ÄãºÃ£¬»¶Ó­À´µ½¾­¹ÜÖ®¼Ò [µÇ¼] [×¢²á]

ÉèΪÊ×Ò³ | ¾­¹ÜÖ®¼ÒÊ×Ò³ | Êղر¾Õ¾

»ùÓÚÆæÒìÖµ·Ö½âµÄÊý×ÖͼÏñˮӡËã·¨Ñо¿_ͨÐŹ¤³ÌרҵÂÛÎÄ

·¢²¼Ê±¼ä£º2015-01-24 À´Ô´£ºÈË´ó¾­¼ÃÂÛ̳
ͨÐŹ¤³ÌרҵÂÛÎÄ Ä¿ ¼ Õª Òª I ABSTRACT II µÚÒ»Õ ¸ÅÊö 1 1.1 Êý×ÖͼÏñˮӡµÄÏà¹Ø֪ʶ 1 1.1.1 Êý×ÖͼÏñˮӡ 1 1.1.2 Êý×ÖͼÏñˮӡ¼¼Êõ 1 1.1.3 Êý×ÖͼÏñˮӡ¼¼ÊõµÄ·ÖÀà 2 1.2 »ùÓÚSVDµÄˮӡËã·¨Ñо¿ÏÖ×´ 4 1.3 ±¾ÂÛÎÄÕ½ڰ²ÅÅ 6 µÚ¶þÕ Ëã·¨Ïà¹Ø¼¼Êõ½éÉÜ 7 2.1 ÆæÒìÖµ·Ö½â£¨SVD£©µÄÌØÐÔ 7 2.2 Arnold±ä»» 8 2.2.1 Arnold±ä»»¼°ÆäÖÜÆÚÐÔ 8 2.2.2 Arnold·´±ä»»Ëã·¨ 9 2.3SVDÓëÆäËûËã·¨µÄ½áºÏ 10 2.3.1 »ùÓÚDCTºÍSVDµÄÊý×ÖͼÏñˮӡ¼¼Êõ 10 2.3.2 »ùÓÚDWTºÍSVDµÄÊý×ÖˮӡËã·¨ 11 2.3.3 »ùÓÚDFT£­SVDÓò¿¹¼¸ºÎ¹¥»÷ͼÏñˮӡËã·¨ 12 2.4Êý×ÖͼÏñ´¦Àí¼°Matlab¼ò½é 14 2.4.1 Matlab¼ò½é 14 2.4.2 Êý×ÖͼÏñ´¦Àí¼ò½é 16 µÚÈýÕ »ùÓÚÆæÒìÖµ·Ö½âµÄÊý×ÖͼÏñˮӡËã·¨ 18 3.1 ˮӡǶÈëËã·¨ 18 3.2 ˮӡÌáÈ¡Ëã·¨ 18 3.3¸Ä½øµÄ»ùÓÚSVDµÄÊý×ÖͼÏñˮӡËã·¨ 20 3.4¹¥»÷²âÊÔÓë½á¹û·ÖÎö 22 3.5 ³¢ÊÔÓëÀëɢС²¨±ä»»½áºÏ 30 3.5.1 Ëã·¨½éÉÜ 30 3.5.2 ¹¥»÷²âÊÔ 30 µÚËÄÕ ×ܽá 35 4.1 ¹¤×÷×ܽá 35 4.2 Õ¹Íû 36 ÖÂл 37 ²Î¿¼ÎÄÏ× 38 ¸½Â¼ 40 Õª Òª Êý×ÖˮӡÊǽ«Éí·ÝÈ·ÈÏÐÅÏ¢»ò±£ÃÜÐÅÏ¢ÏâǶÓÚͼÏñÖеÄÒ»ÖÖ¼¼Êõ£¬¿É¿¿µÄˮӡ¿ÉΪÐÅÏ¢µÄ°²È«Ìṩ¿É¿¿µÄ±£Ö¤¡£Ä¿Ç°Ðí¶àˮӡËã·¨ÊÇÔÚ¿Õ¼äÓò»ò±ä»»Óò²åÈëÊý¾ÝµÄ£¬ÀýÈçÀëÉ¢ÓàÏұ任£¨DCT£©¡¢ÀëÉ¢¸µÁ¢Ò¶±ä»»£¨DFT£©¡¢ÀëɢС²¨±ä»»£¨DWT£©¡£ÔÚ±¾ÎÄÖУ¬²ÉÓÃÁË»ùÓÚÆæÒìÖµ·Ö½â£¨Singular Value Decomposition£©µÄÊý×ÖˮӡËã·¨¡£Í¼ÏñÆæÒìÖµ·Ö½â£¨SVD£©ÓÐÒÔÏÂÐÔÖÊ£º·Ö½âºóͼÏñ¾ØÕóµÄÆæÒìÖµ¼¯Öз´Ó³ÁËͼÏñµÄ¡°ÁÁ¶È¡±£¨ÄÜÁ¿£©ÌØÐÔ£¬¶ø¶ÔÓ¦µÄÆæÒì¾ØÕóÖ»·´Ó³ÁËͼÏñµÄ¡°¼¸ºÎ¡±ÌØÐÔ¡£Òò¶øÆæÒìÖµµÄϸ΢±ä»¯²»»áÓ°ÏìͼÏñµÄÊÓ¾õЧ¹û¡£ ÔÚ±¾ÎÄËã·¨ÖУ¬ÖÃÂÒÓÃÓÚÊý×ÖͼÏñÒþ²ØµÄÔ¤´¦ÀíºÍºó´¦Àí¡£¶ÔÊý×ÖˮӡÐźŽøÐÐÖÃÂÒ·ÖÉ¢ÁËԭʼˮӡÐźŵÄÏà¹ØÐÔ£¬ÔÚÔâµ½¼ôÇй¥»÷ʱ¿ÉÒÔ½«´íÎóÂëÔª¾¡¿ÉÄÜ·ÖÉ¢£¬Òò´ËÓÐЧµØÌá¸ßÁËÊý×ÖˮӡËã·¨µÄ¿¹¼ôÇй¥»÷ÐÔÄÜ¡£±¾Ëã·¨ÊDzÉÓÃArnold±ä»»¶ÔˮӡͼÏñ½øÐÐÖÃÂҵģ¬µ«ÀûÓÃArnold±ä»»ÖÜÆÚÐÔÀ´»Ö¸´Ô­Í¼µÄ¼ÆËãÁ¿ºÜ´ó¡£ËùÒÔÔÚºó´¦Àí¹ý³ÌÖУ¬²ÉÓÃÁËÒ»ÖÖÀûÓÃÄæ±ä»»¾ØÕóÀ´ÇóArnold·´±ä»»µÄËã·¨¡£±¾ÎÄËã·¨»¹¶ÔÌáÈ¡³öµÄˮӡ½øÐÐÁËÁ¿»¯¡£Ê×ÏÈÈ·¶¨ÏñËØֵΪ1µÄÏÂÏÞºÍÉÏÏÞ£¬È»ºó¶ÔÌáÈ¡µÄˮӡͼÏñ½øÐÐÁ˶þÖµ»¯´¦Àí£¬Ê¹×îÖÕµÄˮӡͼÏñЧ¹û¸ü¼Ñ¡£±¾ÂÛÎÄËã·¨»¹³¢ÊÔ½«ÆæÒìÖµ·Ö½âÓëÀëɢС²¨±ä»»Ïà½áºÏ£¬¼´½«Ë®Ó¡Ç¶È뵽ԭͼÏñ¶þάÀëÉ¢±ä»»ºóËùµÃµÍƵ²¿·Ö¡£ÊµÑé½á¹û±íÃ÷£¬»ùÓÚÆæÒìÖµ·Ö½âµÄ±¾Ëã·¨¶Ô³£ÓõÄͼÏñ´¦Àí¹¥»÷¾ßÓÐÁ¼ºÃµÄ³°ôÐԺͲ»¿É¼ûÐÔ¡£ ¹Ø¼ü´Ê£ºÆæÒìÖµ·Ö½â£¬Êý×ÖͼÏñˮӡ£¬Â³°ôÐÔ£¬Arnold±ä»» ABSTRACT Digital watermarking is a technique that can inlay identity information and secrecy information into images. Reliable watermarking provides a pledge for information safety. Many current watermarking algorithms insert data in the spatial or transform domains like the discrete cosine, the discrete Fourier, and the discrete wavelet transforms. In this paper, propose a digital watermarking algorithm based on Singular Value Decomposition (SVD). According to some properties, of SVD, each singular value (SV) specifies the luminance of the SVD image layer, whereas the respective pair of singular vectors specifies image geometry. Therefore slight variations of SV cannot effect the visual perception. In the algorithm of this paper, scrambling technology is used as pre-processing and post-processing of digital image information hiding. The image watermarking is permuted to reduce the relativity of original pixels, so the error bits of the extracted watermarking are dispersed as well. Therefore the resistance to crop attack is improved significantly. In the algorithm, use Arnold transformation to scramble the watermark image, but the image resumption is computational expensive due to the periodicity of Arnold transformation. Therefore, in the post-processing process, used one kind of algorithm how ask the Arnold inverse transformation using the athwart transformation matrix. In the algorithm, but also has carried on the quantification to the recovered watermark image. First determined the lower limit and the upper limit of picture element value is 1, then to the recovered watermark image has carried on two values processing, caused the final watermark image effect to be better. In the algorithm, SVD will try to combine with DWT, and watermark is embedded into the low-frequency part of the original image after two-dimensional discrete wavelet. Experimental result show that the watermarking method based on SVD used to attack the image processing performs well in both imperceptibility and robustness. KEY WORDS singular value decomposition, digital image watermarking, robustness, Arnold transformation
¾­¹ÜÖ®¼Ò¡°Ñ§µÀ»á¡±Ð¡³ÌÐò
  • ɨÂë¼ÓÈë¡°¿¼ÑÐѧϰ±Ê¼ÇȺ¡±
ÍƼöÔĶÁ
¾­¼ÃѧÏà¹ØÎÄÕÂ
±êÇ©ÔÆ
¾­¹ÜÖ®¼Ò¾«²ÊÎÄÕÂÍƼö