عنوان مقاله :
ﯾﮑﺴﻮﺳﺎزي ﺗﺼﺎوﯾﺮ اﺳﺘﺮﯾﻮ ﮐﺎﻟﯿﺒﺮه ﺷﺪه ﺑﺮ اﺳﺎس ﻣﯿﺪان ﺗﻨﺎﻇﺮي در ﺳﯿﺴﺘﻢ ﻫﺎي ﭼﻨﺪ ﻣﻨﻈﺮ
عنوان به زبان ديگر :
Calibrated stereo image rectification based on the correspondence field for multi-view systems
پديد آورندگان :
ﺣﺴﯿﻦ ﭘﻮر، ﺣﺴﺎم اﻟﺪﯾﻦ داﻧﺸﮕﺎه آزاد اﺳﻼﻣﯽ ﻋﻠﻮم و ﺗﺤﻘﯿﻘﺎت - داﻧﺸﮑﺪه ﻣﮑﺎﻧﯿﮏ، ﺑﺮق و ﮐﺎﻣﭙﯿﻮﺗﺮ، ﺗﻬﺮان، اﯾﺮان , ﻣﻮﺳﻮي ﻧﯿﺎ، اﻣﯿﺮ داﻧﺸﮕﺎه ﺧﻮاﺟﻪ ﻧﺼﯿﺮاﻟﺪن ﻃﻮﺳﯽ - داﻧﺸﮑﺪه ﻣﻬﻨﺪﺳﯽ ﮐﺎﻣﭙﯿﻮﺗﺮ، ﺗﻬﺮان، اﯾﺮان , ﭘﻮرﻣﯿﻨﺎ، ﻣﺤﻤﺪ ﻋﻠﯽ داﻧﺸﮕﺎه آزاد اﺳﻼﻣﯽ ﻋﻠﻮم و ﺗﺤﻘﯿﻘﺎت - داﻧﺸﮑﺪه ﻣﮑﺎﻧﯿﮏ، ﺑﺮق و ﮐﺎﻣﭙﯿﻮﺗﺮ، ﺗﻬﺮان، اﯾﺮان
كليدواژه :
ﺑﺎر ﻣﺤﺎﺳﺒﺎﺗﯽ , ﯾﮑﺴﻮﺳﺎزي ﺗﺼﺎوﯾﺮ اﺳﺘﺮﯾﻮ , ﻣﯿﺪان ﺗﻨﺎﻇﺮي , ﺧﻂ اﭘﯽ ﭘﻮﻻر , ﻧﺴﺒﺖ ﻓﻮاﺻﻞ ﮐﺎﻧﻮﻧﯽ دو دورﺑﯿﻦ
چكيده فارسي :
ﭼﮑﯿﺪه: ﯾﮑﺴﻮﺳﺎزي ﺗﺼﻮﯾﺮ ﯾﮏ ﺑﺨﺶ ﻣﻬﻤﯽ در ﺑﯿﺸﺘﺮ ﺳﯿﺴﺘﻢ ﻫﺎي ﭼﻨﺪ ﻣﻨﻈﺮ1 ﻣﯽ ﺑﺎﺷﺪ. ﯾﮑﺴﻮﺳﺎزي ﺗﺼﻮﯾﺮ، ﻓﺮاﯾﻨﺪ ﺗﻄﺒﯿﻖ اﺳﺘﺮﯾﻮ را ﺑﺎ ﻣﺤﺪود ﻧﻤﻮدن ﻣﺴﯿﺮ ﺑﻪ ﺧﻄﻮط اﻓﻘﯽ و ﻣﻮازي ﺳﺎده ﻣﯽ ﺳﺎزد. ﻣﺸﮑﻞ اﺳﺎﺳﯽ در ﯾﮑﺴﻮﺳﺎزي ﻣﺘﺪاول ﺗﺼﻮﯾﺮ ﭘﯿﭽﯿﺪﮔﯽ و ﺑﺎر ﺳﻨﮕﯿﻦ ﻣﺤﺎﺳﺒﺎت ﺑﺮاي ﺑﻪ دﺳﺖ آوردن ﻣﺎﺗﺮﯾﺲ ﮐﺎﻟﯿﺒﺮاﺳﯿﻮن، ﻣﺎﺗﺮﯾﺲ ﭘﺮوﺟﮑﺸﻦ و ﻣﺎﺗﺮﯾﺲ اﺳﺎﺳﯽ اﺳﺖ. در اﯾﻦ ﻣﻘﺎﻟﻪ ﯾﮏ روش ﻣﻮﺛﺮ و ﻣﻄﻤﺌﻦ ﺑﺮ اﺳﺎس ﻣﯿﺪان ﺗﻨﺎﻇﺮي2 ﺑﺮاي ﯾﮑﺴﻮﺳﺎزي ﺗﺼﻮﯾﺮ اراﺋﻪ ﻣﯽ ﺷﻮد ﮐﻪ از ﻣﺎﺗﺮﯾﺲ ﻫﺎي ﻓﻮق اﺟﺘﻨﺎب ﻣﯽ ﮔﺮدد. ﺑﻪ ﻃﻮر ﮐﻠﯽ اﯾﻦ اﻟﮕﻮرﯾﺘﻢ ﻗﺎدر اﺳﺖ ﺗﺼﻮﯾﺮ را ﺑﺮاي ﻫﺮ ﻣﻮﻗﻌﯿﺖ ﺟﺪﯾﺪ دورﺑﯿﻦ ﺗﻮﻟﯿﺪ ﻧﻤﺎﯾﺪ. ﺑﻪ ﺧﺼﻮص ﻣﯽ ﺗﻮان ﻣﻮﻗﻌﯿﺖ دورﺑﯿﻦ ﺟﺪﯾﺪ را ﻃﻮري ﺗﻌﺮﯾﻒ ﮐﺮد ﮐﻪ ﺗﺼﻮﯾﺮ ﯾﮑﺴﻮ ﺷﺪه ﺑﻪ دﺳﺖ آﯾﺪ. ﺗﺼﺎوﯾﺮ ﺷﺒﯿﻪ ﺳﺎزي ﺷﺪه و واﻗﻌﯽ ﺑﺮاي ارزﯾﺎﺑﯽ اﻟﮕﻮرﯾﺘﻢ اراﺋﻪ ﺷﺪه اﺳﺘﻔﺎده ﺷﺪه اﻧﺪ. اﻟﮕﻮرﯾﺘﻢ SIFT ﻧﯿﺰ ﺑﺮاي ﺗﺨﻤﯿﻦ ﻧﺰدﯾﮏ ﺑﻮدن ﺗﺼﻮﯾﺮ ﯾﮑﺴﻮ ﺷﺪه ﺑﻪ ﺗﺼﻮﯾﺮ واﻗﻌﯽ ﺑﻪ ﮐﺎر ﮔﺮﻓﺘﻪ ﺷﺪه اﺳﺖ. ﻧﺘﺎﯾﺞ ﺷﺒﯿﻪ ﺳﺎزي ﻧﺸﺎن ﻣﯽ دﻫﺪ ﻣﯿﺎﻧﮕﯿﻦ اﻧﺪازه ﺧﻄﺎ و اﻧﺤﺮاف ﻣﻌﯿﺎر ﺑﻪ ﺗﺮﺗﯿﺐ ﺣﺪود 25. و 26. ﭘﯿﮑﺴﻞ ﻣﯽ ﺑﺎﺷﺪ و ﺗﺼﻮﯾﺮ ﯾﮑﺴﻮ ﺷﺪه ﻣﺴﺘﻘﻞ از ﻧﺴﺒﺖ ﻓﻮاﺻﻞ ﮐﺎﻧﻮﻧﯽ دو دورﺑﯿﻦ اﺳﺖ
چكيده لاتين :
Image rectification is an essential part in most multi-view systems. It simplifies the stereo matching process by limiting the search path to horizontal and parallel lines. The main problem in conventional image rectification is complexity and mass computation load for calculation of calibration matrix, projection matrix and fundamental matrix. In this paper an efficient and confident method is presented to rectify stereo images based on the correspondence field (CF) that avoids the mentioned matrices. In general this algorithm is able to generate an image for any new camera position. Especially one can choose the new position in such a way that the virtual camera generates a rectified image. Both real and simulated stereo images are used to evaluate the proposed algorithm. SIFT algorithm is used to estimate how close is the generated image to original rectified one. Simulation results show the mean absolute error and standard deviation are about 0.25 and 0.26 pixels respectively and rectified image is independent of focal ratio.
عنوان نشريه :
مهندسي برق و الكترونيك ايران
