CAJ | 학술논문

针对具有一定旋转角度的多幅图像拼接问题，提出了一种基于SIFT-ACO的图像拼接算法。对拼接图像进行预处理后，用SIFT算法提取图像特征，将经典的生物智能优化算法ACO加入到图像配准中，对图像特征点进行配准，最后将多幅图像融合，得到一幅宽视角、无缝、高精度的图像。对多组不同角度拍摄的图像进行拼接试验。试验结果表明：在不影响整体视觉效果的前提下，该算法能有效地提高图像拼接精度。
침대구유일정선전각도적다폭도상병접문제，제출료일충기우SIFT-ACO적도상병접산법。대병접도상진행예처리후，용SIFT산법제취도상특정，장경전적생물지능우화산법ACO가입도도상배준중，대도상특정점진행배준，최후장다폭도상융합，득도일폭관시각、무봉、고정도적도상。대다조불동각도박섭적도상진행병접시험。시험결과표명：재불영향정체시각효과적전제하，해산법능유효지제고도상병접정도。
For problems of multiple images stitching of certain rotation angle, puts forward an image mosaic optimiza-tion algorithm based on SIFT-ACO. After the mosaic image preprocessing, the algorithm uses SIFT algorithm to extract the image features, adds classic intelligent optimization algorithm to image registration and makes image feature points registration, at last fuses multiple images and obtains a wide viewing angle, seamless and high precision image. The experimental result of image mosaic of different angles indicates that the algorithm can improve the accuracy of image matching effectively without affecting the overall visual effect.