£¬ÌØÕ÷±»Ìá³öÀ´¡£¸ù¾ÝDCT£¬Ò»¸öÁ³²¿Ê¶±ðÄ£ÐÍÊÇÓÉÓëSVMS°ó¶¨¹¹³ÉµÄ¡£ÎªÁË»®·Ö·ûºÏÀàÐ͵ķÖÀ࣬һ¶Ô¶àµÄ²ßÂÔÔÚÎÒÃǵÄÄ£ÐÍÖб»Ê¹Óá£Ö§³ÖÏòÁ¿»úÊÇÐÂÒ»´ú»ùÓÚÔÚ¾²Ì¬Ñ§Ï°ÀíÂÛÖеĽü´úÏȽøµÄѧϰϵͳ¡£SVMSÔÚÕæʵÊÀ½çµÄʵÏÖÖÐÌύͼ»­×´Ì¬µÄ±íÏÖ£¬ÖîÈçÎı¾·ÖÀ࣬ͼÏó·ÖÀ࣬С¿éÐÅÏ¢£¬µÈµÈ¡£ºÍĿǰһЩÈËÁ³Ê¶±ð·½·¨Ïà±È£¬±¾ÎÄÌá³öµÄÈËÁ³Ê¶±ðËã·¨¾ßÓнϺõÄÐÔÄÜ£¬ÔÚORLÈËÁ³¿âÉϵÄÐÔÄÜÄ£Äâ±íÃ÷£¬Ëã·¨¾ßÓнϸßʶ±ðÂÊ¡£ ¹Ø¼ü´Ê£ºÈËÁ³Ê¶±ð£»Ö§³ÖÏòÁ¿»ú£»ÀëÉ¢ÓàÏұ任£»ORLÊý¾Ý¿â Research on Face Recognition Technology Based on Discrete Cosine Transform and Support Vector Machines Abstract Face recognition technology (FRT) is front-line task in pattern recognition domain, which has extremely extensive applications such as security systems, criminal identifications, teleconferences and so on. Thus, the study of the technology has been a research focus in pattern recognition and artificial intelligence. In this paper, the actuality of automated face recognition is summarized, the theory of applying SVM into pattern recognition is discussed, and crucial face recognition technologies and difficulties are analyzed and compared. In this paper, A new method of feature extraction to face image, which uses 2-dimension discrete cosine transform to decompose each sub-image, is proposed based on images partition in this paper. And support vector machine is used as classifier to recognize different face image. Based on image partition, a new feature extraction method, which uses 2-dimension discrete cosine transform to decompose sub-image, is proposed. According to DCT, a face recognition model is constructed combined with SVM. In order to classify multi-class classification, one-vs-all strategy is used in our model. Support vector machine is a new generation learning system based on recent advances in statistical learning theory. SVMs deliver state-of-the-art performance in real-world applications such as text categorization, image classification, bioinformatics, and so on. Compared with some current approaches, our algorithm has better performance, the simulation results on ORL database show that our system has a high recognition rate. Key Words: face recognition; Support Vector Machines; discrete cosine transform; ORL database
¾­¹ÜÖ®¼Ò¡°Ñ§µÀ»á¡±Ð¡³ÌÐò
  • ɨÂë¼ÓÈë¡°¿¼ÑÐѧϰ±Ê¼ÇȺ¡±
ÍƼöÔĶÁ
¾­¼ÃѧÏà¹ØÎÄÕÂ
±êÇ©ÔÆ
¾­¹ÜÖ®¼Ò¾«²ÊÎÄÕÂÍƼö