红外技术, 2017, 39 (10): 946, 网络出版: 2017-12-01
对比度阈值自适应的SIFT图像拼接算法
Image Mosaic Based on Contract Threshold Adaptive SIFT Algorithm
摘要
图像拼接技术主要由3 部分构成:采集、匹配和融合,其中匹配最为关键,决定拼接质量。针对SIFT 图像拼接算法的特征点数目局部过高或过低情况,采用对比度阈值自适应的SIFT 图像拼接算法,可得到合理数目的特征点。此方法保证特征点数目在150~300 之间,过高或过低时,可自适应调节数目,进行有效控制。实验证明:改进的算法保证拼接效果的基础上,提高运算效率,具有实用价值。
Abstract
An image mosaic consists of three parts: acquisition, image-matching, and image fusion. Image-matching is the key part, determining the image mosaic quality. For cases where in the feature point number is too high or too low, we propose an image mosaic based on the contract threshold adaptive SIFT algorithm. The proposed method adaptively and effectively controls the feature point number when it is too high or too low, ensuring a value between 150 and 300. Experimental results show that this method, which has practical value, can improve computational efficiency based on mosaic effects.
李尊, 申小萌, 苗同军. 对比度阈值自适应的SIFT图像拼接算法[J]. 红外技术, 2017, 39(10): 946. LI Zun, SHEN Xiaomeng, MIAO Tongjun. Image Mosaic Based on Contract Threshold Adaptive SIFT Algorithm[J]. Infrared Technology, 2017, 39(10): 946.