激光与光电子学进展, 2019, 56 (14): 141502, 网络出版: 2019-07-12 复制并引用该论文  本文已被引 2 次  被引通知

图 & 表

图 1. 双目测距原理图。实线表示左侧摄像机光路图,虚线表示右侧摄像机光路图

Fig. 1. Principle of binocular distance measurement. Solid lines represent light path of left camera and dotted lines represent light path of right camera

下载图片 查看原文

图 2. 程序流程图

Fig. 2. Flow chart of program

下载图片 查看原文

图 3. 关键点探测结果

Fig. 3. Results of key-point detection

下载图片 查看原文

图 4. 关键点匹配结果

Fig. 4. Results of key-point matching

下载图片 查看原文

图 5. 社区发现结果

Fig. 5. Results of community discovery

下载图片 查看原文

图 6. 双目测距结果

Fig. 6. Results of binocular distance measurement

下载图片 查看原文

表 1四种高斯模糊计算方法的消耗时间、复杂度与误差对比

Table1. Comparison of time consumption, complexity, and error for four Gaussian blur calculation methods

Method Time consumption/ms Complexity Error /% n = 11 n= 21 Original Gaussian blur 2452 8990 O(n×n×W×H) 0 Mean blur 589 2437 O(n×n×W×H) 0.04 Separation mean blur 230 435 O(2×n×W×H) 0.04 Integral mean blur 32 34 O(W×H) 0.04

查看原文

表 2本文SIFT运算开销与已有工作的对比

Table2. Comparison of time consumption between proposed SIFT algorithm and previous work

Item Proposedalgorithm Previous workin Ref. [23] Convolution core width 31×31 19×19 Image size /(pixel×pixel) 640×480 640×480 GPU hardware GT940M GTX560 Theoretical floating-pointcomputing power /TFlops 1 1 Time consumption /ms 100 634

查看原文

张志强, 施文华. 改进的尺度不变特征变换算法并行加速双目测距系统及其实现[J]. 激光与光电子学进展, 2019, 56(14): 141502. Zhiqiang Zhang, Wenhua Shi. Research and Implementation of Binocular Distance Measurement System Based on Improved Scale-Invariant Feature Transform Algorithm with Parallel Acceleration[J]. Laser & Optoelectronics Progress, 2019, 56(14): 141502.