CAJ | 학술논문

针对视频拼接存在的边缘失真和融合效率问题，提出一种新的抗畸变高效的实时视频拼接算法。通过建立非线性抗畸变模型，对视频图像存在的非线性畸变进行校正，改善图像边缘过渡的平滑性；采用 SURF配准算法兼顾配准的精度和效率；使用二维标记模板降低融合过程的冗余计算，进行快速图像融合。该算法提升了传统方法的视频拼接质量与融合效率。
침대시빈병접존재적변연실진화융합효솔문제，제출일충신적항기변고효적실시시빈병접산법。통과건립비선성항기변모형，대시빈도상존재적비선성기변진행교정，개선도상변연과도적평활성；채용 SURF배준산법겸고배준적정도화효솔；사용이유표기모판강저융합과정적용여계산，진행쾌속도상융합。해산법제승료전통방법적시빈병접질량여융합효솔。
A novel anti-distortion and efficient video stitching algorithm is proposed to solve video splicing edge distortion and fusion efficiency.The establishment of nonlinear anti-distortion model (NAM)improves the smoothness of the transition edge by correcting the non-linear distortion of the video image.In order to solve the contradiction between accuracy and effi-ciency,speeded up robust features (SURF)algorithm is used.And two-dimensional mark model (2DMM)is presented to accelerate the speed of image fusion by reducing redundant computation.Compared with the traditional method,it promotes the video splicing quality and fusion efficiency.